Survival in An Overmedicated World Look Up The Evidence Yourself 9788770364300
Survival in An Overmedicated World Look Up The Evidence Yourself 9788770364300
Survival in An Overmedicated World Look Up The Evidence Yourself 9788770364300
C. Gøtzsche
People’sPress
Acknowledgements
I am very grateful for the inspiration and advice I have received from patients
and their relatives, colleagues, friends, family, and others. I wish to especially
thank David Hammerstein, Tom Jefferson and my Spanish publisher Enrique
Murillo for their critical readings of my manuscript, and Steven Woloshin and
Lisa Schwartz for commenting on the section about chest pain. I am also very
grateful for the stimulating discussions I had with the participants of two
workshops in Amsterdam, particularly with Sharon Batt, Els Torrele, Irina
Cleemput, Donald W. Light and Silvio Garattini who commented on earlier
versions of a manuscript about abolishing patents in healthcare which I discuss
in Chapter 14.
Disclaimer: Read this book at your
own risk
This book is intended to help you navigate in the bewildering and often
contradictory world of information about diagnosis and treatment. The
information provided in the book is not intended to be a replacement for
consultation with adequately qualified healthcare professionals. This book is
intended to better prepare you for discussing your problems with professionals.
The advice I give is based on my personal views; other professionals might have
other views. Therefore, since my advice is not intended to be a substitute for
consulting your doctor or other healthcare professionals, I cannot be held
responsible for whatever you decide to do, whatever the outcome.
1 Introduction
‘You do not ask a barber if you need a haircut.’ Most people have heard this
expression or something similar. Yet we willingly allow our doctors to subject us
to various diagnostic investigations and treatments which may be financially
beneficial for themselves. Healthcare is riddled with financial conflicts of
interest, and even when your doctor does not directly benefit, there are many
other reasons you should be on the alert. In good faith, doctors use many
treatments which do not work, and since all treatments will cause harm to some
patients, doctors do harm to many people.
Therefore, you must look up evidence in order to protect yourself against being
harmed. Harm is usually caused by drugs but can also be caused by, for instance,
infections, surgeries, Chinese herbs, electroshock, diagnostic testing, and
admission to hospitals - dangerous places because of the many errors made
there.
I cannot write a book about all the issues. I have chosen to focus on a few
common diseases and also on some easily curable diseases that can be deadly if
overlooked. I have also focused on drugs which take many lives unnecessarily
because many of those who died did not really need them.
This book is not about guiding you through particular diseases. It is designed to
give you the confidence to look up answers to questions, yourself. That might
empower you in discussions with doctors and other healthcare professionals and
help you reject some of the diagnostic methods and treatments suggested, e.g.
when you conclude they probably will not work - or are too dangerous - or both.
You will not learn how to play chess, football or tennis by reading books; you
have to practice. Work through the many examples provided in this book by
going online, yourself. Perhaps by changing your search criteria slightly each
time, you will gradually improve and realize that finding the answers you need is
often surprisingly quick and easy. Therefore, please take notes – and use a
highlighter - so that you can quickly find what you need when your skills need
refreshing.
When you are finished reading, please revisit your notes and highlighting. My
apologies for being so schoolteacher-like, but I do have a little experience in
studying - two full academic educations - both as a biologist and a physician.
When you fall ill, you can do a lot to help yourself be more informed when it
comes to deciding which of healthcare’s many offers to accept or decline.
Participating as a partner in the decision-making process, and doing the best you
can, may provide you with enough peace of mind to make dealing with any
situation easier, including accepting any ultimate outcome, be it good or bad. It
may even help you survive and live a better life by avoiding interventions that
may endanger your health. I hope my book will contribute to that.
I wish good luck to the patients who always let their doctor make the decisions
for them. They will need it. Doctors make many errors of judgment, often
because they do not know better and use far too many drugs. We live in a world
that is so overdiagnosed and overtreated that, in high-income countries, our
medications are the third leading cause of death after heart disease and cancer.
This has been demonstrated by several independent studies in Europe and North
America. 1-9 It has also been shown that medical errors, including other than
drug-related errors, are the third leading cause of death even when only counting
hospital patients’ deaths. 10 Most of these deaths are preventable. 10
All these avoidable deaths are a public health disaster - one of the biggest one
we have ever had - and far worse than any other since the Spanish flu outbreak
during World War I. This epidemic of drug deaths is much easier to deal with
than all the other epidemics we have seen (such as infectious diseases) because
we can combat it by using drugs more sparingly. Yet no one does anything that
really makes an impact - these deaths continue to pile up year after year. I have
never understood why we use so many resources on preventing and treating
heart disease and cancer, and so few resources on preventing drug deaths. That
was one compelling reason for me for writing this book.
I provide examples from my own illnesses, and those of my family and friends,
when I feel they are helpful for understanding how we all ended up in the same
boat. And these are also to aid us in perceiving the importance of taking
responsibility for our diseases - our own health and lives - by remembering to
ask crucial questions before it is too late. Several of these examples are about the
three leading causes of death: drugs, heart disease and cancer.
In 2002, the British Medical Journal (BMJ ) published a theme issue entitled,
Too Much Medicine? with articles on the medicalization of birth, sex, and death.
The opening editorial wondered whether doctors could become pioneers of de-
medicalization, handing back power to patients, resisting disease-mongering,
and demanding fairer global distribution of effective treatments. A decade later,
as data on overuse and overdiagnosis mounted, the BMJ re-announced the
campaign, but this time omitting the question mark. 12
Overdiagnosis is common and does matter because, once people are labelled
with a diagnosis, a cascade of medical, social, and economic consequences
follows - some of which are permanent. Medical labels and ensuing treatments
take emotional and financial tolls on people, while also increasing healthcare
system costs. 12
Doctors learn a lot about prescribing drugs but very little or nothing about when
to stop. That doctors are quite poor at stopping their patients’ drugs and tend to
renew prescriptions indefinitely is not surprising. We all know what it means to
prescribe, yet its counterpart – ‘de-prescribing’ – was not a term commonly used
in the scientific literature until 2012. A search of the PubMed medical research
database yielded only 213 articles referring to ‘deprescribing’, yet ‘prescribing’
turned up 36,198 articles - 170 times as many!
Please note that when I describe search results in this book, it must be implicitly
understood that these searches were performed on particular dates. They should
be expected to differ somewhat if repeated. I did all the searches between July
and October in 2017 but I have not listed the search dates. Search dates are only
provided when preparing scientific papers. In most databases, the searches can
be limited to prespecified dates, e.g. covering articles entered into the database
before, say, October 13, 2017, meaning that search results can be precisely
reproduced.
I have debated with doctors who argue we should not be too concerned about the
many deaths caused by our drugs, because some treatments - like for cancer -
carry a well-known risk of premature death, yet they improve overall survival.
Other drugs (they argue) substantially improve quality of life and are therefore
worth taking even though deadly to some people, such as disease modifying
drugs for rheumatoid arthritis.
These arguments are relevant but cannot explain the many deaths we cause.
Most deaths are caused by drugs which many patients do not really need, like
drugs for emotional or physical pain, for instance 11,13 which I shall comment on
later.
Of course, doctors are well-intentioned and want to help their patients to the best
of their abilities. Yet there are many things physicians do not know or cannot see
and, therefore, they can neither act upon nor learn from them.
One issue that makes it particularly difficult for doctors and patients to choose
diagnostic methods and treatments that will provide the best outcomes is that the
scientific literature about diagnoses and treatments is generally of poor quality
and considerably biased. Countless scientific studies have demonstrated this, but
doctors are rarely aware of it, tending to believe what gets published. Therefore,
doctors believe diagnoses to be far more reliable - and treatments far more
effective with fewer harms - than they actually are.
I will discuss these issues throughout the book, both from theoretical
perspectives and when relating to tangible health problems.
In addition to our ignorance about the sad state of healthcare research, another
important reason we use far too many drugs is clinical experience, which is often
very often misleading. Examining many patients makes you a better doctor, but
it is also treacherous. Doctors have their own intervention preferences - usually
drugs - and are aware that most patients improve when taking them. Therefore,
they ascribe improvements to the treatments provided. We call that type of
knowledge ‘uncontrolled experience,’ because no control groups of untreated
patients with which to compare are provided. Even in Hippocrates’s day, it was
understood that clinical experience might be fallacious, and there seems to be no
limit to the number of absurd - often harmful - treatments doctors have
employed. 14
Our personal experiences make a strong impact on us and we tend to not think
about what the course of a disease might have been without treatment. We tell
each other what we believe helped us and recommend the same treatments to our
friends. We see the same phenomenon on TV when journalists have patients in
studios reporting that just the fact these patients are still alive and free from, say,
cancer symptoms is proof treatments work. Perhaps a particular cancer type
usually kills within a year, yet this patient was diagnosed five years prior.
Convincing to most people, it is not convincing if you know about tumour
biology. The patient in the studio is not like a top athlete who, thanks so some
miraculous performance enhancing drug, can suddenly throw a javelin 400
instead of 80 meters. A javelin is always a javelin, but the cancer might not
always be cancer: the diagnosis could be wrong. Even if correct, cancers are
heterogeneous diseases, so some people will be expected to live far longer than
average. Besides, we also now know that some cancers, including breast cancer,
may regress spontaneously and disappear without treatment, 15,16 which you
cannot expect average TV journalists to know. But you could certainly expect
journalists to do their homework by asking experts about the issues instead of
producing misleading broadcasts.
The media is not a trusted source of information. Media outlets sell newspapers,
attracting viewers and listeners by portraying us as sicker and more worried than
we need be. A common trick is to write articles about the diseases we might be
suffering from without knowing. For three months, a Danish newspaper gathered
news stories about what Danish people suffer from and deduced that every Dane
suffers from two diseases, on average. 17 In fact, it is much worse because the
journalists searched for what “ Danes suffer from,” which means that a lot of
‘diseases’ were overlooked. An issue of semantics also exists. You cannot suffer
from a ‘disease’ without symptoms, e.g. you do not suffer from slight increases
of blood pressure, blood glucose or cholesterol. These conditions are not even
diseases - only risk factors, which some doctors, usually on drug industry
payroll, have labelled ‘abnormal.’
The criteria for ‘abnormal’ and needing of drug treatment have been pushed so
absurdly lower over the years that, applying European guidelines for
cardiovascular disease, researchers found that 86% of Norwegian males were at
high risk of cardiovascular disease by age 40, 18 despite the fact that Norwegians
are some of the most long-lived people in the world. In another study, the
researchers found that 50% of Norwegians had a cholesterol or blood pressure
level above the recommended cut-off for treatment by age 24! 19
Doctors cannot learn from what they cannot see. Prescription drugs kill
approximately 200,000 people a year in the United States, and about 3,300 in
Denmark 11 - 20 times as many as traffic deaths. Half of those who die take their
medications as prescribed; for the other half, errors are involved such as
overdoses, drug interactions or treatments with contraindicated drugs. Doctors
rarely realize that some of the deaths occurring in their practices are caused by
drugs they, themselves, have prescribed. These deaths are so common that, on
average, every general practitioner kills one patient each year. 20 Thus, if
practicing for 35 years, a doctor will kill 35 patients. Some drug-induced deaths
occur in hospitals. Yet - by far - most of these deadly drugs are prescribed by
family doctors. Among the biggest killers are NSAIDs (arthritis pain relieving
agents such as ibuprofen) 11 and drugs for mental health issues. 13 In general
practice, the consumption of these drugs is very high: NSAIDs and depression
pills are used 38 times and 76 times as often, respectively, as in hospitals. 21 My
estimate of one death per general practitioner per year is therefore reasonable.
The causes of almost all of these deaths are invisible to doctors and, therefore,
even to their patients, the public and our drug regulators. When a patient dies
due to an NSAID, the drug may have caused a stomach ulcer or a heart attack,
but that might have happened without medical treatment, anyway. Frequent
causes of death due to depression pills and other brain-active drugs are balance
issues. 13 When older patients fall and break their hip, about one in five will die
within a year. Their doctors do not think these deaths may be due to their drugs,
because so many old people not taking medication fall and break their hips.
Most drugs in common use are likely to affect the brain and may lead to falls. In
the elderly, drugs against increased blood pressure, for example, should be used
with caution.
Once the boat capsizes and you, in a panic, are thrown into the waves, it is too
late to regret never learning to swim. We all fall ill from time to time and
sometimes quick decisions are needed in situations where you might be scared to
death and not fully rational. You may not be able to ask the appropriate questions
of your doctor - or of yourself. But you will be able if you acquire the necessary
routines and experience. That is why you should go online and gain experience
while still healthy rather waiting until you ‘fall out of the boat,’ so to speak.
As already noted, I suggest you not only read this book carefully, but also go
online and work through the examples provided. Breaking a mental barrier that
is telling you, “It’s too difficult” is very important. Once you have tried it a
number of times, you will realize that obtaining information to guide your
decisions is possible. And the more you practice, the better you become.
Therefore, use other examples, as well, including your own and keep trying until
you get results. After a while you will feel that you are mastering what you
believed to be impossible.
There is no shortage of good examples to work with. People are very interested
in health issues and all sorts of things come up, even during dinner
conversations. When in doubt, tell your friends you will try finding the answer
until next time you meet. That might impress them so much that they ask you
how they can to do it themselves. That is how we all can help creating better
healthcare and even more agreeable dinner parties. When people have strong, yet
opposing, opinions about healthcare (often the case) discussions can go astray.
Yet if someone says he or she will look it up, that could defuse the situation.
Sometimes, I do that on the spot, during coffee. But I do admit: quickly finding
answers takes some practice.
The public is well aware of the need to be critical. In a survey, two thirds of
British adults agreed with the statement that industry trials were often biased to
produce a positive outcome. 23 Only a third of the public trust evidence from
medical research, in general, whereas two thirds trust what their friends and
family tell them about medications.
Do not. Your family and friends get their knowledge from personal experience,
which is very unreliable, or from their doctors, which are also unreliable, or from
newspapers, magazines, TV and radio, which are equally unreliable. There are
no shortcuts here. You need to look up the evidence yourself.
Utilitarian thinking in public health
Public healthcare is about providing the greatest good for the greatest number. In
medical ethics, that is called utilitarianism. In a randomized trial, if an
intervention led to fewer deaths than in the control group - who might have had
no intervention at all – in all likelihood, it will be popular with politicians and
recommended as a national guideline.
For everyone getting that intervention, that would seem to be a no-brainer, but it
is rarely that simple.
Many questions must be asked. What does 25% lower mortality mean? Many
people interpret this to mean that some fortunate people will not die from a
disease. Yet perhaps death is only postponed, and they die from that disease,
anyhow, just a little later. That is the case for the vast majority of cancer drugs,
for example, yet their effects are still considerably hyped despite their nasty
harmfulness. I will return to this in Chapter 10.
When does the mortality benefit occur? There is a huge difference between
prolonging the lives of children and prolonging the lives of old people who
might soon die of something else, and who might be in such poor condition that
life would not be worth living.
How common is it to contract a disease and die from it? If a disease is rare, the
chances of benefiting from a prophylactic intervention may be so small that most
people would deduce that they should abstain from the intervention.
Public health campaigns are remarkably silent about these issues and usually
have propaganda characteristics. However, turning healthy citizens into patients
comes with a cost. Feeling free, happy and healthy – and being medicine-free -
has great value. Taking these privileges away from people and turning them into
anxious patients needing doctors - perhaps even hospital - can be quite harmful.
Finally, intervention benefits and harms are not measured on the same scale and
decisions about whether benefits outweigh harms are therefore subjective ones.
They are not decisions authorities can make for us. We need to make them
ourselves. Yet we can only do so if properly – and honestly - informed.
The subjectivity of all of this becomes quite clear if we look at traffic deaths.
After we introduced speed limits, road deaths reduced markedly. But how low
should we go? If we set the maximum speed for all vehicles at 30 km an hour,
we could reduce traffic deaths even more, but such a proposal will never gain
public support. Setting speed limits is completely arbitrary, and that also applies
to the limits for what is considered ‘normal’ in healthcare. Unfortunately,
guidelines are often written by people too close to the drug industry, and these
people are also the ones who conduct trials that provide guideline foundations.
11,13
You should ask critical questions if your doctor says you should be treated with a
drug indefinitely because your blood pressure, cholesterol or glucose is too high,
or because the density in your bone scan is too low. You might be better off
doing nothing or even something other than take drugs. I shall comment more on
this later.
Greater flexibility was one advantage of the old system, but that was also a
disadvantage. Different doctors treated similar patients differently depending on
the doctors’ personal preferences, prejudices and experience - and on whether or
not they allowed drug sales representatives to visit. Sometimes doctors used
hopelessly outdated and dangerous drugs because they failed to stay up to date,
e.g. chloramphenicol, a drug that causes bone marrow suppression, with a high
mortality rate.
Before the advent of computers, there was very little oversight of doctors’
activities. Nowadays, prescription data helps health authorities to spot aberrant
doctors who, for instance, use far too many tranquillizers or opiates, allowing
authorities to take action.
That is good, but now the pendulum has swung too far in that direction. We have
a plethora of long, complicated clinical guidelines telling doctors what to do.
Young doctors in general practice feel obliged to follow them, even when
patients are in circumstances which should make them disregard inapplicable
guidelines.
Doctors can ignore guidelines if they explain in patient files why it is appropriate
to do so, but few doctors are prepared to do that. Doing what you are told is
much easier than facing trouble with colleagues. That happened to me just two
years after taking my medical exam. I diagnosed a mild case of type 2 diabetes
in an old man who was admitted for something else. 11 I noted that the common
practice was to start tolbutamide, but since the only large trial of tolbutamide
ever performed was prematurely stopped due to an excess of cardiovascular
deaths, and since those patients who took most of their prescribed medication
were also those that had the highest event rate, I decided not to use tolbutamide.
Here is an example of how gullible we humans are: Mark Spitz won seven gold
medals in swimming at the 1972 Summer Olympics in München. 1 He swam
with a moustache in an era when other swimmers shaved their body hair. A
Russian coach asked him if his moustache slowed him down and he said, “No, as
a matter of fact, it deflects water away from my mouth, allows my rear end to
rise and make me bullet-shaped in the water, and that's what have allowed me to
swim so great.” The following year, every male Russian swimmer had a
moustache. The truth was that Spitz grew a moustache because a college coach
said he was not allowed to grow one. He planned to shave it off before he went
to the Olympics but decided to keep it because everybody was talking about it.
We get more advice from family and friends about healthcare than from any
other source even though very little of that advice is helpful - based on good
evidence. Unfortunately, the same can be said about doctors and other healthcare
professionals. For example, most doctors know little more about drugs other
than what the drug industry tells them. And most of that is incorrect. 2,3 We need
more reliable information sources than that.
For information retrieval in general, we now have Google and Wikipedia, which
are my favourites. I have not been paid to say that. There are other search
engines than Google, but I am not familiar with them. When looking for the
values of diagnostic tools and treatments, I find the Cochrane Library and
PubMed the most useful sources. Reading about these sources is a bit boring, so
before we get to that, I shall provide some examples of important clinical
questions.
I will attempt to explain every step in detail to ensure that you can do it by
yourself, even when it comes to your own clinical questions. Far too often,
intermediate searching steps are missing, especially when guides are written by
experts, because experts believe these steps to be so obvious they need not
include them. Thus, when you are at A looking at C, but cannot find B, you give
up.
Back pain
Formulating questions in the best possible way takes some practice. When we
have health problems, we tend to think in very broad terms such as, “What
should I do about my back pain?” Before we can look up any possible answers,
we need to consider a number of issues, for instance:
Back pain can be located everywhere in the back, but say your pain is in the
lower back, which is the most common location. You can read that nonspecific
back pain is believed to originate in soft tissues such as muscles, fascia, and
ligaments.
That is already important information. Many people with back pain visit
chiropractors or doctors who manipulate their spines. Yet if the pain has nothing
to do with small vertebral misalignments - called subluxations - in the spine,
there is little chance of benefit. I will get back to this in Chapter 13 on
alternative medicine.
Back pain is rarely permanently disabling, and most cases of herniated disks and
stenosis have similar outcomes after a year of rest, injections or surgery. Aha!
You just found out that if you have a herniated disk you should usually refuse
surgery. Too bad for those earning a living on back surgery - but good for you!
As stated, you do not ask a barber if you need a haircut and you should not ask a
back surgeon whether you need back surgery. Most back surgeries carried out by
US surgeons should not have been done. In the 1990s, Republicans tried to shut
down the US Agency for Healthcare Research and Quality after spine surgeons
objected to findings that rest and pain medications work just as well for back
pain as surgery. 5
If you consult a surgeon, always get a second opinion from someone who is not
a surgeon and even look the evidence up yourself. In contrast to drugs, surgery is
irreversible.
According to Wikipedia, the prognosis is good. In most cases, back pain goes
away naturally after a few weeks. About 98% of back pain patients have
nonspecific, acute back pain with no identified serious underlying pathology. For
the remaining patients, we find cases of metastatic cancers and serious
infections. Does that mean that all patients with back pain should be thoroughly
examined for possible cancers and infections with a CT scan, for instance? No.
Virtually every one of us will experience back pain. We cannot examine the
entire population for each and every possible cause of this pain. That would be
hugely expensive; there would be a great number of false positives; and it would
cause a lot of harm. CT scans carry a risk of causing cancer and all those futile
treatments performed on those who had received false positives would also be
harmful. Therefore, detailed examinations are not carried out unless a reasonable
suspicion of a serious underlying pathology is present. You can read about this in
the Wikipedia article.
The pain may radiate into the arms or legs and may include tingling with no
apparent cause (paraesthesia), weakness or numbness, all of which are symptoms
of neurologic impairment resulting from a herniated disc. If these circumstances
continue to progress, or if there are signs of bowel or bladder incontinence, then
the situation is serious and requires immediate hospital admission.
The Wikipedia text tells us that X-rays and other medical imaging scans are not
useful. We are also told that two of the conditions back pain is often attributed to
- lumbar disc herniation and degenerative disc disease - may not be more
prevalent for those in pain than for the general population.
As I will explain in Chapter 13, chiropractors very often take X-rays and tell
patients they can pinpoint their problems. But you must not believe them.
What about treatment? When we come to this crucial issue, I do not recommend
you rely on Wikipedia. Medical literature is far too unreliable to leave to
Wikipedia volunteers. These volunteers rarely have extensive training in
evaluating trials and systematic reviews and, therefore, are unequipped to
explain the benefits and harms of treatments. For that, use reviews with high
standards - like Cochrane reviews.
A systematic review is a review that employs predefined methods for finding and
critically appraising all relevant articles on a given subject. If more than one
article exists on a particular subject, the results are often combined statistically
in a meta-analysis.
The Wikipedia text on back pain illustrates this because it warns against its own
recommendations: “This section needs more medical references for verification
or relies too heavily on primary sources. Please review the contents of the
section and add the appropriate references if you can. Unsourced or poorly
sourced material may be challenged and removed. (January 2016).”
Wikipedia tells us exercise can be effective but should only be done under
supervision of licensed healthcare professionals. I have never seen such a
recommendation before and can see no good reason why people with back pain
should not exercise without being supervised.
It is also odd when the article mentions that, “A study with 80 patients found
magnesium to be useful in those with chronic back pain.” One study? Do other
studies reach other conclusions? What is the likelihood that magnesium is useful
for back pain? About zero. Why would a metal ion, of which we already have
plenty in our bodies, have an effect on back pain?
When the a priori likelihood that something works is very low, we must demand
extraordinarily convincing data that it works. We will not find that in a single
study of 80 patients - the risk of fraud is far greater than the likelihood that
magnesium works. Do not bother downloading a study like this. But - for fun - I
looked up its abstract on PubMed (see below on how to find studies on
PubMed).
We do randomized trials to compare one group with another, but the abstract
only tells us what happened in one of the groups. Only telling readers what
happened in one of the two randomized groups is totally misleading. Usually
patients improve in both groups because they enter the trials when their pain is
worse than usual.
I clicked the link to the full trial report on PubMed and downloaded it. A
significant difference existed between magnesium and placebo at 6 months (P =
0.03), with a value of 7.2 in the placebo group. Yet after two weeks, pain was
reduced from 7.5 to 3.4 in the magnesium group, and from 7.4 to 3.6 in the
placebo group, which is virtually the same (P = 0.28, meaning magnesium had
no effect). Why did the pain go back to baseline in the placebo group but not in
the magnesium group? That makes no sense. A P-value of 0.03 at 6 months
under these circumstances cannot justify using magnesium for chronic back
pain. Therefore, I did not read more of the paper but searched on magnesium low
back pain on PubMed. At the right of the PubMed site, you can see the search
details.
My search yielded 15 records, none of which was a magnesium trial other than
the Egyptian one. I do not believe we will hear about magnesium for low back
pain again.
Now, you need to know a little about P-values. P = 0.03 means that, if a
treatment is no better than the comparator, we will by chance get the difference
we observed - or an even bigger difference - in three out of a hundred trials. If P
< 0.05, we say that the result is statistically significant and probably not a chance
result. However, the literature is full of significant P-values which must not be
believed. Since very few tested treatments are any better than the comparator,
that means – mathematically - most significant P-values are misleading. More
pragmatically, it is much worse than that. Not only the drug industry but even
academics are very good at a game Americans call “torture your data until they
confess.” Data massage is very common and that means the lion’s share of
significant P-values in abstracts is misleading. 7 Even within the articles
themselves, many significant P-values are misleading. In fact, it is more likely
for research claims to be false rather than true. 8 Therefore, do not get impressed
just because a P-value is below 0.05.
I do agree that the patients’ attitudes to pain are very important. Some of us
ignore pain and dismiss it as part of life, while others obsess on it. That is one
reason why some patients who complain about pain somewhere in their body are
beyond therapeutic reach. No matter what we try, they will say that it does not
work. In this type of clientele, we find an abundance of abusers of opiates and
other drugs. So-called pain clinics can be very harmful because they fill patients
with drugs in a non-evidence-based manner, instead of realizing that many
patients should not get drugs at all – they need psychotherapy. And if that does
not work, we might decide to give up. Doctors cannot help everybody, and some
chronic pain patients clearly have psychiatric problems. All doctors have
encountered such patients.
Doctors are good salespeople - too good, in fact. They will often say things like,
“This drug will work for your back pain,” “There are no side effects,” or “This
new arthritis drug will not give you stomach problems.” You really need to be
alert and ask (at least yourself): “Help me in what way and to what degree?”
Most beneficial effects of drugs are trivial and not worth pursuing given the
costs of the drugs, both monetarily and when it comes to the adverse effects.
That is why you need to look up the evidence yourself: your doctor will only
very rarely be able to give meaningful replies to your questions. There are, of
course, no arthritis drugs that will not cause stomach trouble, including bleeding
ulcers. If lower back pain disturbs you so much that you consider taking pills,
you might want to know whether what many doctors say is true: arthritis drugs
(NSAIDs) work better than paracetamol (acetaminophen).
The simplest search you can imagine often leads to useful results. In this case, I
searched “NSAID paracetamol back pain” on Google. Even though I did not add
Cochrane as I usually do, the second entry on the first page appeared to be a
relevant Cochrane review. One of the many useful Google features is you can
see what articles are about without opening the links:
Non-steroidal anti-inflammatory drugs for low-back pain | Cochrane
www.cochrane.org/…/BACK_non-steroidal-anti-inflammatory-drugs-for-low-
back-pa…
Jan 23, 2008 - Non-steroidal anti-inflammatory drugs for low- back pain…. In
patients with acute sciatica, no difference in effect between NSAIDs and placebo
was found. The review authors also found that NSAIDs are not more effective
than other drugs ( paracetamol /acetaminophen, narcotic analgesics, and muscle
relaxants).
If you look up the review, 9 you will find that the authors conclude in their
abstract: “The evidence from the 65 trials included in this review suggests that
NSAIDs are effective for short-term symptomatic relief in patients with acute
and chronic low-back pain without sciatica. However, effect sizes are small.”
Limited pain effects in industry-sponsored trials are not worth pursuing,
particularly when we know these drugs are quite dangerous and take very many
lives. 2
Clinical guidelines
Doctors often rely on clinical guidelines for their practice but whatever the
origin of these guidelines - be very careful. Whether they have been written by
specialist societies, a national board of health, health-technology assessment
agencies, or the World Health Organization - be careful! I rarely read them
because they are often misleading, not reflecting the most reliable evidence we
have, but rather the prejudices and financial conflicts of interests of the authors
or financers, who might be politicians with strong views or a good nose for what
the voters want to hear. Even when guidelines are reasonably good, they may not
acknowledge the poor quality of the studies upon which they are based.
In healthcare, being idealistic and honest - recounting exact findings; calling a
spade for a spade - is not easy. I have seen many very good initiatives, clearly
both in the patients’ and taxpayers’ interest, being closed down or threatened by
drastic funding reductions if its initiators did not ‘behave.’ Whether your
activities have saved your country huge amounts of money does not really
matter. In fact, if you have, the risk you will soon disappear might be even
greater because it means you stood up to powerful financial and political
interests. I have met true pioneers in Canada, the United States, the United
Kingdom, Denmark, Holland, Germany and Australia whose wonderful
initiatives were stopped or amputated.
The Nordic Cochrane Centre, which I established 25 years ago, has also been
threatened, indeed many times. I obtained permanent funding for our institution
from the Danish government in 2001. That investment saved billions of Danish
crowns. Three of our reviews saved taxpayers over 500 million Danish crowns
annually for many years, - approximately 100 times our annual budget. 10 These
reviews were about mammography screening (first published in 2001), 11 alpha-
1 antitrypsin for treatment of patients with lung disease who have this enzyme
deficiency (first published in 2010), 12 and general health-checks. 13 The Danish
National Board of Health asked us to do the mammography screening review
and the Committee on Health in Parliament requested the lung disease review.
The health-check review was our own idea.
In all three cases, our findings were quite negative, had direct political
consequences and threatened powerful interests. Over the years, the Danish
Cancer Society and the Danish Pharmaceutical Industry Association have tried
to convince multiple Ministers of Health that they should close the Nordic
Cochrane Centre. My strategy has been three-pronged: maintaining good
relations with speakers on health in Parliament; demonstrating how useful we
are via our research; and being very visible in the media. Yet there is no
guarantee of fair sailing. When we had finished our review on health checks in
2011, I asked for a meeting with the minister who decided - on the spot - she
would cancel the new government’s plans of introducing regular health checks in
Denmark. However, that same minister threatened my job as director of the
Nordic Cochrane Centre just two years later. 3 I had published a newspaper
article about ten myths harmful to patients in psychiatry and concluded that our
citizens would be far better off if we removed all psychotropic drugs from the
market. Doctors are unable to handle them, and their availability does more harm
than good. I also said that, in coming years, psychiatry should do everything it
could to use psychotropic drugs treatment as little as possible, as briefly as
possible, or not at all. My article was in Danish, but it has also been published in
English. 14
My article caused little stir until two months later when the Danish Psychiatric
Association attempted character assassination and almost succeeded. A media
storm of false allegations ensued which, in reality, was nothing more than a
witch hunt. 3 The minister reacted. In such situations, facts do not matter,
including the fact that psychiatrists and family doctors cause their patients
immense harm and even kill many with psychiatric drugs. 3 Shooting the
messenger is so much easier than changing the system. However, a year later in
2015, I published an entire book about psychiatry where I documented how
harmful the specialty was in detail. 3 This time, I received no threats, and found
great support from many patients and their organizations - and even from some
psychiatrists. The patients nominated me for Dane of the Year and I ended in top
ten. They also made me Protector of the Danish Hearing Voices Network and a
filmmaker made a film about me - Diagnosing Psychiatry - about what I wish to
accomplish in psychiatry. 15
Clearly, people’s views about me differ. Journalists often ask if I get many
enemies. I tell them I do, but also tell them about the friends I have acquired -
some of the best ever imagined.
There is a good deal of ‘feeling good’ in guidelines´ and too little concern about
the harms of the suggested interventions. Doctors and other ‘do-gooders’
experience great difficulty accepting when their efforts do not work, and even
more difficulty accepting when they have nothing to suggest to patients or
healthy citizens. They do not like admitting being superfluous at times and that
they will only do harm if they intervene. Sometimes it is best to let nature run its
course.
In ancient times, a similar visual art phenomenon was called horror vacui (Latin)
or kenophobia (Greek), which means fear of empty space or fear of emptiness.
This anxiety led to filling entire surfaces of spaces or artworks with details.
Nowadays, the horror of emptiness means filling patients with pills, and
nowhere does that lead to more horror than in psychiatry 3 (see Chapter 8).
Perhaps I should say that I have no personal grudge against psychiatry and that I
have never had a psychiatric problem or been under psychiatric care.
A lot has been written about guidelines and their fallibility. I will provide two
examples.
“New guidelines for prostate cancer screening. The new recommendations for
PSA tests say men ages 55 to 69 should ‘make an individualized decision about
prostate cancer screening with their clinician.’ The draft guidelines update a
blanket recommendation in 2012 for no routine screening at any age.”
It usually goes that way. There are always powerful people who - despite clear
evidence of an intervention not working or, as in this case, doing more harm than
good – keep pushing the cart until it gets to where they want to go. We doctors
tend to forget the most famous of all quotes ascribed to Hippocrates: “First, do
no harm.” If I were still a practicing clinician, I would never agree to order a
PSA test for a healthy person. The incidence of prostate cancer corresponds to a
person’s age, so that about 60% of men who are 60 will have cancer. Finding and
treating all these cases of cancers would lead to immense misery in terms of
myriads of impotent and incontinent males. Since we cannot distinguish between
harmless cancers - the vast majority - and dangerous cancers, we treat them all.
We should therefore not look for them in people without symptoms with PSA
tests.
If you google psa cochrane, you will find the Cochrane review. 16 The patient
summary says that screening did not significantly decrease prostate cancer-
specific mortality or overall mortality in a meta-analysis of five trials. People are
warned about overdiagnosis, overtreatment and treatment-related harms. Yet the
review authors recommend that, “Men should be informed of this and the
demonstrated adverse effects when they are deciding whether or not to undertake
screening for prostate cancer.” It appears to me that the authors have confused
citizens and doctors in this sentence. Men cannot require a test of their liking.
You cannot go to a hospital and demand a CT scan. Doctors decide whether or
not they will screen for prostate cancer and men then accept or decline the offer.
But doctors should not offer prostate cancer screening at all.
Our findings were highly unwelcome. After we had approved a version for
publishing in 1998, the editor changed our abstract without telling us, so it
became misleading, favouring the interventions. We incidentally detected this
and complained about it. Some years later, when we updated the review with
new trials, the editor changed our abstract again – once more without our
permission. That is not how Cochrane editors - or any editors - are supposed to
function, and I am pleased to say that this person is no longer editor of the group,
even though his departure was not related to his editorial misconduct.
Here is what happened. The first time, in 1999 in issue 1 of the Cochrane
Library, our conclusion in the abstract was:
Since the reviewers were us, this could not be the "reviewers' conclusions!" The
wording, "there is not enough evidence to show" implies that if there had been
more evidence (for example, including the large, planned UK trial) we would
have shown that the methods were effective. That was seriously misleading. We
had shown with narrow ‘confidence intervals’ (CI - see below what that is) that
we could not have missed a worthwhile effect, and we had explained it carefully
in the discussion section of our review.
We published the most recent version of our review in 2011 (officially dated
2008, because of only one additional trial in the 2011 update). 18 There was still
no trace of an intervention effect, and the large UK trial made no difference to
our results.
So, what did the experts say about our systematic review, which was widely
known and had also been published in the BMJ, when it first came out nine years
earlier? 23 Absolutely nothing. If you cannot beat them, ignore them! There were
hundreds of references after the section that included house dust mites, all
looking very impressive and evidence-based, which they clearly were not.
In a letter to JAMA, I pointed out that the experts did not try hard enough
because all three references offered in support of their recommendations for
house dust mites were irrelevant. I also mentioned that we found that the average
effect of the interventions on the peak expiratory flow rate (the most common
outcome in asthma trials) was exactly zero, with a very narrow confidence
interval. JAMA refused to publish my letter.
Patients should not be lured into wasting their money and energy on useless
interventions such as highly expensive super-vacuum cleaners, mattress covers,
obsessive cleaning, air filters and throwing out carpets. What is most depressing
in all this is that allergy experts should know that interventions against mites
cannot be expected to be effective, because the possible reduction in allergens is
far too small to be effective. 19
Nonetheless, guidelines and review articles continued to proliferate all over the
world, recommending interventions against house dust mites. National boards of
health and patient organizations did the same. We were so appalled that we
published a paper showing how misleading narrative review articles on dust
mites written by asthma specialists are. Inspired by John Steinbeck’s novel Of
Mice and Men, we called it Of Mites and Men. 26 These reviews usually
recommended several methods as being effective, providing a highly selected
and biased sample of references in support of their claims. In 70 reviews, the
most quoted trial had only 7 patients per group. The significant result claimed by
this trial was probably erroneous and it did not even report a clinical outcome.
The recommendations were often based on non-randomized studies, of which
the most quoted study had only included 10 patients per group, yet still claimed
very positive results. In contrast, we have 55 randomized trials and 3121 patients
in our Cochrane review.
You might not have the slightest interest in house dust mites, but I hope you
enjoyed the story. It illustrates how everybody can get it wrong and, deliberately,
continue getting it wrong in direct contrast to scientific data.
To see if it was equally bad elsewhere, I googled “ dust mites guidelines” and
restricted my search to the previous year using Tools. The second link went to
the website of the world-famous Mayo Clinic in the United States. The text was
from May 2017 but utterly hopeless - even more futile and expensive
recommendations than in the Italian survey were on offer. And no references.
Without references, there is no accountability.
The charity Asthma UK ( asthma.org.uk ) performs much better than this. Their
director says that, “If your asthma is triggered by dust mites, the best way to
reduce asthma symptoms is to look after your asthma and make sure it's well
managed, as this reduces the likelihood of you reacting to the dust mite
droppings when you come into contact with them - they are impossible to
avoid."
True. Yet even this organization could not resist the temptation to spread false
hope: “One recent study did find that children using mite-proof covers on their
mattress, duvet and pillowcases were less likely to be hospitalised with an
asthma attack – but only those aged between 3 and 10, living in non-smoking
households, and who were allergic only to dust mites (not pets or pollen, for
example). More research is needed before we know for sure whether (and how)
they could help.”
No! More research is not needed. The reference to this “recent study” is missing
and whether or not it was a randomized trial is not stated. No accountability
again. Further, we are presented with a subgroup result, not the overall result.
That should ring all our alarm bells. Please ignore irresponsible messages like
this. No less than 26 of the trials in our Cochrane review had used mattress
covers. They do not work, for pretty obvious reasons.
Grading evidence by reliability
Many information sources contain statements telling others what to do. These
recommendations need to be graded according to the confidence we have in the
science that lies behind them. The GRADE System is commonly used for that
purpose.
The reliability of evidence is also called the quality of evidence. More than
anything, it depends on the risk of bias in the underlying studies. The most
reliable studies are randomized trials where patients, through chance procedures,
are divided into two or more groups that are treated differently. The results are
then compared. Technically, a table of random numbers is often used and there
are many ways to do so. One of the best is utilizing a central randomization
office to which clinicians send the details of patients they wish to be included in
the trial. If a patient fulfills the inclusion criteria, that patient is randomized by a
click on a computer, making the group assignment appear. The group assignment
is final and cannot be changed by the clinicians if they get second thoughts
believing a patient might be treated more appropriately in another group. We call
this process ‘concealment of the allocation.’
If a study has not been adequately ‘blinded,’ a high risk of bias exists unless the
outcome is objective - such as death. Research has shown by just how much
results can be exaggerated when assessments are not blinded. In trials for a
variety of diseases which had both a blinded and a non-blinded observer,
treatment effects on a binary scale - such as improved or not improved - was
overestimated by an average of 36% (measured as odds ratio) when the non-
blinded observer assessed the effect compared to the blinded observer. 29 In
another such study, which used measurement scales such as the severity of
depression, the effect was overestimated by 68% on average. 30
People who have never conducted randomized trials have no idea about how
much they can - and actually do - go wrong despite carefully written trial
protocols. Patients may leave trials prematurely because they experience adverse
effects or lack of effect, or they may receive additional treatments although
prohibited by the protocol. The general rule is to respect randomization, ensuring
that groups are comparable. Therefore, whatever happens, patients are analyzed
in the groups to which they were randomized even if they get the wrong
treatment by mistake. That is called an intention-to-treat analysis. If the only
patients analyzed are those who do everything they are told and remain until the
last visit, we call this a per-protocol analysis. Such analyses can be seriously
flawed because the groups are no longer comparable.
The best trials are when even the data analysis is blinded, 31 because values and
patients are often missing (lost to follow-up) and many other issues also exist
where decisions need to be made that can potentially bias the trials.
Then we have the issue of publication bias, which means that positive results
have a higher chance of getting published than negative ones. That risk also
needs to be considered. If a trial is very large, it is almost always published, no
matter what the results showed, which is one of the reasons why large trials are
more reliable than small ones. Large trials also yield more precise results. If a
treatment is twice as effective as another, you might find that 10 out of 40
patients in one group improved versus 5 of 40 in the other group. That gives a
risk ratio of 2.00 with a 95% confidence interval (CI) that goes from 0.75 to
5.33. If you conduct several trials, each with 40 patients in each group, you will
reach other results. Yet, based on one single trial, you are 95% sure that the true
risk ratio lies between 0.75 and 5.33. If that trial had been larger, with 1000 of
4000 patients improving in one group and 500 of 4000 in the other, the risk ratio
would still be 2.00 but the 95% CI would be small because it went from 1.81 to
2.21. In this case, you are much more certain that one treatment is twice as
effective as the other.
When grading research, we also consider whether results from study to study are
similar. If results vary greatly, some findings might not be reliable - often the
smallest studies are the problem and should be ignored.
One final issue to consider is whether the trial patients, interventions and
outcomes are similar enough to our own that we can use those results as a guide
for treatment of our own patients.
Studies that are not randomized come in many forms and are often collectively
called observational studies, because what takes place is observed without
interference. In cohort studies, groups of people are followed for periods of time.
If treatments are then instituted or stopped, it is considered common clinical
practice. 32 A comparison cohort not being treated may also be included. Such
studies are much less reliable than randomized trials because cohort members
are not comparable to begin with - at baseline. There are all sorts of reasons why
they receive different treatments and, therefore, any difference in the outcome
should be interpreted cautiously.
When randomized trials cannot be carried out, cohort studies may provide
alternatives. They can be useful supplements to randomized trials, which are
often too small or too short to identify rare or slowly developing harms. They
may also elucidate the benefits and harms of new treatments which have become
routine and utilized by less well-educated clinicians. For instance, laparoscopies
- key-hole surgeries - might be performed less satisfactorily in the hands of less
well-trained surgeons, and a cohort study might disclose that, in some settings, it
leads to more harm than traditional surgery. Cohort studies are also beneficial in
studying patients not included in carefully controlled randomized trials, such as
elderly patients with several diseases and treatments. Finally, such studies can
generate interesting hypotheses that can be tested in randomized trials.
Even less reliable than cohort studies are case-control studies where
investigators select a number of cases and controls, and then tally whether
exposure to supposed causative agents is more common among the cases than
the controls. 32 This design is often used to detect rare forms of harm that would
otherwise require large cohorts to be identified. If, for instance, children with a
particular rare heart defect have, to much greater frequency, mothers who took a
certain drug during pregnancy than mothers of healthy children, we would
suspect that this drug caused the heart defect.
Every type of research design has its place in healthcare research. But if a design
is being used for the wrong purpose, it can go seriously wrong. An infamous
example is mammography screening. Professionals favouring this type of
screening have often used flawed methods to convince others that screening
substantially reduces breast cancer mortality. They far too willingly accepted the
results of case-control studies 33 despite existing consensus among breast cancer
screening experts that such studies cannot reliably determine the effects of
screening. Using data from the Malmö randomized screening trial, it was
demonstrated that the bias introduced by such studies was huge. When properly
analyzed as a randomized trial, the reduction in breast cancer mortality was only
4%, but when analyzed as a case-control study (comparing breast cancer
mortality in attenders with non-attenders within the screening arm), the
reduction in breast cancer mortality was 58%. 34
Conflicts of interest
Healthcare is riddled with conflicts of interest and has been corrupted by
industry money to such an extent that it is likely the most corrupt sector in our
societies. 2 Serious fraud is more common than in other industries, and the drug
industry business model includes organized crime, which is widespread, costly
and deadly. 2,3
Therefore, we should not trust but distrust healthcare individuals and
organizations by default. Sometimes they are right and sometimes wrong; the
problem is we rarely have any inkling either way until we have looked up the
evidence ourselves.
What is abundantly clear is that our drug agencies should be distrusted. If they
had been doing their work properly, our drugs would not have been the third
leading cause of death, and a large portion of the population would not have
been physically and mentally incapacitated, killed or permanently harmed by
rather useless psychiatric drugs. 3
Agencies that were previously respected have eroded over time because they
sacrificed some of the principles that made them respected in the first place.
Treating hepatitis C
Conditional donations to the CDC are earmarked for specific projects. In 2012,
Genentech earmarked $600,000 in donations to the CDC Foundation for its
efforts in promoting increases in testing for, and treatment of, viral hepatitis. 37
Genentech and its parent company Roche manufacture test kits and treatments
for hepatitis C. That same year, CDC issued guidelines recommending screening
for hepatitis C virus in everyone born 1945 - 1965, even though the science
supporting screening was questionable. 37 Two years earlier, CDC formed a
coalition which supports research and promotes expanded testing for and
treatment of hepatitis C, and companies selling tests and treatments for hepatitis
C infection donated over $26 million to the coalition. Conflict of interest forms
showed that 9 of the 34 members of the external working group writing and
reviewing the new CDC recommendations had financial ties to these
manufacturers. 37
By googling sofosbuvir cochrane, you will find the top link goes to the Cochrane
review of sofosbuvir and similar drugs. 38 However, when I first tried this, the
initial link was an abstract of a systematic review from 2014 39 which stated that
the authors had searched for studies in the Cochrane Library. That looked
promising but my interest vanished when I read the whole abstract. All authors
are from China and the fact that medical research fraud is very common in China
has been well documented. 40,41 In 2016, an investigation of data for 1,622 new
drugs submitted for registration to China’s State Food and Drug Administration
concluded that 81% of the applications should be withdrawn because they
contained fabricated, flawed, or inadequate clinical trial data. 41 Apparently, in
one example, data had been recorded by staff who had not yet started their
employment. Another study found that only 7% of 3,137 trials published in
Chinese journals were actually randomized even though all were called
randomized. 42
We cannot dismiss papers solely because they come from China but we can
choose to be more alert. The authors of that review did not distinguish between
randomized trials and observational studies. For instance, they wrote in the
abstract, “One trial and 13 treatment arms/cohorts from seven studies …” They
concluded that sofosbuvir was “effective and safe,” which is a type of industry
jargon that automatically stops my reading.
The next link was a systematic review in JAMA, also from 2014. 43 However, the
abstract looked more like something from a treatment guideline than a
systematic review - there were surprisingly few data and no confidence intervals
or P-values. The last sentence was: “In conjunction with increased screening for
HCV, as suggested by recent Centers for Disease Control and Prevention
guidelines, availability of new therapies may lead to the treatment of many more
people with chronic HCV infection.”
The authors refer to screening which is not an evidence-based recommendation
(see below) and are pleased this might lead to treatment of many more people
with drugs that are so expensive they threaten the national economies - even in
wealthy European countries - if everyone with an infection were to be treated.
That was too much for me. It looked like a glossy brochure aimed at investors
for companies selling these drugs. Furthermore, the outcome was sustained
virologic response, which is not what we want. We want to reduce the
occurrence of liver cirrhosis and liver cancer, and we want to reduce deaths.
Virologic response is a surrogate marker and we cannot know if a decreased
viral load in the blood will have the desired effects on outcomes that matter to
patients. Some antibiotics might have good effects on bacteria in a laboratory
Petri dish, yet patients might die anyway. We also know we can eradicate
malaria from the blood, but malaria may reoccur years later due to parasites
hiding in the liver.
The Cochrane review on sofosbuvir and similar drugs consists of 757 pages but
perhaps the review abstract is all we need. 38 The authors included 138 trials
(25,232 patients) and found a sustained virologic response. They could neither
confirm nor reject any clinically relevant effects of the drugs because the trials
were mostly short-term. Moreover, they considered all trials and results to have a
high risk of bias, so their results presumably overestimated benefits and
underestimated harms. However, there was sufficient information in the trials to
rule out that the drugs compared with placebo reduced the relative risk of serious
adverse events by 20%.
www.bmj.com/content/350/bmj.g7809
Aha! Koretz is concerned, just like the Canadians and me. I have a weakness for
the BMJ, which I consider the best medical journal in the world. And I know
Koretz whom I have met many times. He is a careful researcher and a good
thinker.
The link is to PubMed but there is no link from there to the BMJ paper. If you
google 350:g7809 (the volume and first page number for the article) or just
g7809, you can go to the BMJ and see the article freely available there. 44
The authors describe that not only the CDC, but also the World Health
Organization and the US Preventive Services Task Force, recommend
widespread screening which, under the US Affordable Care Act, mandates
insurers to provide screening without any charges to individuals. That is bad
medicine and a colossal waste of money. New York State even passed legislation
making it mandatory for hospitals to offer testing to all patients born in 1945-65.
This is happening in a country where people die because they cannot pay for
their healthcare; where many still do not have a health insurance; and where
President Donald Trump and almost all his fellow Republicans want to roll back
Obamacare. What goes on there is truly astonishing.
As I had suspected, Koretz and colleagues reported that viral RNA is sometimes
found in body tissues even if the serum is clear, which is only one among many
reasons why sustained virologic response is a poor surrogate for real outcomes.
They estimate that at least 125 million people in the world have active infection.
To treat all these with one course of sofosbuvir (some patients need more than
one course) would cost $10 trillion, about half the US Gross Domestic Product.
That is what the American Academy for the Study of Liver Diseases wants: “all
patients with HCV should be treated to prevent complications of this curable
disease” (fifth link on the Google search list). 45 Who is going to pay for this?
And where is the evidence that antivirals are curative? There is none. This
section of my book is about conflicts of interest and, as you might have guessed,
these authors who criticized the Cochrane review and want to treat everyone
with even a trace of virus in their bloodstream have numerous financial conflicts
of interest when it comes to manufacturers of antiviral drugs.
And yet it is still widely accepted that people on drug industry payrolls can sit on
guidelines committees making recommendations about drugs, medical devices
or diagnostic tests sold by the same companies. The level of corruption in
healthcare is unbelievable. 2 These days, the conflicts people are involved in are
finally out in the open, but they still think everything is fine as long as they
declare their long list of drug industry benefactors. It is not. You do not become
any less corrupt in your academic thinking just because you have declared it.
Money buys opinions - often subconsciously. Follow the money and you can
explain the oddest things in healthcare.
Research has consistently shown that doctors on industry payrolls tend to hold
irrational views about drugs, often preferring expensive drugs not any better than
cheaper alternatives and tending to prefer drugs over other alternatives. 2
Therefore, before you read an article, you should find out if the authors have any
conflicts of interest. These usually appear at the end of the article just before the
references, yet some journals omit these disclosures and others refer readers to
the journal websites. On several occasions, I have not been able to find them.
That appears to be deliberate at times since the same journals can earn a great
deal by selling reprints of untrustworthy trial reports, or reviews of drugs and
devices. It is rarely worthwhile to read a paper if some of the authors have
financial conflicts of interest in relation to the drugs they describe.
And read the acknowledgements. They might reveal papers to be supported by
drug companies, or that a particular paper was not written by any of its authors.
When a person is thanked for “editorial assistance” or for “technical support,” it
usually means that this person wrote the paper. And when someone is merely
thanked for “help,” you can be assured it was for more than just brewing coffee
for the overburdened clinicians analyzing the data. Ghost writing is very harmful
to public health: it misleads doctors about the benefits and harms of drugs. 2 And
it is fraud because doctors are deliberately deceived into believing that the
papers were written by distinguished peers. Ghost-written papers are
subsequently cited in promotional materials and in other ghost papers as if they
provide independent verification of the claimed virtues of drugs. There may be
more ghost papers about a drug than papers written by independent academics. 2
Earlier, authors of at least one paper indexed by PubMed were entitled to upload
comments related to PubMed abstracts. This feature was called PubMed
Commons. In this way, we could at least partly circumvent editorial censorship
and warn readers looking at abstracts when research was not reliable.
When you had uploaded your criticism, the PubMed record would state right at
the top that a comment had been posted: “See comment in PubMed Commons
below.”
Not only financial conflicts of interest but even academic ones can be important.
Journal titles can provide hints. Journal of Medical Screening is not likely to
publish articles critical of screening and is likely to accept papers that tell the
audience what they want to hear - even if the data are flawed. That is exactly
what this journal does. It is really awful. 47
You might think that information from patient organizations can be trusted but
that is not the case. Many patient organizations are financially supported by the
industry and see no problem with this. 2 And sometimes drug companies even
establish patient organizations, doing so secretly. Therefore, you should always
find out where the money comes from before you start reading websites.
You are already receiving salmeterol, but your doctor convinces you to change to
the combination product because it lowers the risk of death - an issue that
worries you because you know your prognosis is worse than your friends’ who
never smoked.
You should always look up the evidence before going to the pharmacy. You will
be surprised. Googling salmeterol fluticasone cochrane leads you to several
Cochrane reviews. The first one compares one combination product with another
- not what you are after. The next has an interesting headline: “Do inhaled
steroids increase the risk of pneumonia in people with chronic obstructive
pulmonary disease (COPD)?” 49 The authors include 43 studies with 30,000
people and found that steroids (fluticasone and budesonide) increase serious
pneumonias requiring hospital admission. Over 18 months, 18 more people out
of each 1000 treated with fluticasone were admitted. The authors also said that
no more deaths were reported overall in the combination groups compared with
controls, and that deaths related to pneumonia were too rare to tell either way.
You now know that fluticasone and other steroids can cause pneumonia and you
already knew that pneumonias can be deadly. You also know without even
opening the review - in contrast to what your doctor told you - that steroids do
not reduce total mortality (they did not increase mortality in the review but, if
they had lowered mortality, it would surely have been stated).
The other Cochrane reviews on the first page of the Google search were about
asthma, not smoker’s lungs. But if we add COLD to our search strategy, we find
a large trial published in the New England Journal of Medicine. 50 You can read
the whole article for free but start with the abstract. The trial aimed at finding out
whether it could be beneficial to give corticosteroids to patients with smoker’s
lungs. GlaxoSmithKline (GSK) randomized 6,184 patients to fluticasone or
placebo, and again randomized all patients to salmeterol or placebo. That created
four groups: placebo, salmeterol, fluticasone, and both drugs together. The
design is called factorial, and its great advantage is that the effect of both drugs,
as well as that of the combination, can be studied in the same trial. Thus, the
investigators can study three research questions with a sample size that would
usually only allow one question to be answered. That is what makes this design
so attractive and powerful.
According to the abstract, “The hazard ratio for death in the combination therapy
group, as compared with the placebo group, was 0.825 (95% confidence interval
[CI], 0.681 to 1.002; P = 0.052, adjusted for the interim analyses).”
Thus, it would seem that your doctor was right when he said that you should take
the combination because it lowers the risk of death. This is the trial the
salesperson from GSK told your doctor about. It is called the TORCH trial,
which means “Towards a Revolution in COPD Health,” as if GSK already knew
the results before beginning the trial! The salesperson also gave him a reprint of
the article. When I teach doctors the principles of evidence-based medicine and
ask them how they will find out whether any given treatment would be good to
use, they often say they would do a search on PubMed. If they find a large trial
published in New England Journal of Medicine, they will believe in it just by
reading the abstract on PubMed.
Articles in this journal are cited vastly more often than articles in other journals,
and most doctors have huge respect for the journal. It does not deserve their
respect. 2
First, trials can be misleading. Trials with exciting results often get published in
our most prestigious journals but such trials might be aberrations. What if other
trials could not confirm reduced mortality? That is why we should always look
for systematic reviews of trials and, in this case, they do not support the finding
in New England Journal of Medicine.
Second, abstracts are often misleading, 7 even in this journal, and even on this
occasion. The editors allowed GSK to present a totally inappropriate analysis in
the abstract that only included half the patients, thereby spoiling the advantage
of the factorial design. The misleading results in the abstract gives the readers
the impression that both of GSK’s drugs should be used.
This is not a lone ‘error’ that can be excused. It appears to be quite deliberate.
Both GSK and AstraZeneca have also not performed the appropriate analyses in
other similar trials. 51
Instead of the above approach, you might have looked for Cochrane reviews via
cochranelibrary.com, Browse by Topic, and Lungs & Airways. There are 17
reviews in the category Chronic Obstructive Pulmonary Disease (exacerbations)
and 76 under Chronic Obstructive Pulmonary Disease (stable).
It takes less than five minutes to browse 76 titles and to find the relevant
Cochrane review: Combined corticosteroid and long-acting beta 2-agonist in one
inhaler versus long-acting beta 2-agonists for chronic obstructive pulmonary
disease. 53 The only thing you need to know is that salmeterol is a long-acting
beta 2-agonist, which is explained if you google salmeterol. More generally,
when you decide to browse the titles of Cochrane reviews, you need to
familiarize yourself with the type of drugs involved first. Cochrane reviews are
usually about a class of drugs - not a single drug - which is why you need to
know what the class of drugs is called.
The review is large. It included 14 trials (11,794 people) with severe COPD. The
combination was not better than the single drug for preventing death. It was also
doubtful whether it carried any advantage in relation to exacerbations and
hospital admissions, and there was an increased risk of pneumonia. Quality of
life, symptoms score, use of rescue medication and FEV1 (the volume of air that
can be forced out in one second after taking a deep breath) improved more on
the combination drug than on the single drug, but the authors reported that the
average differences probably were not clinically significant.
Now you can be absolutely certain that the advice you received from your doctor
was poor. Furthermore, browsing all the titles gives you an additional bonus:
You can see that there are many other Cochrane reviews that might interest you.
If you set aside a little time, chances are you will soon know more about these
drugs than your doctor does.
There is a final twist to this story. When browsing the 76 Cochrane reviews, you
might have easily picked the wrong review. It has the same first author as the
relevant review and a similar title: Combined corticosteroid and long-acting beta
2-agonist in one inhaler versus placebo for chronic obstructive pulmonary
disease, the only difference being that placebo was the comparator instead of
long-acting beta 2-agonists. 54 This illustrates why you need to be very careful
that you look for reviews with the correct comparator.
I also sent my criticism to the Cochrane Library where it was published as part
of the review together with the authors’ reply. 54 As the Cochrane review
continued to be misleading, I submitted a second criticism and complained to the
Cochrane Editor-in-Chief requesting corrections of the misleading statements.
Unfortunately, even after my second attempt that did not happen.
The main problem was that the review authors implied that the combination drug
lowered mortality even though the steroid had no role in this. The authors
replicated the misleading P-value of 0.052 for mortality and recommended that
the combination should be compared with its two components in future trials,
ignoring that this had already been done, not only in the TORCH trial, but in
other similar trials as well. 51,52
3 Information sources
This chapter is a little boring, but you do need to read it. If you skip it now, then
please return to it later.
Google google.com
Google Translate translate.google.com
Wikipedia in English en.wikipedia.org
Cochrane Library cochranelibrary.com
Cochrane Library in Spanish bibliotecacochrane.com
PubMed ncbi.nlm.nih.gov/pubmed
McMaster Textbook of Internal Medicine empendium.com/mcmtextbook
If your mother tongue is not English, Google Translate will often lead you to the
correct term in English. For example, resfriado comun in Spanish, rhume in
French and Erkältung in German are all the common cold.
Drug names
Drugs often have different trade names in different countries, e.g. Prozac and
Fontex both contain fluoxetine - a pill used for depression and many other
ailments. Eli Lilly, an American company, is the inventor of fluoxetine and, in
the US, this substance is also sold under the trade name Sarafem, which has been
approved for treatment of so-called Premenstrual Dysphoric Disorder by the
Food and Drug Administration. 1 In Europe, Lilly was forbidden to promote
fluoxetine for a condition that is not a disease.
Drug regulators approve drugs far too easily, 1 which is a major reason why
medicines are the third leading cause of death. So, when a major drug regulator
somewhere in the world disapproves of a drug, the message is strong: Do not use
it!
Generic drug names are often long and difficult to remember but you need to get
used to that. Imagine an American woman who was prescribed Prozac for
anxiety by her family doctor and Sarafem for menstrual problems by her
gynaecologist. If she does not pay attention to the generic name, she is unlikely
to find out that it is the same drug and might overdose, which increases the risks
of suicide, violence and homicide. 2
You should even use the generic drug names when making searches.
Interventions : People often have very broad questions, e.g. which treatments are
best for low back pain? Such questions need to be narrowed and reformulated,
for instance: Is exercise better than no exercise?
Outcomes : Clinicians and patients must always bear in mind what they want to
achieve. Unfortunately, this most important point is often ignored. Far too often,
people continue with a treatment that does not help them.
Like all psychiatric drugs, ADHD drugs are rather poor drugs. In the short term
they can have noticeable effects on things such as concentrating on boring tasks,
but Timimi explains that, in the long run, these drugs seem to cause more harm
than good and any apparent initially beneficial effects wear off. 2 By discovering
the specifics of what the parents want to see change, Timimi can divert their
interest from drugs to more targeted measures such as developing parental
management skills for children who are more ‘intense’ than most. He helps them
understand the anxieties and stress their children may be feeling, or supports
them getting more structured interventions in schools. He also reminds parents
that one thing is certain about children: they change as they grow older and often
the problems labelled as ADHD (particularly the hyperactivity and impulsivity)
tend to diminish and go away as the child matures during adolescence. 2
Google: google.com
Google is fabulous. Yet some Google automated features makes it difficult to
understand how it does what it does. Google optimizes searches, so results are
better suited individual users even if they had never previously made a search.
There is a commercial reason for this: it enables Google to direct users to
websites with advertisements targeting their individual interests.
Therefore, if you attempt to reproduce my examples, you might not get exactly
the same results. If I write that a particular article is the fourth entry on the first
page in a Google search, it could be the sixth entry in your search. Even the day
we search can affect the outcome.
• Your location
• Device information
That is scary. We are told that Google does not sell our personal information to
anyone. But can we trust this? The Americans spied on German Chancellor
Angela Merkel’s I-phone!
If you want to restrict your searches to recent times, Tools can easily do so.
A general issue with searches is that if they contain more than one word, the
results depend on whether quotation marks are used. "Back pain" means exactly
that and therefore yields fewer entries than if searching for back pain, which
means back AND pain. But you will rarely need to use quotation marks in a
Google search.
Wikipedia: wikipedia.org
Wikipedia articles are written by volunteers. Editorial control is in the hands of
users who create and manage the content. Anyone is welcome to join as a
contributor, editor or author with the understanding that they are legally
responsible for their contributions under US law.
If you look up back pain, for example, the Edit button at the top will tell you:
“You are not logged in. Your IP address will be publicly visible if you make any
edits. If you log in or create an account, your edits will be attributed to a user
name, among other benefits.”
The articles can be edited by anyone and, if you click on View history, you can
see who the authors are, what changes they have made and when they made the
changes. In this case, all the many authors used pseudonyms such as “Doc
James.” Sometimes you can click on these and find out who they are and even
email them. I picked one random name on a long list who turned out to be
someone I knew: James Heilman describes himself as “Canadian ER doc.” He
also says he is a volunteer and does not accept money or honorariums for any
efforts related to Wikipedia.
Young people have no idea what we had to go through in the past. We had to
look up information in encyclopaedias and if we did not have one at home we
usually gave up.
A pertinent question is: Since Wikipedia builds on the work of volunteers who
may have their own horses in the race, or prejudices they wish to propagate to
the rest of the world, can we then trust the articles? Encyclopaedia Britannica,
first published in 1768, is widely respected as a reliable source of information. A
study in the periodical, Nature, compared these two sources. 3Nature chose
articles from both sites on a wide range of topics and sent them to field experts
for peer review. The experts compared these pairs of articles but were not told
which article came from what site. Nature got 42 usable reviews back from its
experts: eight had serious errors such as general misunderstandings of vital
concepts - four from each site. Wikipedia had 162 factual errors, omissions or
misleading statements and Britannica had 123. That makes four mistakes per
article for Wikipedia and three for Britannica. Not bad for a volunteer effort
considering the longer the articles are, the more errors there will be (and
Wikipedia articles tend to be very long). Wikipedia in English has over 60 times
more words than Encyclopaedia Britannica. 4
Cochrane Library: cochranelibrary.org
I have a conflict of interest in my praise for the Cochrane Library: I co-founded
the Cochrane Collaboration in 1993 and I was the director of the Nordic
Cochrane Centre. However, I try to be neutral and objective, and when there is
reason to criticize Cochrane reviews, I do so. Examples of that are found in this
book and in my other books. 1,2
In the future, there will also be reviews of prognosis, which can tell patients
what is likely to happen after they have been diagnosed. One of the first
questions patients want answered is whether their condition is serious. They also
often want to know if it is ‘self-limiting,’ i.e. will go away without treatment,
and how long that will take, on average.
The Cochrane Collaboration is a registered charity and almost all reviews are
undertaken by volunteers without any funding. It is similar to when people
volunteer to write articles for Wikipedia without pay - just much more time
consuming.
Many governments have bought national subscriptions for the Cochrane Library.
Currently, half the world’s population has free access to Cochrane reviews, and
the whole world has free access to all Cochrane reviews one year after
publication.
There is free access to the abstracts, indexed on PubMed, with links to the full
reviews. Even free access to summaries prepared for patients and other non-
professionals is provided. These are often called consumer summaries in English
(and Cochrane jargon), but that is an odd term. Patients do not consume
anything; in fact, diseases and drugs often consume them, and consumption was
the old term for tuberculosis, which ‘eats’ the tissues. When patients with breast
cancer, prostate disease, fractures and HIV were asked, they preferred to be
called patients, not consumers, clients, customers, or anything else. 7 Many
alternatives to ‘patient’ incorporate assumptions (e.g. a market relationship),
which patients may find objectionable. ‘Consumer’ was introduced with good
intentions about empowering patients but this can be achieved without calling
them something they do not want, and which is also quite misleading. The
abstracts of Cochrane reviews are usually longer than patient summaries and
follow strict scientific standards.
Patient summaries and Cochrane abstracts are translated from English into many
languages. Since this work is done by volunteers and is dependent on their
available time, the most important Cochrane reviews have the most translations.
If you look up the Cochrane review Screening for breast cancer with
mammography by googling the title, at the top on the first page, you will see that
the patient summary has been translated into Croatian, Dutch, French, German,
Malaysian, Portuguese, Russian, Spanish, and Tamil.
From the Cochrane Library main page, cochranelibrary.org, there are two ways
of finding reviews. You can do a search at the top right. The Advanced Search
option provides you with many options for focusing your searches and even
saving them. The second way is to click on Cochrane Reviews on the left and
browse titles of reviews by topic (the most useful option – try Cancer ), or by
review group (more for ‘insiders’).
Titles of Cochrane reviews are required to state the interventions reviewed and
the problem at which they are directed, e.g. “Vitamin C for preventing and
treating the common cold.”
All Cochrane reviews have a unique identifier, which is part of the reference. For
example, if you come across this reference in an article or on a website:
“Bjelakovic G, Nikolova D, Gluud LL, et al. Antioxidant supplements for
prevention of mortality in healthy participants and patients with various
diseases. Cochrane Database Syst Rev 2008;2:CD007176,” and you would like
to read the review, you can use the Advanced Search option, move the arrow key
to Search All Text, write “CD007176,” and click Go. It also works well on
Google where CD007176 links to PubMed (see the next section), and from there
to the Cochrane review. Or go directly to the Cochrane review by clicking the
third link on the Google page.
PubMed ncbi.nlm.nih.gov/pubmed
PubMed is a consumer friendly portal to Medline, a huge database created by the
US National Library of Medicine. PubMed is comprised of more than 27 million
citations for biomedical literature from Medline, plus life science journals and
online books. The PubMed journal list includes approximately 30,000 records.
The citations are for abstracts and may include links to full-text content from
PubMed Central and publisher web sites.
PubMed has many other useful resources and one of America’s greatest gifts to
the world was making it free to search this database in 1997. Before then -
before PubMed - it was very difficult to search in Medline and very expensive.
When I opened the Nordic Cochrane Centre in 1993, I bought a subscription to
Medline via a Swiss provider costing approximately $2000 annually. My
investment was a total waste since I could never figure out how to use it and
always needed extensive help from the librarians at our university library.
Databases other than PubMed exist but you do not need to know about them.
The purpose of this book it is not teaching you how to be a researcher but
directing you to information sources that are free of charge.
If you read an article quoting another article you wish to find out more about,
you can usually do so quickly by using PubMed. Google PubMed or go to
ncbi.nlm.nih.gov/pubmed. Under PubMed Tools on the opening page, go to
Single Citation Matcher, write the publication year in the Date field, the first
page of the article in the First page field and add one more piece of information
such as the title of the journal or the name of the first author. When you start
typing the journal title, various title suggestions will appear making it easier to
avoid mistyping.
For instance, if you want to look up the abstract for: “Allen C, Glasziou P, Del
Mar C. Bed rest: a potentially harmful treatment needing more careful
evaluation. Lancet 1999;354:1229–33,” you can type “1999,” “1229” and
“Allen” in the respective fields.
Then you will find a systematic review which is quite interesting. The authors
examined 39 trials of bed rest for 15 different conditions (5,777 patients). In 24
trials investigating bed rest following medical procedures, no outcomes
improved significantly and eight worsened significantly after some procedures
(lumbar puncture, spinal anaesthesia, radiculography, and cardiac
catheterization). In 15 trials investigating bed rest as the primary treatment, no
outcomes improved significantly and nine worsened significantly for some
conditions (acute low back pain, labour, hypertension during pregnancy,
myocardial infarction, and acute infectious hepatitis). It seems to me - whatever
our problems - we should stay in bed as little as possible! Not that long ago,
people with heart attacks were sent to bed in order to protect the heart muscle.
Many of these people were killed due to pneumonia and blood clots in their legs
ending up in their lungs.
PubMed searches can be very elaborate, employing brackets and separators like
AND, OR and NOT (capitalized). You will most likely not need to know this, but
here is an example of it from our Cochrane review: “House dust mite control
measures for asthma” (discussed in Chapter 2). 9 Our PubMed search was:
(mite* AND asthma*) AND (random* OR control* OR blind*). The asterisk (*)
indicates that all subsequent character combinations should be included, e.g.
both “asthma” and “asthmatic.”
That very broad search yielded 2,191 records. If you are not doing research, you
might wish to narrow your search to systematic reviews. Go to Article types on
the left and open Customize. Go through the list and check the Systematic
Reviews box. Be aware that PubMed saves your choices, so you will need to
undo them the next time you search for other article types. Next, click Show.
Nothing happens. The number of entries remains the same. That is due to is poor
programming. To make it work properly, you need to click Systematic Reviews
on the left. You will now see a checkmark (✓) and only 74 entries.
As noted earlier, PubMed abstracts often have links to the full articles, and
sometimes these articles can be read for free. If not, you can likely find them for
free at university libraries. If you try to open an article that is not open access,
you will be asked to pay a fee either to the library that subscribes to the journal
or to the publisher. In the PubMed abstract, there are even links to Similar
articles, which can be very interesting at times.
That was just to give you an idea of a few of the many functionalities in
PubMed. If you look for systematic reviews, Google or the Cochrane Library are
easier.
Medical textbooks
Medical textbooks are useful for acquiring general information about diseases
and about which diagnostic tests and treatments are being recommended by
clinicians. Sometimes these books are sufficient. Textbooks can rarely be
accessed online for free - the McMaster Textbook of Internal Medicine is an
exception ( empendium.com/mcmtextbook ). Since it is still under development,
you might not yet find what you are looking for, but it is worth a try. This
textbook receives donations from people not affiliated with the pharmaceutical
industry, and utilizes the GRADE system allowing explicit differentiation
between strong and weak recommendations.
4 Is the test necessary and the
diagnosis correct?
“The doctor said I had …” Perhaps the doctor did, but the diagnosis could be
wrong. Patients rarely question diagnoses, and even though doctors have some
awareness of ‘diagnostic uncertainty,’ they are much more confident about their
diagnoses than they should be. Scientific studies have consistently shown that
diagnoses are far more unreliable than doctors realize - particularly in psychiatry.
Many doctors have had similar experiences, particularly family doctors who
often encounter patients with diffuse symptoms that do not suggest any
particular diagnosis. A well-functioning system of family doctors is a great asset
because they know their patients and are in much better positions than specialist
doctors when it comes to understanding their diffuse symptoms. These patients
should not be subjected to diagnostic testing because their complaints are
ordinary ones about the multitude of troubles in life. They do not indicate
diseases.
Overtesting is very harmful, as it leads to overtreatment. When doctors do not
know their patients, they tend to order too many diagnostic tests. I have
described the consequences of this elsewhere, 1 and they can readily be seen in
the United States where family doctors are very uncommon, and keeping people
healthy is not a priority because the system is profit-driven. The Unites States
has the most ineffective healthcare system in the developed world and its
citizens have a relatively low healthy life expectancy despite the fact that the US
spends twice as much on healthcare as other industrialized countries (measured
as the percentage of the gross national product).
A 2008 report from the Commonwealth Fund found that the Unites States ranked
last among 19 industrialised countries across a range of healthcare measures . 2
The report tied much of the problem to a weak base of primary care doctors. A
study of 3,075 US counties found that every 20% increase in the number of
primary care physicians was associated with a 6% reduction in total mortality. 3
Since the use of diagnostic testing often leads to harm, you should feel free to
ask your doctor why he thinks a particular diagnostic test is necessary, especially
if you have doubts about it. Does carrying out the test make any difference for
your prognosis? Would you have received the same treatment without the test (in
that case, the test would rarely be necessary)? Is it advisable to wait and see -
without testing - since your problem is not serious and might disappear without
treatment?
I began by saying the biopsy could be painful but that it would not be a major
problem because we would give him a morphine-like drug. He immediately
displayed anxiety. Next, I said there might be bleeding in the abdomen - 1 in 200
patients require blood transfusions - and therefore, I had ordered a few portions
of blood, just in case. Now his anxiety was intense. Finally, I told him 1 in 5,000
patients died from a liver biopsy. Dead silence. The patient did not ask a single
question. It is many years ago, but I think I used these numbers.
I scheduled an appointment for the biopsy later the same day, but I never saw the
patient again. He made a quiet escape and I felt I had done the best I could.
When AIDS was a new disease and patients were admitted in poor condition,
doctors often did not know what to look for. A huge amount of diagnostic testing
was carried out and I decided to find out what the therapeutic consequences had
been by sifting through the voluminous case records of the first 33 deceased
patients in our department. 6 One of the things we discovered was that the many
liver biopsies that had been taken had been totally useless. Many of the doctors
came from liver departments and were used to taking liver biopsies.
Doctors are paid according to the services they provide, so the more tests they
order, the more they will earn. Sometimes they also benefit from tests carried out
elsewhere, e.g. because they own facilities that perform CT scans or ultrasounds.
It is difficult to ask questions that might upset your doctor, but you should put
yourself first. After all, the doctor is there to help you. If your doctor orders a
test you suspect is expensive, you could ask about the cost, whether a cheaper
test is available - and what the doctor earns by ordering the test. You could also
ask whether he receives money or other benefits from the drug industry, has
shares in a company or is visited by drug sales representatives. If your doctor
gets embarrassed, angry or defensive when you ask relevant questions, find
another doctor.
There are many other things you can do. 1 You could start by finding guidelines
in a Google search. I will get back to this just below. Avoid taking drugs unless
they are absolutely necessary – which they rarely are. Remember that you are the
one to decide whether a drug is needed, not your doctor who will not suffer the
harms, and who might not even know what those harms are.
Ask if there are options other than drugs and whether you would be better off -
or reasonably well off - without treatment. Bear in mind that very, very few
patients benefit from the drugs they take. 1 If a drug slightly helps 10% of the
patients, which is very common, that means that nine out of ten are not helped. If
a drug only helps 1%, which is also very common because we are taking so
many drugs for slightly elevated values of blood pressure, blood glucose and
cholesterol, for instance, that means that 99 out of 100 are not being helped.
Ask if there are cheaper drugs than the one your doctor suggests. Kickbacks are
common 1 and, even in countries with little corruption, doctors can still be paid
illegally for each patient they prescribe an expensive drug.
I realize that few healthy people and patients will feel qualified to debate with
their doctors about tests, but anyone can ask simple and reasonable questions. If
you are dismissed with replies like, “That’s what we always do in your
situation,” ask for the evidence even though a meaningful reply is unlikely.
Doctors know very little about diagnostic tests, and they would usually have no
clue about how often tests provide correct or incorrect results for patients like
you. I am not saying that just to criticize my colleagues. That also applies to me.
It is impossible to have detailed knowledge about everything we do, particularly
when the issue is diagnostic tests.
We have to make decisions about where to divide this grey zone into those we
declare ill and those we declare healthy. Yet, no matter where we make the split,
some of those declared ill are actually healthy, and vice versa. And because we
do not want to overlook cancers, we tend to use the cancer label even for
intermediate cell changes even though some are not cancerous, and others will
regress spontaneously.
If a patient has heartburn and a gastroscopy shows an ulcer in the duodenum (the
uppermost part of the small intestine), we accept this result and treat the patient.
But how often is the diagnosis wrong and how often is an ulcer overlooked?
That is difficult to know since there is no gold standard, but we can get an idea
about it by letting two doctors do gastroscopies on the same patients
independently of each other.
In one such study, the doctors agreed there were no ulcers in most of the
examined patients. 8 They also agreed that 10 patients had an ulcer. But in 14
cases, only one of them found ulcers. Thus, by having the patients examined by
two doctors instead of one, the number of positive cases more than doubled. We
cannot know the correct number of patients with ulcers. Ulcers are not as
objective as people think. Scars resembling active ulcers might remain from
previous ulcers, for instance.
For laboratory values, we often define a so-called normal range for results as a
range that includes 95% of the values when a large group of healthy people are
tested. Therefore, 2.5% of healthy people will have values below the normal
range and 2.5% values above. It is actually a misnomer to call it a normal range,
because people with values just outside this range are normal. Therefore, it
would be better to call this range a reference interval.
Therefore, going to regular health checks “to be on the safe side” has little to do
with being on the safe side. In fact, it is more likely to harm you. I will explain
how in Chapter 7.
Questionnaires are the worst of all when it comes to making healthy people sick.
Widely used in psychiatry, the questions are so broad and vague that if each and
every one of us were tested with enough diagnostic questionnaires, virtually all
of us could get at least one psychiatric diagnosis. 9
If we want to figure out whether a particular diagnostic test does more good than
harm, the best thing to do is to carry out a randomized trial where half of the
patients get the test and the other half does not. Since we treat patients based on
the knowledge we have about them, the two groups are not likely to get exactly
the same treatments - precisely the point of this trial. Unfortunately, few such
trials take place.
Diagnostic tests are usually studied in other ways than in randomized trials,
leading to a multitude of problems that I feel fall outside the main scope of this
book. The subject is complicated, and you will probably not need this kind of
knowledge.
On the other hand, you might benefit from a brief introduction to this subject.
One of the methods is employing likelihood ratios: You suspect a certain disease
because of the symptoms and signs a patient displays. This suspicion
corresponds to some subjective probability for having the disease. A diagnostic
test is then performed for that disease. If positive, the likelihood of having the
disease increases, and if negative, it decreases. We are still only talking about
probabilities - we cannot say for certain whether you do or do not have the
disease.
In Chapter 7, I will tell you more about the importance of the a priori probability
of having a disease in relation to screening, which is an issue that confuses most
people. It is also an issue where most doctors get the basic facts wrong.
Americans are proud of their freedom, but they also have the freedom to exploit
the suffering of their neighbours as much as they wish. Greed to the extreme
reigns. Half the people in the world earn less than $4,000 a year.
It should be noted that many doctors do these procedures because they believe
them to be helpful, so the crucial question is whether the test was necessary? All
doctors know chest pain may develop during exercise; we call it angina pectoris,
or just angina, which means chest pain. Angina is caused by a reduced blood
flow to the heart muscle, which is almost always secondary to arteriosclerosis of
the coronary arteries. Although the lawyer’s tests were normal, angina was
probably the issue and drugs can alleviate the symptoms. Most people have
heard of nitroglycerin, an ingredient in dynamite also used to treat angina. Today
guidelines recommend a beta blocker or a calcium channel blocker as first-line
treatment. You will find this information if you google angina treatment and
look up the NICE guidelines (National Institute for Health and Care Excellence
in the UK).
Another test for people with chest pain is coronary angiography. A catheter is
inserted into the artery in the groin and moved toward the heart, and an X-ray of
the coronary arteries is obtained by injecting contrast. I once discussed clinical
cases with colleagues in the United States and they told me of a patient who got
chest pain when running uphill, which led to a coronary angiography. I
questioned this and asked why they did not intervene without the test. They
replied that if the coronaries were too narrow, one or more stents could be
inserted to widen them.
I would argue that since we already knew the patient had angina - defined as
chest pain during exercise - then why do a stress test? The patient already did his
own stress test by running so why repeat this on a treadmill or stationary bicycle
at the clinic? And if the test was negative, it would be a false negative, because
we already knew the patient had angina.
Next, I tried exercise stress test cochrane. There was a link to a Cochrane
review, but the abstract tells us that the review is about screening patients with
chronic kidney disease for the risk of coronary artery disease. Not what we are
after.
Another link looked more relevant - a systematic review and meta-analysis of
prospective studies of the accuracy of exercise stress test for coronary artery
disease [CAD] 12 - perfect. The authors included studies investigating “a
representative sample of patients over the age of 18 years, who had symptoms
suggestive of CAD (for example, chest pain or breathlessness), or who were
asymptomatic but had risk factors for CAD (e.g. diabetes mellitus,
hypertension).”
The Google link brought me to the methods section of the article, which is not
the original, but a copy published by Medscape. If you click Abstract and
Introduction, you will be asked to register for free at Medscape in order to read
the article.
That is easy but there is a quicker way. Go back one step to the methods section
where the full reference of the article is found at the top. Open a new window in
your browser and go to PubMed, ncbi.nlm.nih.gov/pubmed, PubMed Tools and
Single Citation Matcher. Enter a few details from the reference of the article,
e.g. Int J Clin Pract in the Journal field, 2012 in the Date field, and 477 in the
First page field; click on Search, and the abstract appears.
The authors included 34 studies with 3,352 participants, which is a very large
sample for a study of diagnostic test accuracy. They reported that bicycle echo
(ultrasound) testing performed better than treadmill echo testing, which
outperformed both treadmill ECG and bicycle ECG. They documented this by
using likelihood ratios. A positive likelihood ratio of 11 for bicycle echo testing
means that people with CAD are 11 times more likely to have a positive test than
people without CAD. Conversely, a negative likelihood ratio of 0.2 means that
people with CAD are 5 times less likely to have a negative test than people
without CAD.
The link to the full article is to the right of the PubMed abstract and access is
free. What you always need to ask is: Were the patients who participated in the
studies similar to the lawyer (or yourself)? You can already see in the methods
section on Medscape that they were not, since some of them did not have angina
symptoms, only risk factors for coronary artery disease.
There are other reasons why this good review may not give you the answers you
are looking for. In their abstract, the authors conclude that exercise testing is
more useful at excluding CAD than confirming it, which suggests that some
patients using this test may be spared angiographies, which are more expensive.
However, in the main body of the article, the authors say exactly the opposite,
namely that exercise testing is probably more useful at confirming CAD than
excluding angiographies. So, what is correct?
Despite these issues, I think we may come to useful conclusions by reading the
paper. The authors say that only 8 of the 34 studies were blinded to the reference
standard and the test. Blinding the investigators is the most important precaution
we have, and when they are not blinded, there is a risk that they will report
socially desirable answers rather than what they would if blinded. Thus, the
studies probably exaggerate the accuracy of the test. Even so, the seemingly
impressive likelihood ratios are not so impressive when we look at more readily
understandable numbers. The median prevalence of CAD in men in the studies
was 62%. So, before doing the test, a man's risk of having CAD was 62%. A
positive exercise test increased his risk of having CAD to 82% and a negative
test decreased the risk to 37%. Moreover, our patient does not belong to a group
where 62% has CAD; he belongs to a group where virtually everyone has CAD,
and where a positive test result would, therefore, not add much to what we
already knew.
The crucial question is: Are there any trials that have compared outcomes for
patients with angina treated without stress tests or angiographies, with patients
who underwent tests before treatment? Such trials would give us the answer we
are looking for.
You could search on PubMed using “ exercise stress test” in the search field. It is
important to use quotation marks. When I did this search without using quotation
marks, 89,087 entries appeared and only 1,251 came up with quotation marks.
Then I went to Article types on the left, Customize, checked the boxes
Randomized, Controlled Trial and Systematic Reviews, undid any other
checkmarks there were on the list, clicked Show, and clicked on Randomized
Controlled Trial and on Systematic Reviews so that the checkmarks (✓) were
visible.
105 titles came up which took me six minutes to browse. Most of them were
dealing with drug therapy - not a single one was relevant.
Another category also came up under Heart & circulation called Non-specific
chest pain, which is not relevant since the lawyer had specific chest pain.
I shall now tell you my own story, which illustrates the diagnostic difficulties in
this area and how difficult the practice of medicine can be.
In 2010 at the age of 60, I ran fairly fast in a 5 km relay. After 4 km, I got a
strange and unpleasant feeling in my chest that prevented me from running.
Since I had never had heart problems or any risk factors for heart disease, I was
surprised that this could happen. After a short break, I resumed running but was
forced to stop again because of cardiac arrhythmias and the same strange feeling.
I was left with no doubt whatsoever that I had coronary heart disease, despite all
my years enjoying various sports and my lack of risk factors. I was encouraged
to take a small dose of aspirin every day but, after looking up the evidence, I
refused taking it.
How can you find out whether you should accept or refuse aspirin? Please think
about that before you continue reading, then do the search and see what you find.
I will not give you the ‘solution’ until later.
Because my symptoms, the cardiac monitoring results, and the stress test all
complimented each other very nicely, no one had any doubt I had coronary heart
disease - one of the most common causes of death.
It took some time to get used to my new situation. I looked up various reviews
and found out by how many times my risk of death had suddenly increased. You
might see this a little masochistic, but I have a habit of wanting to know as much
as possible about everything. I felt I stepped out of my usual, “Don’t worry be
happy” attitude to life and walked straight into the antechamber of death. It felt
like that until I got used to my new situation.
I was put in a hospital bed next to a young man with blocked coronaries due to a
hereditary condition. His father had died very young. I felt very sorry for the
man and I even felt a little sorry for myself. So, these people are my new
companions, right? One of the many bad things about being hospitalized is you
are no longer in healthy company. Now it is all about sick people. We should
spend as little time as possible at hospitals because staying there does not bolster
self-confidence.
The next day I was supposed to have one or more stents inserted into my
arteriosclerotic arteries. I am still surprised I said yes. It happened so quickly, I
had little time to think. I already knew - contrary to what people think – that
coronary bypass operations do not prolong life; they merely have a symptomatic
effect. Therefore, I supposed the same was true for stents and I did not like the
idea of having tubes inserted into my coronaries. Furthermore, my ailments were
minor. I had no problem playing tennis. I could give up running and, instead,
race around on my bike.
Why did I accept the stenting? I just do not know. I should have studied the
evidence first.
There I was, lying on the table with a catheter in my groin, awaiting the
inevitable. I would be leaving my wonderful world of freedom and entering the
land of sickness and dependency, destined to die too soon. Life would never be
the same.
Then the cardiologist said, “Turn the screen so that Peter can see his arteries.” I
was stunned. My coronaries were smooth - without a trace of arteriosclerosis.
They might have belonged to one of my medical students.
What on earth was this? My cardiologist said I was false positive and
recommended that I continued running as much as my symptoms allowed. I
agreed. I asked the cardiologist who had subjected me to the stress test to explain
it. He could not. We all knew that, under certain rare circumstances, people
could have spasms in their arteries. I assumed that was the case for me. Since
there were no arrhythmias during the positive stress test, that could not be the
reason. I might never find an answer.
I still run - more than ever - sometimes 8 km every day of the week together
with my wife who has run half-marathons. I am too lazy for that. When we run
too fast and I get uncomfortable chest arrhythmias, I just pause for a few
seconds. Life is now far better than in 2010, demonstrating that it does not
always go downhill - it does improve at times.
Good doctors become increasingly humble with age because they realize that
many of their patients do not correspond to what they have read in their
textbooks. Conversely, poor doctors become increasingly arrogant as time
passes. I was lucky enough to be looked after by good doctors.
Did you find out how to search for the effects of aspirin? Found anything
helpful? It is not an easy task. Since aspirin is used for many things, we need to
focus on heart disease. Is the issue about using it for therapy or prevention? I
would say prevention because aspirin lowers the risk of blood clots, which is
why it is recommended for people who have had heart attacks.
You could try googling aspirin heart cochrane. If you type aspirin coronary
cochrane, the first two entries will be the same.
The first link goes to a Cochrane review of aspirin for primary prevention of
coronary artery disease 13 - not directly relevant for my case because I believed I
already had the disease. Yet what people usually want to know is how to prevent
heart attacks, so perhaps this review is of interest after all. It turns out to be only
a protocol for a Cochrane review. Since it is from 2004, the review should have
been completed and published long ago. (Therefore, I have written to the
Cochrane review group listed just under the title of the review, Editorial Group:
Cochrane Heart Group, and asked them to remove this outdated protocol from
the Cochrane Library.)
The next Google entry was a review that compared aspirin to aspirin in
combination with another drug for preventing cardiovascular disease. 14 This is
not relevant because we want to know the effects of aspirin versus placebo. On
the other hand, it takes only a few minutes to read the background section. It
says that a meta-analysis has shown that the relative risk reduction of death,
myocardial infarction and stroke in patients at risk of cardiovascular events is
approximately 20%, and that protection with antiplatelet therapy - e.g. aspirin -
for patients at high risk of cardiovascular disease remains unsatisfactory in
absolute terms.
Although I was at risk of cardiovascular events, I felt that the risk was not great.
If I had been taking aspirin and I fell off my bike or stumbled on a root and hit
my head while running in the forest, it would not be pleasant because aspirin
might cause a brain haemorrhage.
Other people might have reached other conclusions than I, but this example
illustrates the importance of not treating everyone the same way - which
guidelines tend to do.
One of the most important of all questions is this: “Am I similar to the patients in
this review?” In the entire world, is there not a single, large trial comparing
aspirin to placebo in patients like me? Not a trial comparing patients who never
had angina (primary prevention) and not patients who had previous heart attacks
or other serious events - something in between.
If you search on aspirin angina in PubMed, it corrects into aspirin AND angina
(2,273 entries). Limit your search to Clinical Trials or Systematic Reviews : 631
entries. Still a bit too many but they could be browsed through in a day. For
Systematic Reviews alone (check the Clinical Trials box to make it disappear) –
you get 122 entries. At the top, sort by clicking Most Recent. The first entries are
about all kinds of irrelevant issues such as patients with diabetes. We can narrow
the search to include only entries with angina in the title: aspirin AND
angina[ti]. PubMed has several such useful features. Now you are down to 24
links, most about unstable angina. If you wish to eliminate these, you can search
for aspirin AND angina[ti] NOT “unstable angina”, and you are left with three
documents - two narrative reviews and one guideline - none of which are
relevant.
Did I have stable or unstable angina? Googling stable angina provides the
answer. The first link went to the American Heart Association website. Stable
angina was what I had. Pain or discomfort occurred when my heart worked
harder - it did not come by surprise, episodes of pain tended to be similar,
usually brief (5 minutes or less) and relief was attained through rest or
medication. You might experience it while running and normally this type of
chest discomfort is relieved through rest, nitroglycerin or both.
One question remains. Do stents improve survival? I did a basic search of the
Cochrane Library for stent. Stents are used in many places in the body, e.g. even
in the biliary duct, but the 58 titles that came up were quickly browsed. One
study compared two types of stents. I read the background section: 15 adverse
events associated with percutaneous coronary intervention include death,
coronary artery complications, e.g. perforation of the artery, distal embolization
(passage of an intravascular mass capable of clogging capillaries), or stent
thrombosis, myocardial infarction, bleeding or infection of the access site,
bleeding in the abdomen, stroke, and acute kidney failure. Holy smoke! I had
exposed myself to all those risks for no good reason.
Do stents improve survival in those who need them? That is difficult to find out,
because the many trials that have been performed compare stents with something
else, like bypass or balloon angioplasty, not with doing nothing. How do you
search on doing nothing? That is really difficult. I decided to cut this Gordian
Knot which, to me, always means Google: do stents improve mortality. It
worked. A paper in New York Times described what we need to know with
reference to a recent meta-analysis. 16
Three trials enlisted stable patients after myocardial infarction, and five trials
enlisted patients with stable angina or ischaemia on stress testing. 17 Interesting. I
finally found patients like me.
“Prescribing beta blockers, ACE inhibitors, statins and daily aspirin — now
standard for treatment of stable coronary artery disease — was just as effective
as stent implantation for prevention of chest pain, heart attack, the need for a
future PCI and death.” 16
One of the authors of the meta-analysis said that more than half of the patients
with stable coronary artery disease are implanted with stents without even trying
drug treatment. They believed the reason was financial. 17 “In many hospitals,
the cardiac service line generates 40% percent of the total hospital revenue, so
there’s incredible pressure to do more procedures … When you put in a stent,
everyone is happy — the hospital is making more money, the doctor is making
more money — everybody is happier except the health care system as a whole,
which is paying more money for no better results.”
The cost of the procedure varies from about $30,000 to $50,000, and more than
one million are performed every year in the United States. That makes a total of
around $40 billion a year for doing something unnecessary that can kill you. The
risk of death is about one in a thousand. 16
Dr. Harlan Krumholz, a cardiology professor at Yale who was not involved in
the study, said that the findings contain a lesson for doctors treating heart
patients. “When people are making decisions, it’s important to disclose to them
that this procedure - outside of an emergency - is not known to be lifesaving or
to prevent heart attacks … The vast majority of people who have this procedure
have the expectation that it will help them live longer. That belief is out of
alignment with the evidence.”
Angina is often the symptom that convinces doctors and patients that medical
therapy is not enough and that a stent is required. Yet, in this review, 29% of
people who had a PCI still had angina, compared to 33% of those on medication
- an insignificant difference. 16 One of the authors declared that, “interventional
cardiologists use the analogy of a pipe blocked in a house — it’s a terrible
analogy, but patients accept it. It’s simplistic and erroneous.”
But what about those who really, really need stents - those with seriously
blocked coronaries? Well, again the data are disappointing. The first truly
placebo controlled trial was published in November 2017. 18 Researchers
inserted catheters in all 200 patients but only stented half of them. All patients
had severe (≥70%) single-vessel stenoses. The primary outcome was quite
ridiculous: exercise time. Yet that utterly irrelevant outcome is recommended by
both the US and the European drug regulator. Stenting was not beneficial. The
exercise time increased a little more in the stent group, but the difference was not
statistically significant, 16.6 seconds (95% CI –8.9 to 42.0, P = 0.20). Moreover,
the average exercise time was 510 seconds, or 8.5 minutes, before the
intervention, which increased by half a minute in the stent group and by a
quarter of a minute in the placebo group. So what?
What I described above about stress tests for chest pain - although shocking - is
not an example carefully selected among others. I read the story about the US
lawyer a few days before I decided to write this book and thought it might be an
interesting example to work with. What I have demonstrated here is how
important it is to be clear about the questions being asked. The more specific you
are, the better your searches will be and the more likely you will be able to
answer your questions, or at least become clear about just how much you need to
extrapolate from the information you find.
The story is by no means atypical. Doctors routinely use diagnostic tests, rarely
thinking about whether it has ever been demonstrated that the tests are useful,
i.e. whether they make any difference to therapeutic decision making, or whether
their use leads to more benefits than harms.
The same can be said about many of the interventions doctors employ, in this
case stents. Yet this could also simply be a matter of prioritizing profits over
patients - or a matter of convenience for the staff, which the next example
demonstrates.
One reason why it makes sense to take daily temperatures is that hospital
acquired infections are very common. If you google this term, one of the
suggestions is “hospital acquired infections statistics 2016.” You can see that the
US Centers for Disease Control and Prevention issued a report saying that, on
any given day, about one in 25 hospital patients have at least one healthcare
associated infection. The report also says that hospital acquired infections are
major, yet often preventable, threats to patient safety. In plain language: hospital
infections kill many people!
I asked my superiors back then if there was any evidence that the new
thermometers were equally reliable as the old ones. Although no one had
investigated this question, they did not believe so. They also failed to protest a
hospital decision that was made by people who were not clinicians. It seemed to
me that this fundamental issue had never been tested. The reliability of
diagnostics should be measured in a clinical practice setting, but these
thermometers had only been tested in laboratories. The electronic oral
thermometers gave reproducible and accurate results when dipped into warm
water but, in practice, they proved very unreliable compared to rectal readings.
That was demonstrated in 1991. 19
I raised the issue again when oral thermometers were substituted by electronic
temperature measurement in the ear. Sometime later, one of my colleagues
finally did the study I had called for. 20 He compared an infrared tympanic
thermometer with electronic rectal thermometer measurements in 121 patients
admitted to a geriatric department. Five per cent of the differences were larger
than 1°C, which means that ear thermometers should not be used. Eight years
later, it was shown in a systematic review that infrared ear thermometry would
fail to diagnose fever in one-third of children with rectal temperatures of 38°C or
above. 21 That review was published in 2006. Yet my hospital and, I believe,
other hospitals in Denmark, still do not measure body temperatures and instead
employ readings obtained by ear thermometers.
This story and the one about chest pain earlier illustrate that - in issues big and
small - we often fail to live up to the principles of evidence-based medicine. The
issue about mercury is no longer relevant because we now have rectal
thermometers without mercury. For more than 30 years, we have used unreliable
thermometers despite the fact that hospital acquired infections are significant
causes of death. If you google hospital acquired infections deaths, you will find
a report that says that hospital acquired infections affect 5-10% of the patients in
the US each year, and result in 99,000 deaths plus an estimated $20 billion in
healthcare costs. Considering the high death toll and all the hugely expensive
high-tech equipment we have in our hospitals, not wishing to know the body
temperature of our patients does not make sense.
Thirty years and billions of dipsticks later, I suggested that one of my PhD
students should look into this. We asked several public funders for support, but
no one was interested in financing our project which I therefore paid for out of
my government budget. We published a Cochrane review and concluded: “We
found no evidence assessing the benefits and harms of screening with urinary
dipsticks, which remain unknown.” 22 You can easily find this review by
googling dipstick cochrane.
If someone asks you to pee for a dipstick test, with no suspicion that you suffer
from a disease in that region of the body, you should decline. You should also
tell your healthcare provider to read the Cochrane review about urinary
dipsticks.
5 Infections
Infections are very common, particularly when there is overcrowding and poor
hygiene. This is often the case in nurseries and kindergartens, and many small
children and their parents suffer immensely from virus infections, which in my
country are called “nursery plague.” I was particularly hard hit as a parent and
sometimes would cough for months and sleep badly, too. I also developed
secondary bacterial pneumonias that were treated with antibiotics. I often
wondered whether having all this misery for so many people was really the best
way to arrange childcare. At one point, we had daycare in our own home, with
another family’s child, and were in much better shape.
We finally could not take more and, in the most diplomatic way we could, talked
to the head of the nursery. My wife is a clinical microbiologist who knows how
important hygiene is for limiting the spread of infections. Yet we got nowhere -
the nursery head was highly defensive and totally resistant to our arguments.
However, we learned years later that disinfectant alcohol dispensers had been
installed in the institution which the staff actually used. Still, there had been a
great number of rather strange discussions about whether people might become
allergic to the disinfectant or whether the children might be harmed by it.
Hygiene is the best public health remedy we have yet. Being so low-tech, it is
difficult to convince even health professionals of its importance. Males seem to
be less hygienic than females. At an international surgery congress, for instance,
20% of the males did not wash their hands after visits to toilets. 1
When I have a cold, I tell people why I avoid shaking their hands. Washing your
hands before eating is also a good idea. You might have picked up infectious
particles in the air, and virus infections can often even spread from solid objects
you have touched. Many people think it is mostly a matter of avoiding being
near other people who are coughing or sneezing, but proper hand hygiene is
probably most important.
Doctors use antibiotics far too often, even for infections that in all likelihood are
caused by viruses. In many countries, antibiotics can be bought over the counter,
adding to the problem with the development of resistance. In Northern Europe,
the occurrence of resistant bacteria is relatively rare, e.g. less than 1% for the gut
bacterium Klebsiella, whereas over 50% are resistant in Greece. 2 In Latin
America, Africa and Asia, it can be far worse, because of the many species of
bacteria including the much-feared methicillin-resistant Staphylococcus aureus,
also called the yellow Staphylococcus.
Should you pause with exercise when you visit such countries? Perhaps
outdoors. It is safer to run on a treadmill in a fitness facility but also very boring.
The kilometres feel much shorter when you are outdoors in a varying landscape.
Vaccines in general
Some people are ideologically opposed to vaccination. I have never understood
the idea behind this fundamentalism, especially when parents are not only
exposing their own children but even other children to avoidable risks by
forgoing common childhood vaccinations. ‘Herd immunity’ is important. To
prevent the occurrence of epidemics such as measles, vaccinating a large
percentage of the population is necessary.
The arguments I have heard most often remain unconvincing. One of them is
that it somehow would be better to let your child suffer from measles rather than
avoiding measles through vaccination. This ‘back-to-nature’ romanticism is
untenable. If we accept the general premise that it is better to allow nature to run
its course, we would have to abandon developed society, which is very far from
being ‘natural.’ We are no longer big apes in Africa - we are doing far better than
that and surviving longer than we ever did. If we do not vaccinate our children,
there will be far more cases of severe brain damage and deaths than if our
children are vaccinated. This is so well-documented that I will not even bother to
look up the evidence.
One relevant argument against vaccines is that they are not as safe as people
think because the drug industry has cheated with its research. Several readers of
my book about the drug industry 5 – essentially an industry of crime - asked me
why I did not write about vaccines in the book. The reason is pragmatic. I could
not write about everything and it did not occur to me that vaccines would be
particularly interesting because I had learned that vaccination is one of the
greatest advances in healthcare. But vaccines should be scrutinized with the
same rigour as drugs and, after I took an interest in the HPV vaccine controversy
(see below), I came to realize that data manipulation, cheating and cover ups can
be important for our perception of vaccines.
Earlier in the 1700’s, inoculation with live smallpox was already standard
practice in England, but it involved serious risks - both to the recipient and
others - because those inoculated could pass on the disease after becoming
carriers. Inoculation involved introducing material from smallpox pustules into
the skin, which generally produced a less severe infection than naturally
acquired smallpox, yet still induced immunity to it.
Whether an infection is deadly or not has much to do with the initial exposure to
microorganisms. Danish anthropologist Peter Aaby has done ground-breaking
research in this area. Aaby did studies in Africa and elsewhere that disproved the
dogma that malnourishment played any significant role in measles mortality. 8
By using patient files from the department of infectious diseases in Copenhagen
from 1915 to 1925, he confirmed his findings from Africa that the more children
there were in a family, the higher the death rate during measles epidemics. 9 He
concluded that this was because conditions of overcrowding resulted in more
intensive exposure within families, transferring greater doses of the virus, and
that the children therefore died before they had mounted an effective immune
response. Aaby’s findings were met with similar disbelief by ‘the learned bodies’
as Jenner’s.
Aaby has published other important research. Vaccines can affect non-targeted
diseases both positively and negatively. The immune system is immensely
complicated, and it is not possible to predict which ancillary effects a targeted
vaccine might have - only empirical studies can show what they are. Bacillus
Calmette-Guérin (BCG) for tuberculosis and the measles vaccine probably
reduce mortality from pneumonia and sepsis. In contrast, the combined
diphtheria, tetanus and pertussis (DTP) vaccine is suspected to double overall
mortality from unrelated infections in low-income countries, which is worrying
because pneumonia and sepsis are bigger killers than the targeted diseases in
such settings. 10,11 These findings did not make Aaby popular at the WHO
headquarters. Public health messages become difficult when such totally
unexpected results appear.
In Western settings, I have no doubt that the combined DTP vaccine is beneficial
and that we should all receive it. Before we had these vaccines, many people
died from diphtheria, tetanus and pertussis. I clearly remember what it was like
to have whooping cough - truly horrible. I coughed constantly, and I sounded
like a sea lion. I also know what it is like to have mumps. The pain from my
enlarged salivary glands was so intense that I cried a great deal. I could not eat
and smile without experiencing unendurable pain.
It is cruel to expose our children to the horrors of infections that can be so easily
avoided through vaccination. But what about all the other childhood
vaccinations? My private rule of thumb is that if a vaccine is part of the official
vaccination program in some countries and not in others of similar standing, then
it is not so important to get your child vaccinated.
These are the key issues for vaccinations: What is the risk of getting infected and
what is the risk of getting seriously harmed or dying? The WHO speaks about
countries with high fatality rates but what about your own country?
It could work like this: Imagine you find out the risk of getting infected during a
two-week stay in a tropical country is about 1:1000 (this is tricky because it is
very difficult to estimate risk). And the risk of developing a serious complication
to the disease is 1:50. Then your risk of getting badly harmed without
vaccination will be 1:50,000. If you read the vaccine package insert or find
similar information online, you will discover that the risk of serious harm from
the vaccination is 1:10,000. That means you can travel to such places five times
unvaccinated without exposing yourself to a greater risk of being seriously
harmed than if you had been vaccinated.
Japanese encephalitis
We can see how this works out in practice. Japanese encephalitis vaccination is
sometimes recommended, but how common is it? When Googling Japanese
encephalitis, various suggestions will appear while typing, including incidence,
which refers to the number of cases in a given year. The first link goes to the
World Health Organization: “The annual incidence of clinical disease varies both
across and within endemic countries, ranging from <1 to >10 per 100,000
population or higher during outbreaks.” If we use the high-risk estimate, 10 per
100,000, which is 1 per 10,000, we should divide this risk by 26 to get the risk in
two weeks - only 1 per 260,000.
The WHO site explains that the case-fatality rate can be as high as 30%.
Permanent neurologic or psychiatric sequelae can occur in 30–50% and there is
no cure. The WHO also mentions that, “Safe and effective vaccines are
available.” You should never believe such statements. Nothing is safe, and you
should try to find out how often people get seriously harmed with and without
the vaccine, and what the harms are.
In medical jargon, people rarely use the term harms; they talk about side effects
which is a way of toning down the inevitable - that all treatments can do harm.
As long as this tradition prevails, it is best to search for side effects. Adverse
reactions can also be a useful term.
Since serious harms are often downplayed or omitted (most entries I looked at
were about common side effects), I included the FDA (the US Food and Drug
Administration), often highly useful, in my search: japanese encephalitis
vaccine side effects FDA. The first entry was the Product Information for a
vaccine distributed by Sanofi Pasteur.
So, many people are harmed by the vaccine. And since we can die from it, there
must be situations where more people actually die from being vaccinated than
not; this depends entirely on the risk of getting infected. Furthermore, Japanese
encephalitis is transmitted by mosquitos, so bed nets and insect repellents offer
good protection. In my view, that means vaccination for Japanese encephalitis
should not normally be recommended.
I have never been vaccinated against Japanese encephalitis because I knew that
the risk of infection was extremely low. But the WHO recommends travelers get
vaccinated when spending extensive time in endemic areas (i.e. areas where the
disease regularly occurs). The Danish Handbook for Doctors says that adults and
children over 2 months of age traveling to areas in Southeast Asia and the rest of
Asia at certain times when Japanese meningitis occurs with increased frequency
should be vaccinated. The US Centers for Disease Control tell you to get the
vaccine if recommended and, “Talk to your doctor about your travel plans: Your
doctor can help you decide if you need the JE vaccine based on the length of
your trip, the areas where you will be traveling, and your planned activities.”
When I was young, evidence-based medicine had not been invented. I received
most of the recommended vaccines including those against smallpox (now
eradicated), yellow fever, cholera, typhoid, and hepatitis A. I also received
gamma globulin before going on a primitive trip to Kenya in 1980, even though
I was skeptical about getting an injection of antibodies from multiple donors
because I could not know whether an infection might be transmitted that way. I
would definitely not get such an injection today and I believe it is not
recommended. The FDA Product Information for Gammagard (immune
globulin) only mentions people with immunodeficiency as recipients of the
product, not travelers; it also lists pretty horrible harms, e.g. thrombosis, renal
failure and death. This is not a product to be taken lightly but, in 1980, my
doctor did not care.
People’s perceptions of the risks they are willing to take vary enormously, which
is why we decide differently given the same evidence. For instance, although
rabies is extremely rare, I would get vaccinated for rabies if bitten by a dog or
bat in tropical countries, or by a squirrel in North America. It takes time for
rabies to kill you, and post-exposure vaccination is effective.
When I was a child, there weren’t many vaccines. I would not hesitate for a
moment to recommend common vaccines like those against measles, mumps,
rubella, diphtheria, polio, tetanus, whooping cough, and pneumococci (which
protects against pneumococcal meningitis, among other things).
HPV vaccines
HPV vaccines are intended to prevent deaths from cervical cancer by reducing
the risk of getting infected with the human papillomavirus. Important
uncertainties have made them controversial. A great deal of public debate has
been sparked, particularly in Denmark where research at the Syncope Centre
suggested that some of vaccinated girls may have been seriously harmed by the
vaccines. These harms include postural orthostatic tachycardia syndrome
(POTS), complex regional pain syndrome (CRPS) and chronic fatigue syndrome
(CFS).
The Ombudsman’s decision came a year later, and we uploaded a final report
commenting on that decision two weeks later. The various documents are
available at http://www.deadlymedicines.dk/category/blog/. Our most important
observations are these:
EMA asked the manufacturers to search for possible harms in their databases and
EMA did not react to the fact that the search strategies used by the companies
were grossly insufficient and must have overlooked many cases.
EMA declared that the adjuvants used in the vaccines to boost the immune
response were safe, yet the five references EMA provided in support of this view
were either non-accessible or irrelevant. We investigated the issue ourselves and
found no evidence that adjuvant safety studies had ever been carried out.
EMA allowed the manufacturers to lump the ‘placebo’ groups in their trials even
though none of the trials were genuinely placebo controlled. In almost all the
trials, the placebo was either an adjuvant or a hepatitis vaccine. If these active
‘placebos’ cause harms similar to the HPV vaccines, it would be difficult or
impossible to use the trials to find out if the HPV vaccines cause the suspected
rare harms.
EMA did its own literature searches but withheld the results from its own
scientific advisory committee. These withheld reports revealed that HPV
vaccines, other vaccines and perhaps also the adjuvant (in combination with an
otherwise harmless virus infection) might potentially cause POTS or CRPS in
some people.
It seemed like a cover up that EMA withheld this information from its own
expert committee which, moreover, wasn’t necessary because the committee
worked under a life-long confidentiality clause.
The WHO Uppsala Drug Monitoring Centre and the Danish Health Authorities,
both of which had found signals of harm, were dissatisfied with how their
observations and reports were dismissed by EMA.
We felt that EMA had committed scientific misconduct in relation to the HPV
vaccines.
Suppose you know nothing about the HPV vaccines. How do you find reliable
information that can help you decide whether you should get your 12-year old
daughter vaccinated?
The usual approach even works this time. When I googled hpv vaccine cochrane
in February 2019, the uppermost entry was a link to a criticism our research
group published in July 2018 of the Cochrane review of the HPV vaccines 14
(which was the second link from the top).
The Cochrane review did not find an increase in serious neurological harms. We
did, because we used clinical study reports we had obtained from the European
Medicines Agency, which are much more reliable than what drug companies
publish. Using a blinded observer, we also found more symptoms associated
with POTS and CRPS in the HPV vaccine groups than in the groups treated with
a hepatitis vaccine or with the adjuvant. We communicated our results at our 25
th Anniversary Research Symposium on 12 October 2018 and they are included
The Danish National Board of Health reacted with considerable arrogance which
has not furthered their case. They declared that the HPV vaccines are unique,
being the first vaccines against cancer, which they are not: the hepatitis B
vaccines protect against liver cancer. They launched a public campaign in 2017
where they emphasized the importance of the evidence hierarchy which has
systematic reviews of randomized trials at the top and consensus statements at
the bottom. However, since we do not have placebo-controlled HPV vaccine
trials, neither can we have reliable systematic reviews of the harms.
Furthermore, the Danish Board of Health refers to EMA when saying that the
vaccines are safe, even though EMA’s work produced a flawed report where the
participants were legally bound to reach consensus - the bottom of the evidence
hierarchy.
On its website, the Board of Health provides a list of references in support of its
position, but one glaring omission is an important study carried out by the WHO
Uppsala Monitoring Centre. 19 This institution found that a combination of
headache and dizziness, with either fatigue or syncope, was more commonly
reported in HPV vaccine reports than in non-HPV vaccine reports for females
aged 9-25 years, and that this disproportionality remained when those countries
reporting the signals of CRPS (Japan) and POTS (Denmark) were excluded. The
Uppsala Monitoring Centre also avoided any possible media attention influence
by including only cases reported before 2015. Even so, they identified a greater
number of potentially undiagnosed cases than the total number of cases that had
been labelled with one of these diagnoses by the drug companies when asked by
EMA to find cases in their databases. Contrary to EMA’s reassuring statements
to us, 20 it is not “a very conservative approach” to let the drug companies
exclude a large amount of cases diagnosed by a skilled Danish clinician without
inspecting the underlying raw data, or to trust drug companies when they report
far fewer cases than the Uppsala Monitoring Centre found.
The director of the Danish Board of Health talked about alternative facts when
researchers were skeptical of the work done by the drug companies and by
EMA. He launched the idea that opposition to these vaccines is the result of that
we are living in “a postfactual society.” The authorities often do what they can to
shut down uncomfortable debates with the mantra, “It’s time to move on…” but
denigrating remarks like these can have the opposite effect.
The public health perspective is that cervical cancer is a terrible disease, that we
can avoid many deaths through vaccination, that the harms are trivial compared
to the benefits, and that everyone in a given age range should get vaccinated.
The declining rate of vaccination in Denmark has been described by the Board
of Health as a catastrophe lurking just around the corner if we do not get that rate
up.
However, the public health perspective is misleading and not much in line with
the evidence.
First, it is difficult to see any lurking catastrophe. Dying from cervical cancer is
rare. Only some 100 people in Denmark die from it every year - about 15,000
people die from smoking. Thus, if we want to do our best to keep women alive,
we will do far better if we use our resources convincing young girls not to start
smoking rather than convincing parents that their girls should get vaccinated
against HPV. Increasing the price of cigarettes has proven highly effective but
we have seen no such initiatives.
Second, what is the effect? We actually do not know. The vaccines were
approved because they lower the risk of infection with some HPV strains that are
known to cause cancer. They also lower the risk of cell changes that are
precursors to cancer. But they are only about 70% protective against these
particular HPV strains and there are other strains that can cause cancer. We do
not know if these other strains will take over and cause cancer, or for how many
years the vaccines are protective. Furthermore, most cell changes will disappear
if left untreated. 14 Therefore, although it is very likely that the vaccines will
reduce deaths from cervical cancer, it has not been documented.
Third, a key question is always: When will the benefit come? Most people will
be surprised to learn that about half of those who die from cervical cancer are
over 70 years of age. Official statistics show that only about 12 women in
Denmark under 45 years of age die of cervical cancer each year. If we assume
that all 12-year old girls become vaccinated and that the vaccines are 70%
effective, then about 8 women will be spared each year. We cannot deduce the
numbers of those who are seriously harmed by the vaccines because the
manufacturers have made it very difficult to study the incidence of harms in their
trials where there are no placebo controls. Even so, a systematic review of the
published trials from 2017 found more deaths in the vaccine groups than in the
control groups (14 vs. 3, p = 0.01) and more serious systemic adverse events in
girls receiving the 9-valent dose than in those receiving the 4-valent vaccine (3.3
vs. 2.6%, p = 0.01). Yet none of the serious adverse events were judged to be
vaccine-related. 21 I find it quite ‘interesting’ when clinical investigators – who
probably have conflicts of interest with sponsoring drug companies - decide that
none of the serious adverse events are vaccine-related.
There are also reports of deaths caused by the vaccine. In Spain, a young woman
with asthma had a severe exacerbation when she received the first shot of the
vaccine. Despite that, she got a second shot a month later and developed severe
dyspnoea and seizures 12 hours later. She was admitted to an intensive care unit
where she died two weeks later. The judicial ruling acknowledged a causal link
to the vaccine. 22
In 2013, the WHO introduced new criteria for the assessment of causality of
individual adverse events following vaccination. 23 These criteria have been
heavily criticized - and rightly so in my view. They make it almost impossible to
detect signals of serious harm - including deaths - after vaccinations. 24 I have
never seen so many comments on PubMed as those related to the abstract about
the new WHO criteria. Both papers make for chilling reading. 23,24
Added to that, propaganda about the safety of HPV vaccines have the effect that
many doctors do not report their suspicions of serious harms. They either ignore
them due to public reassurances about safety, or because they worry about
getting into trouble if they report.
In September 2008, Kent Woods, head of the UK drug regulator - the MHRA
(Medicines and Healthcare Products Regulatory Agency) - sent a letter about the
HPV vaccine to healthcare providers. The immunization program for Cervarix
had just started and Woods asked healthcare providers to report any suspected
adverse reactions via the Yellow Card Scheme, while discouraging them from
reporting fainting that occurred during or very shortly after vaccination, as these
were usually a psychogenic response to needle injection.
One year later in October 2009, Woods sent a second letter mentioning that there
had been stories in newspapers about adverse events including chronic fatigue,
but - based upon reported events against expected background rates - there was
no reason to believe a causal link existed between the vaccine and chronic
fatigue.
Since doctors had been encouraged to report adverse events one year earlier and
were now told that their reporting had not identified problems, the second letter
probably discouraged some doctors from reporting not only chronic fatigue, but
also symptoms of POTS and CRPS as adverse events. The letter was sent right
in the middle of the study period of an investigation that was used by EMA in
2015 to dismiss concerns about serious neurological harms of the HPV vaccines.
25 One of EMA’s key arguments, which appeared no less than ten times in its
official report, was that there was no difference between what was observed and
the expected background incidence of serious harms. All the authors of the paper
reporting on this investigation were employees at the UK drug agency. The main
analysis was based on spontaneous reports compared to the background
incidence, and since the data collection period for the study was the first two
years of the HPV program, it included the period when doctors had been
discouraged to report their suspicions. That was not mentioned in the paper.
I must admit, after studying what goes on behind the scenes, I realized why so
many people are skeptical about vaccines. Some of the most hard-core skeptics
are not rational - but some surely are. We should ask questions. There are many
examples of doctors being actively discouraged by the authorities from reporting
suspected serious vaccine harms, of which the recent WHO initiative mentioned
above is probably the worst.
The public health perspective has been much hyped, and the potential harms
downplayed or omitted altogether. The individual perspective is very different. If
I get vaccinated, will I benefit? That is highly unlikely, and there is a risk of
serious neurological harm, about which we know too little. How many do we
need to vaccinate to save one life? That number has never been disclosed. In
Denmark, around 32,000 girls are 12 years old. Assuming we can save 8 of them
each year (see above), that means the NNT (the number needed to treat) is
approximately 4,000. Thus, the benefit potential is only one per 4,000. What
number does it take to seriously harm one person? We do not know. When
authorities say the vaccines do more good than harm, it is probably correct. Yet
we actually do not know.
What can a girl do instead of getting vaccinated? She can attend regular
screenings. This method is not ideal because it leads to many conizations
(removal of part of the cervix) due to the frequency of detected cell changes. If
you want to know what the risk of preterm birth is after conization, you can
google conization risk preterm birth meta-analysis, which leads you to a recent
meta-analysis that found that the risk doubled, from 5.4% to 10.7%. 26 Since
many cell changes never develop into cancer, it should be possible to adopt a
wait and see approach, thus substantially reducing the amount of conizations.
Unfortunately, the debate about the HPV vaccines has become highly polarized
and oversimplified, often reduced to a matter of whether you are for or against
the vaccines. The Nordic Cochrane Centre has never expressed any views about
whether the vaccines do more good than harm apart from saying that we find
that likely. My deputy director vaccinated his two girls and my wife vaccinated
our two girls. But my wife and I - who both have a background in infectious
diseases - were in doubt when she vaccinated our eldest daughter in 2008. Even
that far back, thousands of reports had been submitted to the authorities of
adverse effects including nausea, paralysis and death. Many parents and
physicians worry about the vaccine health risks, even though it is difficult to
know whether these adverse events had anything to do with them. 27 Scientific
uncertainty is illustrated by the fact that my deputy would still vaccinate his girls
today. We would not.
Upon our eldest daughter’s 12th birthday, we received a letter from a doctor
asking us to enroll her in an HPV vaccine trial conducted by GlaxoSmithKline. I
told my colleague we were considering letting her participate but we needed to
see the trial protocol in order to make a fully informed decision. Although our
request was completely reasonable, I did not expect it to be granted. Many years
later, when we asked to see protocols for ongoing trials in Denmark for use in a
research project about informed consent, our request was declined by several
local research ethics committees. 28 We complained to the national research
ethics committee, which then offered us full access provided we signed a
confidentiality agreement stating we would not share commercially confidential
information in the protocols with others. Despite this reassurance, several
companies refused to provide their protocols and involved their lawyers. Sanofi-
Aventis sued the national committee but lost the court case.
To my big surprise, my colleague sent the HPV vaccine protocol 29 after I had
signed a confidentiality agreement and promised to return it after reading. I
explained to him that I had two reservations:
“There is nothing about harmful effects in the 105-page protocol, only some
non-informative comments like ‘generally safe and well-tolerated.’ The readers
are referred to the Investigator's Brochure about this. In the parents’ information
you can read that the vaccine has ‘affected the nervous system, blood cells, the
thyroid and the kidneys.’ It would be relevant for us to know what that means
and the frequencies of such potentially serious harms. If this information is
included in the Investigator's Brochure, hopefully, it would be possible to hand it
over to us. You cannot make an informed choice if you do not get statistics on
the side effects.”
“It appears on pages 79-83 in the protocol that Glaxo owns the data and the
investigators do not have any realistic possibilities of publishing the trial without
the company's permission, among other things, because the company must
approve publications and because individual investigators will not gain access to
all data from the trial - only their own data.”
My colleague responded that it was not possible for him to send me the
Investigator’s Brochure, unfortunately. He did not explain why.
Several parents of our daughter’s girlfriends had also been contacted and asked
me if they should enroll their daughter in the trial. I told them not to and
explained why I declined.
Some doctors have suggested that these girls suffer from psychiatric problems.
That may be the case at times, but I doubt if it is common, and postulating that
all of them are having these issues is not only wrong but even deeply insulting.
When I was a child, women with menstrual cramps were called hysteric (
hysteros in Greek means womb). That changed when it was discovered that
prostaglandins caused the pain and that prostaglandin synthetase inhibitors had
an effect on this pain. Yet old habits die slowly.
Influenza-like illnesses
Vaccination against influenza is also a hot topic. When people say influenza,
they mean influenza-like illness. The main Cochrane review is easily found (
influenza vaccination cochrane ). 31 The preventive effect of influenza vaccine
on healthy adults is small: at least 40 people would need vaccination to avoid
one case of influenza-like illness and 71 people to avoid one case of influenza.
The vaccination did not show an appreciable effect on lost working days or
admission to hospital.
As the virus mutates quite rapidly, we cannot be sure that the effect obtained by
vaccination will be the same as in the randomized trials. Furthermore, as for all
interventions, it is important to consider the harms. When doctors first alerted
their colleagues to the possibility that Pandemrix, one of the influenza vaccines
used during the 2009-2010 pandemic, could cause narcolepsy in children and
adolescents with a certain tissue type, their reaction was to ridicule these doctors.
It has now been firmly established that Pandemrix can cause narcolepsy, a very
serious condition where people suddenly fall asleep, up to several years after
vaccination of children and adolescents, and that the disease is immune-
mediated.
There is a review of people aged 65 and older, but since only one trial was found
- no conclusions could be made. 32
Some fundamentalists, particularly in the United States, can only see positive
consequences of these vaccinations. When such people are in power, things go
badly wrong. In 2017, a senior faculty member at New York University School
of Medicine, who did not do any clinical work, had her faculty appointment
terminated because she did not get an influenza vaccination. 33 The university
stated that, "immunization against the flu is critical for the protection of our
patients, visitors, and colleagues. Regrettably, since we have not received
evidence of your vaccination, your non-compensated faculty appointment will be
terminated effective immediately."
One Cochrane review is about vaccination of healthcare workers who care for
elderly people. 34 Almost 13,000 healthcare workers were included in the review,
but the authors could not find any conclusive evidence of benefits of this
vaccination program on laboratory-proven influenza, lower respiratory tract
infection, hospitalization, death due to lower respiratory tract illness or all-cause
mortality.
Canadian researchers also recently reviewed the evidence and agreed with the
Cochrane researchers. 33 There is no valid evidence to support the supposition
that vaccinating healthcare workers protects patients from influenza. One
researcher mentioned that vaccination for the H3N2 subtype resulted in a
vaccine effectiveness of around 40%, which means that three out of five
vaccinated healthcare workers remain as susceptible to H3N2 as if they were
unvaccinated. She added, "In that context, to focus exclusively on the risk posed
by unvaccinated workers - treating them as outcasts or, worse, terminating their
employment - while overlooking the risk posed by vaccinated workers,
potentially jeopardizes patients.”
Indeed. Vaccination may provide staff with a false sense of security. That might
reduce their level of handwashing and potentially increase, rather than decrease,
the risk of infecting patients.
There are other Cochrane reviews about influenza vaccination in patients with
various diseases which you can look up if you are interested.
I have written about this in another book 5 and shall only repeat a little of it here.
Roche has committed what to me looks like the biggest theft in history, but no
one has dragged the company to court for fraud. Roche omitted publishing most
of their clinical trial data and refused to share them with Jefferson and other
independent researchers. Based on unpublished trials, Roche claimed that
Tamiflu reduces hospital admissions by 61%, secondary complications by 67%,
and lower respiratory tract infections requiring antibiotics by 55%. I consider
that fraud. As mentioned earlier, fraud means a wrongful or criminal deception
intended to result in financial or personal gain, typically by unjustifiably
claiming or being credited with accomplishments or qualities.
The FDA sent Roche a warning letter that the company should stop claiming that
Tamiflu reduces the severity and incidence of secondary infections while,
curiously, the European Medicines Agency swallowed the bait and accepted their
claim that the drug reduced lower respiratory tract complications.
Many people have wondered why the WHO selected people to write guidance
about influenza drugs who were paid by the companies marketing the drugs, yet
did not disclose this in their guidance reports. There was so much secrecy, it was
not possible for outsiders to even get information about who was on the WHO
committee.
The scandals seem endless. Following criticism of the US Centers for Disease
Control (CDC) and its foundation for accepting a directed donation from Roche
for the agency’s Take 3 Flu Campaign, which encourages the public to “take
antiviral medicine if your doctor prescribes it,” the CDC posted an article on its
website entitled: Why CDC recommends influenza antiviral drugs. 37 The agency
cited multiple observational and industry funded studies, including a meta-
analysis it described as “independent,” even though it was sponsored by Roche.
All four authors had financial ties to Roche, Genentech, or Gilead. Despite its
extensive list of studies, the CDC did not cite the Cochrane meta-analysis.
The CDC director told the public that these drugs could “save your life,” which
looked like classic stealth marketing where the industry places messages in the
mouths of trusted third parties. There is no reliable evidence that these drugs
save lives and it is not even likely that they do.
Roche refused to give the Cochrane researchers access to its unpublished trials
or to its clinical study reports but, after a four-year campaign supported by BMJ,
Jefferson and colleagues succeeded in getting the data they needed. As a
consequence, they completely changed their conclusions compared to their
previous Cochrane review which was based on published trial reports. There
were serious discrepancies between what had been published and what was only
available to drug regulators. As usual. 5
I often use this kind of reasoning when people tell me this or that has miraculous
effect on something. I cannot know everything, but what I do know is, if it were
true, I would have read about it in my favourite medical journal - the BMJ.
We could stop here. Forget about magazine articles: endure your colds like the
rest of us and accept that nothing works. We cannot do hard work finding the
evidence or the lack thereof every time someone speaks of miracles. We need
some decision-making shortcuts. Since the common cold will not kill us, it is no
big deal.
If you wish to do more, you know what to do: Google common cold cochrane
and look up the Cochrane review about vitamin C. The first entry you see is the
summary for patients. 38 It tells you that over 200 viruses can cause common
cold symptoms and antibiotics are useless.
Vitamin C became particularly popular in the 1970s when Nobel laureate Linus
Pauling concluded from placebo-controlled trials that vitamin C would prevent
and alleviate the common cold. However, Pauling’s book about this is an
appalling example of selective citation, of which we might not expect a Nobel
laureate to be guilty. 39
Although this trial has been criticised, 38 we cannot disregard the fact that bias
can easily explain small differences in highly subjective outcomes. Furthermore,
even if the 8% difference in duration was true, it would not be clinically
relevant. If a cold lasts 12 days without treatment, it would last 11 days with
treatment.
And we have not even looked at the harms yet. According to the Cochrane
review, the published trials did not report adverse effects of vitamin C, but that
does not mean that there were none. Google vitamin C side effects and you will
see. According to the Mayo Clinic, “High doses of vitamin C have been
associated with multiple adverse effects. These include blood clotting, death
(heart-related), kidney stones, pro-oxidant effects, problems with the digestive
system, and red blood cell destruction.”
Vitamins are not regulated the same way as drugs and there are many fraudsters
at large who see them as goldmines because many people naïvely assume that
they must be good for them. You will therefore not find much by googling
vitamin c fda. Yet the first entry links to an interesting warning letter from 2017.
41 The FDA writes to the Vitamin C Foundation that it violates the law by
• The ability of vitamins C, E, and selenium orally to prevent most cancers, and
to decrease the mortality of pancreatic cancers, stomach cancers, prostate
cancers and other cancers.
• Almost all vitamin C experts agree, hydrogen peroxide produced by high levels
of ascorbate (vitamin C) kill cancer cells.
Vitamin C was also supposed to work for all forms of cardiovascular disease:
• In 1994 Lunes Payling [ sic ], Ph.D, and his associate Matthias Rath, MD,
patented LP(a) Binding Inhibitors for the prevention and reversal of common
cardiovascular disease cause by a sub-clinical vitamin C deficiency…
• [H]igh-dose vitamin C treatment…for all forms of cardiovascular, disease,
including congestive heart failure, heart disease and stroke.
It seems we need no other drugs than vitamin C. All that is total nonsense and
not even the spelling is correct. Rath is a German doctor who promoted
nutritional supplements for people with HIV in South Africa and argued that
antiretroviral drugs were harmful. 42 He did not work with “Lunes Payling,” but
Linus Pauling on the possible therapeutic effects of micronutrients and claimed
to be his nominated successor. The Dr. Rath Foundation sells micronutrient
supplements through a website. In 2002, the UK Advertising Standards
Authority ruled against Rath over a newsletter entitled Good health: Do it
Yourself, which claimed that nutritional supplements could prevent cancer and
heart disease. In 2005, Rath began to distribute one of his products, VitaCell, to
HIV positive people in Khayelitsha in South Africa, while he attacked the drug
industry as profit hungry and unscrupulous. The Treatment Action Campaign,
supported by Médecins Sans Frontières whose doctors opened the first
antiretroviral clinic in South Africa in Khayelitsha, said some people taking the
drugs stopped in favour of the vitamin supplement. Campaigners and medical
staff have given evidence that some people died.
The authors of the Cochrane review did not have the guts to tell us what their
review really showed. They concluded:
For heaven’s sake, no! This is the American, “Talk to your doctor,” nonsense.
How on earth would people find out if “therapeutic vitamin C is beneficial for
them.” That is impossible. People should not take vitamin C.
If you google cough cochrane, you will find a Cochrane review with 29 trials
and 4,835 people. 43 Very impressive. If these drugs worked, no more than 100
patients would be needed to demonstrate it. But they do not work. The number
of studies in each category of cough preparations was small and many studies
were poorly reported making the assessment of the bias risk difficult. There was
also a lack of reporting of blinding of outcome assessors and of whether cough
outcome measures were validated. In addition, studies supported by
pharmaceutical companies or other providers were more likely to have positive
results.
The Cochrane authors’ conclusion it pretty typical for doctors: “There is no good
evidence for or against the effectiveness of OTC (over-the-counter) medicines in
acute cough. This should be taken into account when considering prescribing
antihistamines and centrally active antitussive agents in children; drugs that are
known to have the potential to cause serious harm.”
We do not need to go further than this. Considering all the bias in drug trials, and
that most trials are industry funded and manipulated, a negative finding in a
review of almost 5,000 coughing people is an overwhelming reason to scream
NO!
But that is immaterial to the drug industry. In the United States, over-the-counter
cough and cold medications were used by 39% of households during a period of
three years. 44 Many of the drugs came on the market over fifty years ago when
there was little control of medications. Yet, over a period of seven years, poison
control centres reported more than 750,000 concerned calls related to such
products, and the FDA identified 123 deaths of children under six in its database.
The harms of the drugs included cardiac arrhythmias, hallucinations, depressed
consciousness and brain disease.
The manufacturers claimed that the injuries could be prevented through parent
education - a horrendous lie. In 2011, the FDA announced that it would pull 500
cold medications from the market. 45 It was recognized that some drugs - opiates,
for example - may slow down breathing, decrease cough and allow mucus to sit
in the chest where it can cause lung infection. This is still the case today for
many cough medicines.
After the FDA crackdown on the use of over-the-counter cough and cold
medications in children younger than two, the number of emergency room visits
for adverse events decreased by 50%. However, Dr. Harold Nelson, an allergist
at the National Jewish Health in Denver, said: “These are well-recognized drugs
that have been used for decades and there is no reason to suspect that there is a
risk involved.” This foolish statement illustrates quite well how many doctors
actually think. When I tell doctors how dangerous many of their drugs are, they
often dismiss the danger by saying that they never saw anyone in their practice
die on that particular drug. Well, I have never seen anyone die in the traffic.
Does that mean there are no traffic deaths? The industry said, in the five years up
to 2016, revenue was anticipated to grow at an annualized rate of 1.8% to $8.5
billion (which seemed to be in the United States alone apparently). 46 Quite a
revenue for something that does not work and kills children. I cannot see much
difference between pushing these drugs and pushing narcotics on the street.
Very many people use fever reducing drugs - antipyretics - when they have a
fever. I learned many years ago from a colleague in infectious diseases that this
is a bad idea because the body’s immune defense becomes ten times more
effective if the body temperature goes up a couple of degrees Celsius - a
dramatic increase in efficiency. I could not remember whether he talked about
white blood cells or other issues, so I decided to find out when writing this book.
I played around with a few search strategies on Google until I found a highly
relevant paper. Since I had visited many websites that day and could not
remember the search words, I retrieved my ‘history’ on the browser page (e.g.
Firefox or Explorer) by pressing ctrl + h and selecting the option that shows the
most recently visited sites first. The first relevant one was leucocytes fever -
Google Search. I had also tried leucocytes activity temperature and immune
response temperature. It did not take long to find what I was looking for, even
though I expected it to be difficult. I read a few abstracts on the first Google
page I found with my first two search strategies, but my third strategy was right
on target because a very informative paper appered. 47
It was a review article. I have extracted the most important parts from the first
two pages:
The fever response is a hallmark of infection and has been shaped through
hundreds of millions of years of natural selection. The increase of 1 to 4°C in
core body temperature that occurs during fever is associated with improved
survival and resolution of many infections. For example, the use of antipyretic
drugs to diminish fever correlates with a 5% increase in mortality in human
populations infected with influenza virus and negatively affects patient outcomes
in the intensive care unit.
The fact that fever has been retained throughout vertebrate evolution strongly
argues that febrile temperatures confer a survival advantage. A long-standing
mystery relates to the protective mechanisms by which fever wards off attacks
by invading pathogens. One mechanism involves direct effects of febrile
temperatures on the infectious potential of pathogens. For example, temperatures
in the febrile range (40–41°C) cause a greater than 200-fold reduction in the
replication rate of poliovirus in mammalian cells and increase the susceptibility
of Gram-negative bacteria to serum-induced lysis.
In my household, we have never taken antipyretics like aspirin or paracetamol
when we had a fever, and I now know why it is prudent to avoid such drugs. A
mechanism that can be seen all over the animal kingdom, that has likely been
around for more than 600 million years, must have a very strong survival value,
which even experimental studies have suggested.
What will you find when looking up systematic reviews of antipyretics? They
lower the body temperature and perhaps have other small benefits like reducing
headaches. But these are not important outcomes. When we have an infection,
the most important thing is to survive, and the second most important thing is to
cure the infection.
Therefore, do not consider the pros and cons of taking antipyretics, and ignore
the usual nonsense: “Ask your doctor whether an antipyretic is right for you.” It
is not right for anyone, not even for a reptile or an insect.
Meningitis can be caused by several bacteria and even by other organisms, e.g.
viruses and amoebae. The most feared form is the one caused by meningococci (
Neisseria meningitidis ), which can also cause life-threatening sepsis.
The doctor in charge decided to wait and see how the condition developed,
because he felt the boy most likely had viral meningitis. I saw the situation very
differently. I did not have children at the time but felt that if that cute little boy
was mine, I would immediately give him a high dose of penicillin intravenously.
I could see no good reason for waiting since we had already drawn blood for
culture of bacteria and had done a lumbar puncture.
I was a newly qualified doctor, and this happened less than two years after I took
my medical degree. The hospital system is highly hierarchical making it difficult
for young doctors to argue with more experienced colleagues. I cannot recall our
conversation, but I did express my worries about postponing treatment.
However, being a young doctor at the frontline of things, you must not argue
your view too persistently: you need to accept what more senior colleagues say.
Yet the doctor in charge was too academic and that had tragic consequences.
When meningococcal meningitis was no longer in doubt, it was too late. The boy
died. That kind of experience never leaves you, and I still reproach myself for
not speaking out to my senior more loudly, even though there was little chance to
change his mind.
There is a third incident I will never forget. I was now a senior doctor still at the
department of infectious diseases when I got a call from the emergency
department. I was very busy with acutely ill patients and the emergency
department was so far away that I could not abandon my own patients. However,
it was not necessary to see that patient in order to conclude that this man very
likely had sepsis with meningococci. He was fully awake and felt very well but
had small haemorrhages in his skin. I told the junior doctor who had called me
on the phone in no uncertain terms that he should immediately give the patient 5
million units of penicillin intravenously and I even explained why.
After a couple of hours, I had stabilized by own acutely ill patients and went to
see that patient.
I was deeply shocked. I was greeted by the patient: a man in his thirties with a
big smile who looked forward to going home after we took care of his bleeding.
My heart sank when I saw his skin and the large confluent haemorrhages, and I
found out that the junior doctor, despite my clear directions, had not started
penicillin. I saw a happy, young man who did not know he would soon die.
The fact that the hierarchical system was respected in one of the cases and not in
the other resulted in the same tragic outcome. Infection with meningococci is a
hyperacute emergency and whether or not the patient survives can be a matter of
minutes.
Recently, three young boys in Denmark aged 16 to 18 with meningitis died and
huge media attention arose questioning why these deaths had not been avoided.
The official body, the Patient Compensation Board, concluded that all three boys
might have survived if the doctors had treated them adequately. 48 Here is one of
their stories: 49
In the middle of the night, Trine Baadsgaard awakens because her fever-stricken,
16-year old son Mathias vomits. Mathias looks at his hands and says: “Mom,
look at this.” Red spots are on his hands. Worried, Trine finds her computer and
googles to find out what the spots might be. She very quickly finds out that it
could be meningitis. She is shaken.
The dark spots make her use the so-called ‘acute’ phone in Copenhagen. But 35
people are in line ahead of her, so she calls emergency services. She reports that
she thinks her son has meningitis. An ambulance arrives. The ambulance file
states: “Sick child, fever over 38.5 C, seems in poor condition and has a prickly
rash that cannot be removed on pressure.”
A doctor from the ambulance attends Mathias, but even though he has spots on
his body, the doctor rejects the meningitis diagnosis and does not administer
penicillin. The ambulance takes Mathias to the paediatric department at Herlev
Hospital, a major university hospital in greater Copenhagen. Now Mathias has
even more spots. He is being examined, blood samples are taken, but before any
results come back, a new doctor rejects the possibility of meningitis. Instead, the
doctor believes it is a less serious disease, Schönlein-Henoch, which typically
disappears spontaneously. Mathias was only examined for this disease and not
for meningitis.
Mathias is sent home after half an hour at the hospital. His mother is happy
believing he does not have meningitis: she trusts the doctors' assessment. They
talk about going on a skiing holiday a few days later and his mother assures him
that he can go.
The hospital calls and says that there are “some elevated bacterial numbers”
which is not an emergency but that she should bring back her son. Meanwhile,
Mathias is getting worse. His mother helps him get dressed and he vomits. He
has more spots on his legs, he is dizzy, has a headache, pain in the joints and a
lump on his forehead. However, the doctor still believes that he does not have
meningitis or sepsis.
While Mathias is getting worse, his mother wonders why he is not given
antibiotics. At one point, she shouts that her boy should get drugs. She does not
understand what prevents the doctor from giving him penicillin just to be on the
safe side.
Mathias’ father arrives and Mathias greets him. These are the last words he says
to his father. Shortly thereafter, Mathias goes into severe pain, becomes
paralyzed on the left side of the face and becomes more and more confused.
Five hours after Trine’s first call, Mathias finally gets antibiotics. His brain dies
a few hours later. Despite huge media attention, the other two young boys die a
year after Mathias.
Unsurprisingly, The Board of Patient Safety ruled that Mathias had evidence of
meningitis when he arrived at the department and that the staff should have
started antibiotic treatment before receiving the test results. It is unbelievable
that this could happen, and Mathias might have survived if he had been given
penicillin in time.
What is even more unbelievable is that the paediatric department, even after
heavy criticism, did not do enough to improve its routines for identifying
meningitis. Furthermore, the hospital’s report about the case contained several
errors, which the hospital admitted. Unfortunately, this is all too typical. Systems
protect themselves. They always put themselves first, even though we hear so
much about focusing on patients.
I can find no excuses for the wrong-doing in this case. Mathias had classical
symptoms of sepsis with meningococci and every doctor has learned how
important it is not to overlook or dismiss this. On top of that, his mother said she
suspected meningitis all along. Not all doctors are good doctors but dismissing
her concerns is unbelievably arrogant, especially without any good reason for
doing so.
Leg pains, cold hands and feet, and abnormal skin colour are signs of sepsis.
If your doctor does not want to start penicillin based on the suspicion of
meningococcal disease, it will not help to shout. Yet it might help if you say that
you will report the doctor to the hospital administration and to the National
Board of Health if anything untoward happens - and that you have good
connections with journalists. I also think it could help if you ask the doctor to
repeat what he just said while you use your mobile phone to record it.
Our medical education should be radically changed. I read 20,000 pages before
becoming a doctor. I knew virtually everything about receptors and how drugs
were supposed to work. I could write the formulas for the 22 essential and non-
essential amino acids and state the Latin or Greek names for countless
anatomical structures and knew where they were located in the body, as if I was
planning to become an all-round surgeon. But what about those relatively few
conditions that are often lethal if overlooked and curable if not? They must be
repeated again and again, in all possible ways, with patient stories and real
patients, so no doctor could possibly overlook them.
Getting back to the research: Why did it take so long for someone to decide to
study which symptoms are important for diagnosing patients with
meningococcal disease? Far too much emphasis has been placed on one
symptom - neck stiffness - and many doctors have dismissed the possibility of
meningitis because that symptom was not presenting and, in doing so, caused
many deaths that could have been avoided.
Throughout a long career, I have consistently observed that the most important
projects I have ever undertaken - those most relevant to patients - could not
obtain funding. I had to complete them without funding or by using my Nordic
Cochrane Centre government funding. If a science proposal is so complicated
and high-tech that no one really understands it, it is much easier to fund than if
you merely wish your research to be useful to patients and doctors.
Malaria
Malaria continues to take many lives. According to the WHO, around 300,000
African children die each year before their fifth birthday due to malaria. In the
UK, only about six people die annually from malaria. So why mention malaria in
this book? Primarily because the book is written for an international audience
which includes people living in malaria infested areas. Furthermore, like
meningitis, it is unbearable when people die of an illness that is easy to treat.
A fever should always be taken seriously if people live in, or have recently
returned from, a malaria infested area. Falciparum malaria can progress into a
severe and life-threatening illness, including cerebral malaria, if it is not
diagnosed and treated promptly. Travelers returning from risk areas should
therefore seek urgent medical advice for any symptoms, especially fever, if these
occur during their trip or in the twelve months following their return home. A
blood sample should be promptly examined for the presence of malaria parasites
by microscopy. Other types of malaria are benign, but since an outbreak can
occur up to 30 years after exposure to the mosquitos, few doctors would suspect
that a fever might be due to a visit to the tropics so many years earlier.
The night was very hot and they became dehydrated. The slimmest of the two
hallucinated and became seriously disorientated. He saw animals not there and
he wanted to take a shortcut through the desert much longer than their planned
route where they would be very difficult to find. He weaved from side to side
and he was difficult to understand when he talked. He did not notice that the map
he tried to follow was upside down. Later, his companion also became confused;
he suggested another shortcut and could also see non-existing animals.
The natives who found them the next day could not understand why anyone
would be foolish enough to walk in the desert without carrying a spear or other
weapons. Wild dogs were about, and lions sometimes became so thirsty and
desperate they killed people to drink their blood.
When the trip was over, each of us had lost around 6 kg, because where we were,
there was little to eat. I kept a diary and read it aloud in the evenings. My
companions found it so interesting they suggested I wrote a book. I did. 51
I had prepared well for the trip and read books about how to survive in the
tropics. Therefore, I knew I should be on alert when I came down with a fever a
month after my return. Back then, chloroquine was the recommended drug for
malaria prophylaxis even though it was well-known that the parasites had often
developed resistance to the drug.
During my illness I called for doctors who came to see me twice. They did not
find it necessary to examine my blood for malaria, even though I asked to be
admitted to the hospital for examination. They knew that I had just come home
from Kenya after a journey under primitive conditions. They also knew that I
studied medicine and that I was very scared I might have falciparum malaria.
That did not impress them in the least.
So just by looking at me, this wise guy could tell that I did not have malaria!
That is when my story becomes interesting for others. What exactly should we
do in such situations?
I have been anti-authoritarian all my life and have always questioned ‘know-it-
alls.’ I also knew that I must have my blood examined. Yet this doctor’s self-
confidence paralyzed me. I simply did not know what to do. Not wanting to
bother other people unnecessarily, it did not occur to me to take a taxi and go to
an emergency ward. And the thought of calling an ambulance was even more
remote. After all, I had not had a heart attack.
Today, I fail to understand how I, of all people, could be so gullible. The rational
part of me was screaming that my doctor was wrong and bloody arrogant
besides. I still remember his name and am tempted to call him. But that will not
make either of us happier, so I guess not.
I cannot say this strongly enough: If you feel uncomfortable about your doctor’s
advice, do not accept it. Take care of yourself! It might mean the difference
between life and death. Do not feel afraid, like I was, of burdening the healthcare
system. Better to burden it once too many times than once too few. The
exception is if you suffer from anxiety of the type previously called neurosis.
Such people sometimes burden others all the time and they should be treated
with psychotherapy rather than subjected to all sorts of superfluous tests that will
only increase their anxiety.
When I tell this story to colleagues, a common reaction is that people do not
survive falciparum malaria and, therefore, I cannot have had that disease. But
that argument is faulty. Most children in Africa survive, and most European
explorers, missionaries and scientists recovered even before we had any
remedies for malaria. My colleagues’ reaction is an example how few nuances
doctors sometimes have in their thinking, which is another reason why patients
should follow their own hunches.
Several people living close to European airports who have never visited malaria
infested areas have had malaria. This happens because infected mosquitos are
brought to Europe by aircraft. Another possibility of getting malaria in Europe is
through blood transfusion. Malaria outbreaks have also occurred during warm
summer months in countries as far north as Finland. Yet the risk of contracting
malaria in Europe is, of course, extremely small.
Cases traced to stowaway mosquitos are rare, since the cabins of all aircraft
leaving malaria infested regions are routinely sprayed with insecticide. I once
flew out of Heathrow to India with Air India and was highly surprised to see a
steward spraying some foul-smelling aerosol. I asked to see the canister but there
was no declaration of the ingredients on it. I also asked why he sprayed the
entire cabin, which he said was to prevent insects from being brought to India.
That made a lot of sense to me and I told him about the few cases of malaria near
Heathrow. India has plenty of its own home-grown malaria so why import extra?
There were other amusing surprises on that flight to India. A Sikh steward asked
me if I was non-vegetarian. I responded that I was not a ‘non-thing;’ I was
human and ate most things. I told him I was not a non-female or a non-
homosexual, either. He did not appreciate my humour but a man sitting nearby
did. He was Indian and explained that many Indians, Hindus for instance, only
ate meat once a week or not at all, and that the Sikh steward’s question was
therefore reasonable.
Who is right? If you google statins, you will immediately find yourself in a
minefield. Five of the first nine entries are about side effects. That is highly
unusual for any drug and may suggest that the harms of statins have not been
taken seriously. Or is it just a sign that the war is ongoing, with fierce believers
on both sides - and a lack of rational thinking?
If you type statins cochrane, you will see controversy even here. As you write,
one option Google suggests is cochrane statins controversy. The first article
under that heading is Cholesterol Confusion and Statin Controversy. 1 The
authors argue that “the role of blood cholesterol levels in coronary heart disease
(CHD) and the true effect of cholesterol-lowering statin drugs are debatable. In
particular, whether statins actually decrease cardiac mortality and increase life
expectancy is controversial. Concurrently, the Mediterranean diet model has
been shown to prolong life and reduce the risk of diabetes, cancer, and CHD.”
Well, that message is controversial in itself. Although it has been argued for
decades that the benefit of statins might not have anything to do with their effect
on cholesterol, few people question whether statins lower mortality. I am not one
of them even though newer trials have been less convincing than old ones.
The controversy has mostly been about whether people with no pre-existing
cardiovascular disease should be treated with a statin. A Cochrane review on this
is available, 2 but when I see such reviews, my first question is always: Were
these people healthy like me, or were they not as healthy as they should be
considering the amount of discussion about primary prevention of a disease?
The Cochrane review is from 2013. When I wrote my book about the drug
industry in 2013, 3 and looked at the earlier version of the review from 2011, 4 I
noticed that the average age of the participants was 57 years and they were not
that healthy at the outset. Some trials only recruited patients with diabetes,
hypertension or increased lipids, and others additionally included patients with
previous cardiovascular disease. Further, the rate of smokers ranged from 10% to
44% in trials providing data on this issue. Finally, I always look up whether the
trials were funded by industry. If results are disappointing, trials may never be
published - or deaths and heart attacks may be deliberately removed from the
intervention group. 3 In fact, fraud has been demonstrated many times in industry
funded cardiovascular trials. 3
Only one of the trials providing data on all-cause mortality was publicly funded.
It seems to me, and this was even confirmed by the Cochrane review authors in
the discussion section, that the 16% reduction they found in all-cause mortality
was greatly exaggerated. For example, a large, publicly funded trial, the
ALLHAT-LLT trial - not included in the review because more than 10% of the
patients had pre-existing cardiovascular disease - did not find a reduction in
mortality: risk ratio 0.99 (95% confidence interval 0.89 to 1.11, meaning that we
are 95% certain that the true effect lies somewhere between an 11% reduction
and an 11% increase in total mortality).
No important differences exist between the 2011 and 2013 Cochrane reviews.
The number of trials and patients were 14 and 34,272 in 2011, and 18 and
56,934 in 2013. The reduction in all-cause mortality was 16% in 2011 and
almost the same in 2013 (odds ratio of 0.86, which is a 14-15% reduction). Yet
the authors’ conclusions had changed. In 2011, they wrote that some trials
included people with cardiovascular disease (CVD) and that, “Only limited
evidence showed that primary prevention with statins may be cost effective and
improve patient quality of life. Caution should be taken in prescribing statins for
primary prevention among people at low cardiovascular risk.” These caveats
were gone two years later when the authors merely wrote that statins reduced all-
cause mortality, major vascular events and revascularizations with no excess of
adverse events among people without evidence of CVD treated with statins.
Dr. John Abramson, one of the researchers who pointed out that statins are not
devoid of harms, published a paper in 2015 about the controversy. 5 The core
issue is the level of risk a person should have for acquiring cardiovascular
disease within the next five or ten years to justify a statin prescription. In 2014,
the NICE guidelines lowered the threshold of CVD risk for statin therapy in
healthy people from 20% down to a 10% ten-year risk. That was because of a
meta-analysis by the Cholesterol Treatment Trialists’ (CTT) Collaboration,
published in 2012. 6 This meta-analysis demonstrated a consistent reduction in
major vascular events, regardless of baseline risk, and claimed that men and
women, old and young, and people with or without CVD all appear to benefit,
and that the evidence resolved concerns about possible serious adverse effects
and potential sources of bias in the randomized trials highlighted in the 2011
Cochrane review.
The General Practitioners Committee of the British Medical Association was not
happy with the new NICE guidelines and stated they did not have confidence in
the recommendation to reduce the risk threshold for prescribing cholesterol
lowering drugs. 5 Abramson and colleagues published a paper in BMJ pointing
out that, for people with less than a 10% five-year CVD risk, there was no
significant reduction in overall mortality. 7 They also wrote about muscle
symptoms (and misinterpreted an epidemiological study which they later
corrected), diabetes, liver dysfunction, acute renal failure, cataracts, sexual
dysfunction and psychiatric symptoms.
Sir Rory Collins from the CTT claimed that really good data from over 100,000
people showed that statins are very well-tolerated and only have “one or two
well-documented [problematic] side effects.” However, a year later Collins
admitted that his research team had assessed the effects of statins on heart
disease and cancer, but not on other side effects such as muscle pain. 5
There are two major problems with the CCT meta-analysis. 5 First, the CCT had
agreed that the patient-level data upon which their analyses would be based,
“will be held in strict confidence.” Thus, the CTT, part of the Clinical Trial
Service Unit of Oxford which is heavily dependent on pharmaceutical company
money for its research, was given sole access to patient-level data without
allowing them to be reviewed by independent experts. Second, although
evaluation of the effectiveness of medicines must always be counterbalanced by
evidence of harm, the CTT’s arrangement to receive effectiveness data did not
have such balance. They only received adverse events data on cancer and the
reasons for ending study treatment. The frequency of adverse events cited in
their 2012 meta-analysis relied upon published reports rather than patient-level
data, and we know that harms in published reports of industry sponsored trials
are underreported to such an extent that these data are virtually useless 3 (see
also Chapter 8).
Abramson concludes by saying, “the bottom line is that practically all that we
think we know about the efficacy and safety of statins has been brought to us by
commercial interests that hold the actual data as proprietary secrets. Since we
cannot verify commercial claims of efficacy and safety, healthy people should
use statins with caution.” 5
I agree. Statins have many adverse effects and some of them are serious. 3 And
who knows how large the mortality benefit is, given that drug deaths are
regularly missing in industry trials? Even if we choose to believe in the industry
trials, the benefit is small, also in high-risk people. The Cochrane authors noted
that 2.8% of the trial participants died. 4 A 16% reduction from a rate of 2.8%
gives a rate of 2.35% and an NNT of 1/(2.8% - 2.35%) = 222. However, the
authors noted that some trials were stopped early when the benefit was large, and
that selective reporting of outcomes was common. This does not surprise me
given the many billions that can be earned by cheating. The 16% is undoubtedly
an overestimation.
What should we do, considering all this confusion? In my view, there are too
many uncertainties to allow us to draw firm conclusions, apart from the fact that
statins lower mortality in some people. I feel, however, this discussion should
not primarily be about effect sizes and the likelihood of benefits and harms
occurring. It should be more of a philosophical discussion. What is it we want
from our lives? People live vastly different lives and we should not generally
interfere with that. Some people are not at all concerned about some future
minor risk of dying from a cardiovascular event and even less interested in
taking a drug that might lower that risk minimally for the rest of their lives.
Some people try to climb Mount Everest even though 10% of them will die in
the attempt. Some people simply feel not taking pills that change many normal
body processes is one of life’s treasures. Finally, there are much better ways of
reducing cardiovascular risks than taking pills, exercising and avoiding
overweight, for example. Pills are poor substitutes for that.
Hypertension
In many ways, the issue with drugs for hypertension is similar to the statin issue.
3 They cause many harms, which patients may not notice or might believe is due
to advanced age. And the benefits for low-risk people is small. Furthermore,
exercising and not getting overweight are excellent remedies. Finally, many
people with a hypertension diagnosis are not hypertensive. Their blood pressure
just happened to be increased when they visited their doctor. Therefore, it is
prudent to try to taper off antihypertensive drugs, one by one, at regular intervals
and see how it goes.
When I was admitted to the same hospital four years earlier on suspicion of
coronary artery disease (see Chapter 4), I declined taking aspirin after having
studied the scientific literature. However, the second neurologist who saw me
thought I suffered from paroxystic atrial fibrillation and suggested I start taking
both aspirin and dabigatran (Pradaxa). I told him I refused to take Pradaxa
because no antidote existed if I were to start bleeding. An antidote does exist for
warfarin, a similar drug. I also said that I was a sportsman who sometimes
sprinted 100 m at full speed in the forest and drove as fast as I could on my
racing bike. Therefore, I was worried about being anticoagulated because I
might fall and get a head injury. He explained that Pradaxa was the drug
recommended in their guidelines and he tried convincing me to take it, arguing
that warfarin was a rat poison that would never have been approved by the drug
agencies if it had been discovered today.
I felt he must have listened too much to salespeople from Boehringer Ingelheim,
the maker of Pradaxa, or their paid allies among doctors, because this kind of
speech is exactly what you would expect from unreliable sources. I believe
warfarin would be approved if it were a new drug today. I told him Boehringer
Ingelheim had concealed their knowledge, that if the plasma concentration of
dabigatran was monitored, major bleedings could be reduced by 30-40%. I also
said it was a marketing trick to give doctors the false impression that, in contrast
to warfarin, no monitoring was needed for dabigatran. He was surprised by this
information which he did not know. I sent him a BMJ paper about these issues. 8
Even so, the neurologist tried quite hard to convince me that I should take
Pradaxa and referred to the department’s guidelines several times. I was
unmoved and called a colleague who is a specialist in internal medicine and
discussed it with him. He agreed entirely with me. Two of his patients on
Pradaxa had recently died.
When the neurologist looked at the notes from my admission with chest
discomfort, he found it was not certain that I had suffered from atrial fibrillation.
He then changed his opinion and told me I should take aspirin and clopidogrel
because of my TCI attack.
There was one more very important reason for my reluctance to use drugs. I
went through a good deal of the science during the 24 hours I spent at the
hospital, and I took into account that I have no risk factors whatsoever for TCI. I
suspected that many of the patients in the trials summarized in the Cochrane
reviews had various risk factors because it is in the drug industry’s interest to
include high-risk people so that they can show that their drugs work. Therefore,
the Cochrane results could not be extrapolated to me. I do not have diabetes, my
blood pressure was only 118/76 on admission, S-cholesterol was only 5.3
mmol/l. That was not even a fasting cholesterol, but one taken after my lunch.
Furthermore, during the 16 hours I was monitored, nothing appeared on the
ECG. I had sinus rhythm all the time.
My head was put into an MR scanner, but since the nurse did not tell me what to
expect, my experience was not a good one. I wanted to know how long it would
take but decided not to ask since it would change nothing. There were very loud
noises; sometimes my couch was shaking quite a bit; and there were long pauses
when I thought it was over. It lasted about 45 minutes and, even when it was
over, I laid there motionless, waiting for a very long time while the doctor
examined my results. It was very strange, like being in a futuristic fantasy world.
With my head still in the magnet, I shouted several times to find out when it
would end but received no reply, probably because the nurse was not in the
room. I was alone and had a small balloon near my hand which was some sort of
alarm I could press if I was in distress. I was, but not enough to press the
balloon.
Finally, the nurse came back. She declared that they needed to do more
recordings. I said that I had enough and that the doctor would need to make
decisions based on what I had already provided. I later asked about the purpose
of the MR scan and was told that, if there were signs of earlier thromboses such
as dead brain tissue, they would look more carefully for any possible source of
the blood clots.
The next day, I requested to leave the department because I was feeling well and
needed to give an important lecture in another town (which was filmed and has
been seen by over 60,000 people on YouTube: https://www.youtube.com/watch?
v=i1LQiow_ZIQ&t=176s ). That was agreed even though it was routine to
monitor people’s hearts for 48 hours. I came back two days later for an
ultrasound of my neck arteries despite boldly telling the staff that they would not
find the slightest sign of arteriosclerosis in them, because I had coronaries like a
medical student. As expected, they were smooth.
I felt the scientific literature was not of much use because my risk of getting
another TCI must be very low compared to other patients with all their risk
factors. Therefore, I told the chief physician that I was not interested in taking
pills. We negotiated a little and I agreed to take baby aspirin for three months. I
was not even keen to do that. After my loading dose of aspirin at the department,
I did not take the drug for more than a week (the effect lasts very long) and then
divided 500 mg pills with a kitchen knife and took a small dose each day.
Just prior to leaving the hospital, a nurse came into the common room where I
worked on my computer and measured the blood pressure on my left arm. It was
only 99/70. I told her that could not be correct because I never had a blood
pressure below 100. I suggested she tried my right arm. There, the blood
pressure was 130/100. Neither could that be true, but the nurse explained that
blood pressures were not necessarily the same in both arms. “I know,” I said,
“but there should not be such a great difference.” I therefore asked to have my
blood pressure measured on both arms in my hospital bed where they had a
stationary apparatus. As expected, it was now the same in both arms and similar
to the low level when I arrived. I then told the nurse that they needed to get a
technician to check the mobile apparatus. You see how easy it is to get a
diagnosis of hypertension that is wrong!
There were other surprises during my 24-hour hospital visit. The first neurologist
told me a little about my blood tests and said that my homocysteine was
approximately 20 and that the upper normal range only went until about 15. She
deduced that I might suffer from vitamin B12 deficiency. I told her not to worry,
that I was fine and that I could just put the matter out of my mind. She seemed to
agree with that. After I had come home, I looked up homocysteine on Google
and found out that it is a fairly unreliable indicator of anything and, therefore,
should not be used as an indication that something is wrong.
The first neurologist also told me that my MR scan had shown an aneurysm, but
she added that this was quite common. I fully realize that you cannot be
subjected to an MR scan without a considerable risk of finding something fairly
normal. So that did not worry me. But it did worry me a little that the nurse who
scanned my two neck arteries told me that one of them was quite twisted. Does
that not predispose to blood clots in the arteries - mechanical obstructions
causing turbulence? I am quite sure I learned that during my medical studies. But
if true, I can do nothing about it. Why inform me about something I cannot act
upon and which can, therefore, only be harmful because it might make me
anxious?
No one really knew what was wrong with me. It was just like when I was
admitted with angina four years earlier. Amaurosis fugax can be caused by an
embolus, but according to the chief physician, the fact that printed letters were
disappearing suggested a thrombus in the occipital lobe. How strange that I am
such an odd person who continually defies the textbooks.
I left the hospital and put it all behind me. I hate being a patient because we lose
our autonomy and become dependent on others. I made the refrain in the song by
Brian McFerrin, Don’t Worry, Be Happy, my motto. It can be heard on YouTube
in a fabulously funny video.
The staff at the Department of Neurology was excellent and very kind. I met no
less than three nurses with the same name: Julie. That could be rather practical if
you were demented.
I did look up the recurrence risk, of course. The risk of a TCI recurring is about
7%. A year later, I developed diplopia that lasted half an hour, and two years
after that I suddenly got blurred vision which lasted for about five minutes.
Perhaps I should take aspirin again. But I still doubt it and whenever you are in
doubt about drugs: Don’t take them. "Don’t worry, be happy." We all die
anyhow.
7 More on screening
Screening healthy people for disease is very appealing. It seems the right thing to
do. “Prevention is better than cure.” “Catch it early.” “If you haven’t had a
mammogram, you need more than your breasts examined,” as the American
Cancer Society put it in their advertisements. 1 Deals for Mother’s Day
mammograms as presents have been offered, and the American Cancer Society
claimed that 80% of women aged 35–50 years were at high risk and they
recommended getting baseline mammograms. That made one epidemiologist
state that claiming more than half to be at high risk was mathematically absurd. 2
Philosophically, we might ask: Does screening really reveal diseases? Not really.
The person was healthy and is still healthy. If for example, screening detects
prostate cancer, the person is still healthy, even in relation to his prostate. We
know from autopsy studies that most old males have cancer in their prostates,
and as stated earlier, the prevalence corresponds somewhat to the age:
approximately 60% of 60-year-old men have prostate cancer. 3 Yet only 3% die
from it. Hence the saying that most men die with prostate cancer and not from it.
Thyroid cancer
Autopsy studies have shown that probably every single person above a certain
age has cancer in their thyroids, yet dying from it is extremely rare. 2,3 In the
United States, only 0.08% die from thyroid cancer, nonetheless, busybodies like
the Light of Life Foundation recommend that Americans get screened for
thyroid cancer with the slogan “Check Your Neck.” On the front page of the
foundation’s website, an audio recording is available of rock star Rod Stewart
who has been treated for thyroid cancer. He says that it is the fastest growing of
all cancers in men - a cancer that can strike anyone at any time - luckily, early
detection saved his voice and his life. He tells you to ask your doctor to “check
your neck” for thyroid cancer. Beneath Stewart’s propaganda are black-and-
white photos of Cindy Crawford and Brooke Shields with no explanation why -
but they sure do look pretty.
The website explains that thyroid cancer “can occur in anyone,” which is kind of
amusing because thyroid cancer does occur in everyone. You can donate - of
course. The list of sponsors and partners include Eisai, Bayer, Veracyte, Sanofi,
Shire, and Interpace Diagnostics.
This website is typical for disease awareness campaigns. In the United States,
most body parts have advocacy groups, politicians or athletes with public
awareness campaigns and slogans promoting screening. 5 The Light of Life
Foundation provides celebrity endorsement and frightening information, yet no
mention of what the screening harms are, or whether screening saves lives.
There is no statistical information whatsoever on the website and no references
to any evidence. If there was, it would be easy to see that screening for thyroid
cancer is an exceptionally bad idea.
In Korea, the National Cancer Centre and many university affiliated hospitals
recommend ultrasound to screen healthy people for thyroid cancer. 6 That has
created an epidemic of pseudo-disease. The incidence of thyroid cancer has
increased every year by about 25%, making this the most common type of
cancer in Korea with 15 times more new cases per year than in the UK.
However, mortality rates have remained almost constant during the past 30
years. It is therefore clear that this man-made epidemic of cancers has been very
harmful for the population. More than 90% of people diagnosed with this cancer
had their thyroids surgically removed. Ironically, despite Rod Stewart’s
propaganda, one of the harms of thyroid surgery is losing your voice! There are
many other surgical harms that you can look up on Google.
Not well understood, it is a fact of nature that cancer can be detected in every
one of us from when we are quite young - if only we would get screened enough.
However, most cancers we harbour in our bodies are totally harmless. They
either disappear again spontaneously or grow so slowly that they will not cause
any symptoms before we die from other causes. 2,3
Therefore, with rare exceptions, the only thing screening can accomplish here
and now - if it leads to a positive finding - is to harm people. No longer healthy
in a psychological sense, getting diagnoses negatively effects people.
In that way, screening is similar to drugs. What we primarily know about both is
they can harm you. We should therefore only go to screening or take drugs if
these interventions do more good than harm on average, and we should require
reliable randomized trials demonstrating exactly that. This is rarely the case,
even for mammography screening.
Mammography screening
Mammography screening is highly controversial. But if you google
mammography screening, you will end up thinking it is a good idea to get
screened - a testimony about the effectiveness of propaganda. The initial Google
page has 12 entries. One is called, What You Need to Know, two are actually the
same one from the US National Cancer Institute, two are from radiologists, two
from the American Cancer Society, one from Web MD, one from Cancer
Research UK, one from the UK National Health Service, and one from the
WHO.
What about the other 11 entries? Websites with headlines like What You Need to
Know are usually untrustworthy, but in this case, that was not so: “In November
2011 the Canadian Task Force on Preventive Health Care (CTFPHC) released
their updated guidelines on breast cancer screening. The reaction was a heated
debate on the pros and cons of mammography, with doctors, breast cancer
organizations and breast cancer survivors on both sides of the issue. Rethink
Breast Cancer agrees with the CTFPHC guidelines and we’re here to help clear
up any confusion around the question: Should I get screened?” That looks much
better than I thought. No data are presented but the guideline recommendations
are quoted:
There it was again. They love this in North America: Talk to your doctor.
Passing the buck to your doctor instead of saying no, which is plain, simple and
correct. When these guidelines were about to be published in the Canadian
Medical Association Journal, the journal’s editor asked me if I could write an
accompanying editorial. I accepted and the title was: Time to stop mammography
screening? 8 I found these guidelines more balanced and more in accordance
with the evidence than any previous recommendations I had seen, but I
suggested that it might be best to avoid screening altogether, at any age, and
explained why. Four years later, I pulled no punches and published the article,
Mammography screening is harmful and should be abandoned. 9
The US National Cancer Institute gets it wrong in its headline: What are the
benefits and potential harms of screening mammograms?
The error is a big one, yet we see it all the time. Instead of emphasizing the
benefits and potential harms, it would have been more correct to emphasize the
potential benefits and harms. Since all screenings have harms, it is wrong to talk
about potential harms, yet it would be correct to talk about potential benefits,
because these benefits, e.g. a reduction in mortality for a certain type of cancer,
might never materialize. Aside from the headline, the information was quite
good, but no numbers were provided, and we need numbers to make evidence-
based decisions.
Inside Radiology from Australia does not ask whether one should attend
mammography screening but assumes that is the case: Why Would My Doctor
Recommend Screening Mammography? Typical of doctors who earn their livings
from offering diagnostic tests, the numerical information is flawed. The text
refers to an independent breast screening review from the UK which found that
three times as many women will be overdiagnosed as those who avoid dying
from breast cancer. 10 But it then adds, “another estimate is that for every case of
overdiagnosis, 2 to 2½ lives are saved by mammographic screening.” The
website does not inform us that this estimate was provided by people who are
not independent and have huge conflicts of interest when it comes to
mammography screening. 11 Nor does it state that the estimate is grossly
unreliable - which we have demonstrated. 12 The authors’ benefit-to-harm ratio is
20-25 times more favourable than the estimate we published in the Cochrane
review of the randomized trials. 13 Our Cochrane review is the most thorough
review of the screening trials ever made. It has been peer reviewed countless
times, also after we first published it in 2001, and even by people in favour of
mammography screening.
The paper got it wrong by a factor of 20-25 times - a kind of a record – and is
horrible to a point far beyond what most researchers would think possible. The
data are cherry-picked; some of the numbers are demonstrably false and in stark
contrast to official data; the authors contradict themselves; and an unclear
method is described in a footnote. Further, the graphs are seriously misleading
and hide important data telling a very different story, and the authors adjusted for
a trend that does not exist. No conflicts of interest were declared even though
one of the authors founded Mammography Education Inc. in Arizona in 1980
and declared a 5 million SEK income in Sweden in 1999, which is an
extraordinary amount by Nordic standards. Their article 11 is a masterpiece in
manipulation. 12 I cannot remember having seen anything in the medical research
literature as bad as this article. Of course, it was published in the preferred
journal of screening zealots - Journal of Medical Screening. If you are interested
in screening, do not read anything in that journal.
The other radiology website is the Radiological Society of North America. It
talks about benefits and risks of mammography screening, which is wrong. A
risk means that something might happen but not necessarily that it will happen.
However, screening always leads to harm, which is worse than just at risk. Under
risks, only radiation exposure from the mammogram and false positive findings
are listed. Although the website is from 2017, nothing is stated about the most
important harms, overdiagnosis and overtreatment, even though unnecessary
treatment of healthy women kills some of them. The website is called
RadiologyInfo.org - For Patients. In my country, it is against the law to inform
patients about medical issues without informing them of the most serious harm
of the recommended intervention. Furthermore, healthy women attending
screening are not patients, they are citizens.
The American Cancer Society has always very aggressively promoted cancer
screening and very often in disregard for the facts. 2 They say that women ages
40 to 44 should have the choice to start annual breast cancer screening with
mammograms if they wish to do so. Women age 45 to 54 should get
mammograms every year. Women 55 and older should switch to mammograms
every 2 years or can continue yearly screening. Screenings should continue for
as long as women are in good health and expected to live 10 more years. They
also say that some women should be screened with MRI’s along with
mammograms.
“Talk with a health care provider about your risk for breast cancer and the best
screening plan for you.” Talk to your doctor who might give misleading advice
in order to earn more money.
The other website from the American Cancer Society is called Mammogram
Basics. “Why do I need mammograms…” leaves no doubt that you should be
screened. No talking to your doctor here. We are told that mammograms are
safe, but I could find no information about the benefits and harms of screening.
That is less than “basics.”
The WebMD from 2015 notes that conflicting guidelines from leading medical
groups have made this issue murkier than ever and says that the main expert you
need to check with is your doctor. That does not make sense. When people
disagree so much, what you are told will vary greatly from doctor to doctor.
“Talk to your doctor” just passes the buck again.
WebMD is honest about the harms - very few websites are. It says that false
positives and overdiagnosis are the biggest concerns: “Getting called back for
another mammogram or a biopsy can be stressful. In one survey, 40% of women
who had this happen described it as ‘very scary’ or ‘the scariest time of my life’
… some women may get surgery, radiation, and chemotherapy that they don't
really need because doctors want to be cautious.” It also says, “One study found
that as many as 10 women might be getting overdiagnosed for one death
avoided.” That is what we reported in our Cochrane review.
Cancer Research UK reports the benefits and harms of screening, and notes that,
for every life saved from breast cancer by screening, around three women are
overdiagnosed. 10
I have many reservations about the statements in this WHO report. For example,
it notes that women requiring further investigations after screening experience
significant short-term anxiety. It is not short-term. Even after three years, women
who had experienced false-positive diagnoses had an anxiety level and other
psychological problems that was somewhere between that of women with breast
cancer and women told they did not have cancer. 14 The report dismisses total
mortality with the arguments that breast cancer mortality constitutes very little of
this and that the follow-up in the trials was only 11 years. However, breast
cancer mortality is the wrong outcome 9 not only because it is biased in favour of
screening but also because the treatment of overdiagnosed, healthy women
increases their risk of dying. 9 Radiotherapy may cause deaths from heart disease
and lung cancer, and these iatrogenic deaths (deaths caused by doctors) are not
counted as breast cancer deaths. If we take into account the deaths caused by
radiotherapy and rather generously assume that screening reduces breast cancer
mortality by 20% and results in only 20% overdiagnosis, in accordance with the
Independent UK report, there appears to be no mortality benefit. 15 That result
can be discussed. Modern radiotherapy, for instance, may be less harmful, but
considering that screening does not reduce the rate of advanced cancers and
therefore cannot work, 9,16 -18 it seems likely that screening increases total
mortality.
I have gone into this in some detail because I wanted to show how confusing it
all is if you do not do what you have learned in previous chapters. We started out
by googling mammography screening which led to a lot of misinformation. The
best thing to do is to google mammography screening cochrane, which brings up
the Cochrane review as the first entry. The second entry is a mammography
screening leaflet we wrote in 2008 and updated in 2012, because the information
offered to women in all countries was very one-sided, biased and often just plain
wrong. 19 -23 Our leaflet became so popular that volunteers in 16 countries
translated it into their own languages, including Chinese, Russian and Arabic. 24
Should you get screened with mammography? No. Given the uncertain benefit
and the certain harms, an overall utility analysis cannot turn out positive, i.e., in
favour of mammography screening. 9 Accordingly, well-argued articles from
independent researchers now call for a stop to screening. 25
You might conclude that the number needed to treat to harm one person is one,
because 199 of 200 rounds up to one.
It is easy to calculate the NNT if you know the relative risk and the absolute risk
of experiencing an untoward event. Here are the results of a randomized trial
with 200 patients in each group, which corresponds to the numbers just
mentioned:
Drug Placebo
Died 3 4
In order to save one life, we need to treat 200 people with the drug, because the
number of deaths drops from 4 to 3. More formally, it goes like this:
The risk ratio (also called the relative risk) means the risk on drug divided by the
risk on placebo: 1.5%/2% = 0.75. That is what you typically find in journal
articles and other types of propaganda. The risk decreased by 1 - 0.75 = 0.25 or
25%. A risk ratio of 1 means that the drug is no better than placebo.
What if the only thing you know is that the risk is decreased by 25%, e.g. if a
leaflet from a national board of health encourages you to get screened for breast
cancer claiming a 25% reduction in the risk of dying from breast cancer? 2 You
can work out what the NNT is by looking up official statistics on the mortality of
breast cancer. Googling risk of dying from cancer produces statistics from the
American Cancer Society as the first entry where the deaths from many types of
cancer are listed. The lifetime risk of dying from breast cancer is 2.7%. You
might wonder why there is so much fuss about this cancer - pink ribbons, runs
for the cure and a breast cancer awareness month - when so few die from it. I too
wonder. Ten times as many women die from cardiovascular causes, so running is
fine, just not for some non-existing cure for breast cancer, and you should run
more than once a year. You should run every day to lower your risk of dying
from cardiovascular causes.
If it were true that breast screening reduced breast cancer mortality by 25%, you
could calculate the NNT for screening. We will increase the death risk a little
because most US women go to mammography screening. So, let us say that the
2.7% already represents a 25% reduction, which means that the risk without
screening is 2.7%/0.75 = 3.6%. Since the risk difference is 3.6% - 2.7% = 0.9%,
the NNT becomes 1/0.009 = 111. Thus, if 111 women attend mammography
screening regularly for many years, one will be spared dying from breast cancer.
But that is not true and neither did we take into account the fact that screening
increases mortality from other causes. No overall mortality benefit from
attending screening has ever been demonstrated and it is not likely it even exists.
Health checks
Health checks, called annual physicals in the US, are similar to regular car
checks. What they have in common is that a lot of issues are detected that should
remain untreated. This results in huge bills for our old cars, because parts are
replaced that do not need replacement. In the hands of the mechanics, we are
chanceless. A friend of mine never had his car checked - he had it repaired when
there was a problem and saved a huge amount of money over the years. Very
little is needed to keep a car running. Of course, checking the brake disc
thicknesses once in a while is wise, and a few other things, too. Yet I do wonder
why we all do not follow his smart example, even when it comes to our health.
Health check cochrane on Google reveals our Cochrane review as the first entry.
26 When we embarked on this review, also published in 2012 in the BMJ,27 we
did not expect to find much. We were highly surprised to find 14 trials with
relevant outcome data that had compared health checks with no health checks in
adults unselected for disease or risk factors. We updated the review in 2019 and
added one more trial. 26 The median follow-up was long - ten years - and there
were 21,535 deaths, which is a lot. Health checks did not lower total mortality,
risk ratio 1.00 (95% confidence interval 0.97 to 1.03), cardiovascular mortality,
risk ratio 1.01 (0.92 to 1.12), or cancer mortality, risk ratio 1.05 (0.94 to 1.16).
We did not find an effect on clinical events or other measures of morbidity, but
one trial found an increased occurrence of hypertension and
hypercholesterolaemia with screening and another trial found an increased
occurrence of self-reported chronic disease. A third trial found a 20% increase in
the total number of new diagnoses per participant over six years compared to the
control group. No trials compared the total number of prescriptions, but two of
four trials found an increased number of people using antihypertensive drugs.
Two of four trials found small beneficial effects on self-reported health, but that
could be due to reporting bias since the trials were not blinded. We did not find
an effect on admission to hospital, disability, worry, additional visits to the
physician, or absence from work, yet most of these outcomes were poorly
studied. No useful data were reported on the number of referrals to specialists,
the number of follow-up tests after positive screening results, or the amount of
surgery. Important harms, such as the number of follow-up diagnostic
procedures or psychological effects, were often not studied or reported and many
trials had methodological problems. Unfortunately, doctors forget to register
even the most obvious harms all too often.
Adding all this up, we concluded that general health checks are unlikely to be
beneficial. That had political consequences. In 2007, the Danish Association of
the Pharmaceutical Industry lobbied politicians in the Danish parliament and
convinced some of them that regular health checks were good for preventing
diseases. When asked by a journalist whether the issue was more about selling
more drugs, the industry spokesman - in a rare bout of honesty - admitted that
was the case. 28 In 2011, our new government had regular health checks on the
menu, but I told the minister of health that our Cochrane review, just completed,
found no effect on mortality. I had invited one of my colleagues to the meeting
and he informed the minister about a large Danish trial he just finished which
also failed to find an effect. 29
We told the minister that health checks are probably harmful, leading to more
diagnoses, more drugs, more adverse effects and even psychological problems
because people are told they are less healthy than they think. The minister
aborted her plans on the spot and said it was the first time the new government
had broken a pre-election promise in an evidence-based manner.
When people draw conclusions before doing their homework, terrible mistakes
can be made. Infamous statistician Bjørn Lomborg denied the existence of
climate change in his book, The Skeptical Environmentalist, arriving at this
conclusion by being highly selective with the references. He arranged the
Copenhagen Consensus Conference in 2011 where three health economists
concluded that health checks would give most health for the investment, namely
26 Danish crowns for every crown invested. 30 Quite an impressive gain for
something that does not work.
Once something has been introduced, it is very difficult to stop. The reactions in
the UK to our review were - with British understatement - “interesting.” The
NHS’s universal health check for everyone between 40 and 74 years of age
tested for cardiovascular disease, diabetes and chronic kidney disease and,
“Evidence confirms that this is both clinically and cost-effective.” In a slide
show, we were told that yearly health checks would prevent at least 9,500 heart
attacks and strokes, 2,000 deaths and 4,000 people from developing diabetes,
thus preventing significant illness and premature death. That message showed a
graveyard at sunset with two crosses in the background, so no one could miss the
gravity of what would happen if they did not attend health checks.
Our Cochrane review came out in October 2012, and the same month, a
Department of Health representative told BBC News: "By spotting people who
are at risk of heart attacks, diabetes, stroke and kidney disease, we can help
prevent them. The NHS Health Check program is based on expert guidance.”
I see. The NHS Health Check program was based on evidence until our review
came out and showed what the evidence was.
One year later, after we had had enough of all the official UK nonsense in
defense of their indefensible health check programs, we published a letter in The
Times, which resulted in a front page interview: NHS checks on over-forties
condemned as useless. It covered almost half a page – and ran next to an equally
large photo of Prince William, his wife, child and a royal dog. Elsewhere, Public
Health England announced that an expert panel was to be established to review
the effectiveness and value-for-money of NHS Health Checks, and that was
because of repeated calls for the program to be scrapped because evidence
indicated checks wasted time and money. According to the Daily Mail, ministers
insisted that 650 lives a year could be saved - a sharp retreat from the previous
claim of saving 2,000 lives. Barbara Young, chief executive of Diabetes UK,
also supported routine checks saying they could uncover an estimated 850,000
people with undiagnosed type 2 diabetes. What a funny remark: labeling
hundreds of thousands of healthy people as diseased has no value in itself. We
must find out whether screening for diabetes is helpful. Our review found that it
is not. Several of the trials had done just that: screened for diabetes.
This soap opera got so bizarre that I decided to participate in the amusements. 32
For the headline, I used a quote from episode four of Yes, Minister, a BBC series,
where Minister of Administrative Affairs Jim Hacker says to Sir Humphrey, "I
don't want the truth, I want something I can tell Parliament!" My statement in the
BMJ was:
Public Health England will establish an expert panel to review the effectiveness
and value for money of NHS Health Checks, and it will refresh the economic
modelling behind the programme. We are furthermore told that "although we
recognise that the programme is not supported by direct randomised controlled
trial evidence, there is nonetheless an urgent need to tackle the growing burden
of disease which is associated with lifestyle behaviours and choices."
The truth, that health checks don't work and are likely harmful, is too much to
bear for Public Health England, it seems. An expert panel is the modern version
of the Oracle in Delphi, and statistical modelling is like whispering in a wizard's
ear which result you would like to hear. Saying that there is an urgent need to
tackle the growing burden of disease as an excuse for going against clear
evidence from randomised trials reminds me of another episode of "Yes,
Minister" where it was skilfully argued why a huge number of administrators
were needed for a hospital that had no patients.
We have seen this before. The Marmot report on mammography screening was
also a "Yes, Minister" episode. Even using the too optimistic estimates from the
report, Mike Baum showed that radiotherapy kills at least as many healthy
overdiagnosed women as the report estimates screening saves. Like health
checks, mammography screening is harmful, but such trifles don't affect the
leaders of the NHS or the UK Government.
When life gets too absurd, I revert to my "Yes, Minister" DVDs and get a big
laugh. It is probably more healthy than crying.
A month after this, we published a letter in the BMJ about the lack of fair play. 33
The NHS Diabetes and Kidney Care and the Department of Health had
published an eBulletin, Response to the Cochrane review, on the NHS Health
Check program’s website, which appeared to be serious criticism of our work. It
was, however, unfounded and misleading. We sent our response to the eBulletin
to the national director of the NHS Diabetes and Kidney Care and asked for it to
be published on the website. That was declined. The letter we got from the NHS
stated that the government had already decided that “NHS Health Checks will be
carried out as a national priority;” that the website is not a forum for debating the
merits of such checks; and that “there are other more appropriate places to
discuss government policy.” Total censorship in a country that calls itself
democratic. We also wondered, since the website was not a forum for debating
the merits of health checks, why the NHS had done exactly that, yet still denied
us the opportunity to respond? Why did the NHS program not publish its
criticism in BMJ, where we had published our review, so we could respond to it?
The answer is obvious: the NHS would lose that battle.
The absolute low point in all this came five months later when NICE (National
Institute for Health and Care Excellence), supposedly an independent institution,
prostituted itself to the NHS. 34 It issued a press release stating:
“Helping local authorities to encourage people to attend NHS Health Checks and
support them in making changes needed to improve their health, is the focus of a
new NICE briefing published today … This new publication is part of a suite of
briefing documents which NICE is producing to provide support to local
government … while also providing the best value for money … A report from
Public Health England found that checking blood pressure, cholesterol, weight
and lifestyle of people in this age group could identify problems earlier and
prevent 650 deaths, 1600 heart attacks and 4000 diagnoses of diabetes a year …
NICE recognizes the debate on the effectiveness of health checks that is ongoing
at the time of publication of this briefing. The NHS Health Check programme is
currently part of the health delivery infrastructure in England, so NICE seeks to
support its effective delivery.”
Prevent 4000 diagnoses of diabetes a year? Diabetes UK had just claimed that
routine checks could uncover an estimated 850,000 people with undiagnosed
type 2 diabetes. How does that add up? Are we supposed to find 850,000 or
supposed to avoid finding 4000?
Two days after the NICE press release, one of my UK colleagues called this
“Stalinism in the NHS” and referred to a new article. 35 Members of parliament
were not as gullible as NICE, but singled out NHS Health Checks as a cause for
concern in a highly critical Health Committee report on Public Health England,
which included reports that health professionals were being leaned on to refrain
from criticizing the project in public. Regarding the poor uptake (only about
50% attended), Public Health England said it was aiming to drive the acceptance
rate up to 70-75%. That seems to me to come close to forced treatment - which
we otherwise only know from psychiatry - now in the form of health checks.
What happened to informed consent and individual freedom? I think the general
public is wiser than Public Health England.
If you do not google health check cochrane but just health check or annual
physical, it will be more difficult to find our Cochrane review. Based on our
experience with Public Health England, and on the fact that annual physicals in
the United States are so common that it has the same ritual status as having your
teeth checked, I would expect to find very misleading websites with such short
search terms - but I was pleasantly surprised. When our review came out, people
paid attention to it and media interest was colossal. Many of the websites, even
in the US, questioned health checks. Yet that was not what we found when we
researched Danish websites in 2009. 36 We found 36 different tests on 56
websites about health checks. Twenty-one tests were offered on at least 10% of
the websites and 17 of these 21 tests were unsupported by evidence, or evidence
against them for screening purposes existed. None of the websites mentioned
any harms and they presented a median of only one of the 15 information items
recommended when screening healthy people by the World Health Organization
and the Danish National Board of Health.
Screening for other diseases
Most screenings recommended by our authorities or patient organizations do
more harm than good. Only three programs easily come to mind for being
reasonable. Screening for phenylketonuria in new-born infants can prevent
development of serious mental retardation and much more through a special diet.
Screening for cervical cancer finds precursors to cancer, and can therefore
prevent cancers from occurring. Screening for colon cancer by testing for blood
in the stools or by looking for polyps and cancers with sigmoidoscopy can
prevent deaths from colon cancer and does not lead to overdiagnosis - by finding
polyps, it actually lowers the incidence of colon cancer. 37
I have lost respect for several organizations which, until recently, enjoyed high
repute. As just noted, NICE helps Public Health England provide regular health
checks even though they do not work and are probably harmful. The National
Health Service in the UK recommends screening for dementia, even though it is
very likely harmful. 38
The profession’s organized denial of the facts about how poor and dangerous
psychiatric drugs are has been successful in convincing people that these drugs
are good for them. 1 However, it seems to only be a matter of time before this
house of cards gets blown away. Opposition to the status quo - even among the
psychiatrists themselves - is growing.
I know many excellent psychiatrists and one of the best is Peter Breggin. He has
taught at several universities and has had a private practice in Ithaca, New York
for 50 years. Although not being in an academic environment, he has written
numerous scientific papers and some 20 books which are very instructive and
well referenced. Breggin does not use psychiatric drugs except in rare cases
where the abstinence symptoms during withdrawal from psychiatric drugs
instituted by some of his colleagues become unendurable for his patients. Then
he might prescribe a drug for a short period to slow withdrawal. Breggin once
said publicly that, if you have a psychiatric problem, you should not consult a
psychiatrist – it is too dangerous. Before I started reading some of his books, I
had already come to the conclusion that the use of psychiatric drugs does far
more harm than good. Nonetheless, I have been greatly inspired by him. My
conclusions are:
You should not take psychiatric drugs. The only exception I can imagine is a
seriously disturbed acute situation but, even here, the drugs should quickly be
tapered off. If you are already on one or more psychiatric drugs, you should
seriously consider getting off them.
Warning! Psychiatric drugs are addictive. They should not be stopped abruptly
because abstinence reactions may consist of severe emotional and physical
symptoms that can be dangerous.
Many patients want to come off their psychiatric drugs and they would also get
better if they did. Unfortunately, most doctors don’t know how to do it in the
best and safest way, and official guidelines are either totally reticent about this or
recommend tapering far too quickly. That is a remarkable omission considering
that, in Western countries, about 5% of the entire population has become
dependent on psychiatric drugs. 1 It is therefore appropriate that science
journalist Robert Whitaker called one of his books, Anatomy of an epidemic. 2
The psychiatric drug epidemic is likely the most harmful epidemic we currently
have - even worse than our obesity epidemic since obesity does not change our
brains.
Withdrawal may take months or even years and some patients continue to have
abstinence symptoms long after they came off the drugs; or they can suddenly
reappear after a symptom-free period, e.g. if the patient is stressed. In the worst
cases, the patients cannot come off the drugs because they have developed
permanent brain damage, which means they are bound to life-long treatment
with the drugs even though they will cause further harm.
I could have called this chapter Mental health, Mental disorders or Psychiatry,
but Emotional pain is the best term I can come up with. Emotional pain is central
to suffering from mental health issues and it is part of the definition of most
psychiatric disorders. There are exceptions, of course. A manic person may not
feel his emotional pain but will likely feel its return when the mania passes and
he realizes what he has done and gets embarrassed.
If someone has physical pain, e.g. because of a broken leg, we can test whether
aspirin has an effect on the pain compared to placebo. This is the case, but it
does not make us believe we fixed the problem. What is needed is not aspirin,
but surgery and a plaster cast. As another example, we do not treat headaches for
years on end, without trying to find out what causes the headache, e.g. a brain
tumour. But we do treat mental health issues with drugs for years, usually
without bothering to find out what traumas might have caused them.
The most prescribed type of psychiatric drugs - depression pills - illustrate this.
Effects are measured on a depression scale composed of various symptoms
scored according to their severity and the points are tallied for a total score.
After a few weeks, that score drops slightly more on drug than on placebo 1,3 and
the psychiatrists declare the drug has worked.
At best, psychiatric drugs can provide some relief in acute situations. I have
done research into psychiatric drugs for ten years and currently have five PhD
students studying their effects, yet I have been unable to find reliable evidence
that any psychiatric drug can cure any psychiatric disorder. These drugs have
purely symptomatic effects, and therefore, it is immensely misleading to call
them antidepressants or antipsychotics suggesting they are similarly curative as
antibiotics for infections, or to compare the effects of the drugs with the effects
of insulin on diabetes, which psychiatrists commonly do, unfortunately. 1
Long-term, psychiatric drugs are very harmful. 1-6 All of them impair higher
brain functions, which is what makes us human – our abilities to think, feel,
function, remember, love, have empathy, and care for ourselves and others. All
of these drugs can probably cause permanent brain damage and even addiction
leading to horrible abstinence symptoms when patients try to get off them.
Almost all increase mortality 1 and may cripple people physically and mentally.
And they can cause the same disorders they were supposed to alleviate, plus
additional disorders. Very few other drugs or substances can cause such terrible
harms - except opiates and street drugs. The main difference between street
drugs and psychiatric drugs is that the latter require prescriptions from doctors.
However, after buying these drugs at pharmacies, people sometimes sell them on
the street.
When patients get worse from the drugs they take - some of which are forced
upon them - psychiatrists rarely realize the drugs caused the harms. They believe
the disorder is worsening or that an additional disorder is presenting itself. So,
they increase the dosage (to no avail, as raising dosages increases harms, not
benefits) or they add even more psychiatric drugs. Doctors also routinely
increase dosages after a few weeks if the patients are not complaining too much
about harms. It almost never increases the effect - only the harms - and the risk
of dying.
That is the main reason why so many patients become chronically incapacitated,
with many diagnoses and drugs. In the end, no one can remember how it all
started and what life was like before the patient became ill. Patients are being
turned into chemically induced, artificial products with brains and personalities
that are no longer the same. And drug use has become part of their identity - just
as it is for drug addicts.
I call this the fly paper of psychiatry. The more noise the patients generate by
flapping their ‘wings,’ the more stuck in more diagnoses and drugs they become.
Many patients learn to refrain from telling their psychiatrist certain things to
avoid even more drugging than they are already experiencing.
That means if your interest is psychiatric drugs, you cannot use what I have
taught you in previous chapters. Hundreds of Cochrane reviews of psychiatric
drugs are available, but they are generally misleading because the trials included
are unreliable. Even when Cochrane authors try hard to do their best, they cannot
make up for all the flaws in the published trial literature. 1 We have done reviews
of depression pills based on 64,381 pages of clinical study reports we obtained
from drug regulators, and we were able to demonstrate harms omitted in the
published literature (see below). It is - really - very depressing. Psychiatric drug
trials are huge wastes and they abuse patients’ altruistic willingness to contribute
to clinical research.
Since I have written a book about this, 1 I will only repeat some of the most
important issues here and add a few more.
In the vast majority of psychiatric drug trials, the patients were already on a drug
similar to the one being tested against placebo. After a short wash-out period -
typically one week - the patients were randomized to the new drug or placebo.
The cold turkey that some patients in the placebo group go through in this type
of trial harms them, and therefore, it is no great wonder that new drugs
outperform the placebo in patients who have been harmed. Introducing longer
wash-out periods does not help much. If people were permanently brain
damaged before entering the trials, wash-out periods cannot compensate, and
even if they were not, they could suffer from abstinence symptoms for months or
years. 4,6
Thousands of trials of neuroleptics have been carried out, but when we recently
searched for placebo-controlled trials in psychosis that only included patients
who had not received such a drug earlier, we only found one trial. 9 It was from
China, and since it appeared to be fraudulent, we could not use it. Thus, all
placebo-controlled, randomized trials of neuroleptic drugs in patients with
schizophrenia spectrum disorders are biased, which means that the use of
neuroleptic drugs cannot be justified based on the evidence we currently have. 1
To find out how long patients need to continue taking their drugs, so-called
maintenance (withdrawal) studies have been carried out. These studies are also
highly misleading because of cold turkey effects in the placebo group. A large
meta-analysis of 65 placebo-controlled trials (6,493 patients) found that only
three patients needed to be treated with neuroleptics to prevent one relapse after
one year. 10 That looks very impressive, but the result is unreliable. The apparent
effect of continued treatment with neuroleptics decreased over time and was
close to zero after three years. Thus, what was seen after one year was mainly
iatrogenic harm - even though it was described as a benefit. When follow-up is
longer than three years, the results might be the opposite of the short-term
results: discontinuing neuroleptics would be best. And that is precisely the case.
In a randomized trial with seven years follow-up, patients who had their dosages
decreased or discontinued fared far better than those who continued taking
neuroleptic drugs: 21 of 52 versus 9 of 51 had recovered from their first-episode
of schizophrenia. 11
Lack of blinding
Because of the conspicuous side effects of the drugs, trials labeled double-blind
are not really double-blinded. Quite a few patients - and their doctors - know
who is on the drug and who is on placebo. 1 It does not take much unblinding in
a trial before the small differences recorded can be explained purely through the
bias in the outcome assessment on a subjective rating scale. 1
Therefore, there is a risk that investigators report something other than what
actually occurs when they medicate patients in trials supposedly double-blinded.
That was the case in a famous trial funded by the US National Institute of
Mental Health in 1964 that is still being cited as evidence that neuroleptics are
effective. It was a six-week study of 344 newly admitted patients with
schizophrenia randomized to phenothiazines such as chlorpromazine, or to
placebo. 12 The investigators reported that the drugs reduced apathy, improved
motor movements and made patients less indifferent – exactly the opposite of
what these drugs do to patients which the psychiatrists had admitted a decade
earlier. 2 The investigators felt that the drugs should no longer be called
tranquillizers - which is what they really are because they sedate people - but be
called anti-schizophrenic drugs. Their study contributed to shaping the erroneous
beliefs that schizophrenia can be cured with drugs and that neuroleptics should
be taken indefinitely. 13
However, the truth is that these drugs do not have a clinically relevant effect on
psychosis. Even helped by all the formidable biases - including cold turkey, lack
of blinding, and industry funding leading to serious data manipulation 1 - the
outcomes of schizophrenia trials have been poor. 1 The least clinically relevant
effect corresponds to about 15 points on the Positive and Negative Syndrome
Scale (PANSS) 14 - commonly used in these trials. Yet, what was obtained in
recent placebo-controlled trials submitted to the FDA is far below this minimum
improvement - only 6 points on the PANSS score 15,16 - even though scores
easily improve significantly when someone is knocked down by a tranquillizer,
leading to less frequent expression of abnormal ideas. 15
For depression, the situation is quite the same. The drugs do not work. The
smallest effect that can be perceived on the Hamilton depression scale is 5-6, 17
but only approximately 3 is obtained in flawed trials. 1 Several meta-analyses
have found that the effect of depression pills is larger if the patients are severely
depressed, 3,18,19 and the pills are generally recommended for severe and
sometimes even moderate depression. However, the reported effects are small
for all severities of depression, e.g. in the most recent meta-analysis, it was 2.7
for patients with a baseline Hamilton score above 23 which is considered very
severe depression, 19 and 1.3 for milder degrees of depression. 3 Moreover, it is
likely just a mathematical artefact that the effect seems to be slightly larger in
severe depression. 20 The placebo-controlled trials have not been adequately
blinded, and the bias caused by the lack of blinding can be large. It was 68% on
average when non-blinded observers were compared to blinded observers in the
same trials in a review that included all diseases, 21 but it does not need to be
large in order to explain the results in the meta-analyses of depression pills.
Since the baseline scores for severe depression are larger than for mild
depression, any bias will influence the measured result more in patients with
severe depression than in those with mild depression. For instance, if we assume
the unblinding bias is 10% when estimating the effect in the drug group and, for
the simplicity of the example, that there is no bias in the placebo group and no
improvement between baseline and the final visit, then a Hamilton baseline score
of 25 would still be 25 after treatment. But because of the bias, there would be a
2.5-point difference between drug and placebo. If the baseline is 15, that
difference would only be 1.5.
It is really very simple and therefore surprising that, as far as I know, I was the
first to describe this mathematical artefact. 20 But then again, what we do not
understand about psychiatrists is a great deal. What seems so obvious to other
people, is often ignored or denied. 1-8,15,22
The small effects of depression pills measured in flawed trials disappear if the
placebo contains atropine, which has similar side effects as the pills, e.g. dry
mouth. 23 It would seem that almost anything with side effects ‘works’ for
depression, 24 which suggests that bias is what is measured in depression trials
on a rating scale. It also seems that everything that numbs people or makes them
euphoric ‘works’ for depression, including neuroleptics, antiepileptic drugs and
stimulants. Three of the 17 items on the Hamilton scale are about insomnia, for
instance, and that issue alone can yield six points on the scale. 25 If a person goes
from maximum anxiety to no anxiety, eight points can be earned. Thus, alcohol
and morphine would surely ‘work’ for depression, but we do not prescribe
alcohol or morphine for people with depression or call them antidepressants.
Irrelevant outcomes
A score on a rating scale says very little - or nothing - about how well a patient is
functioning. Over a thousand placebo-controlled trials of depression pills have
been carried out, but I have not seen a single trial that measured whether the
patients were cured because of the drug, i.e. whether they came back to a normal
productive life with normal relationships to other people. If any such trials
existed, we would have known about them. Unless, of course, they showed that
the drugs did not work - or actually made the situation worse, which they seem
to do - and were therefore buried deep in company archives for no one to see. 1
I recently found a trial that had such outcomes, but it was not designed as other
placebo-controlled trials. 26 Eli Lilly sponsored it. Patients treated with
fluoxetine, sertraline or paroxetine for at least four months underwent five-day
periods where treatment was interrupted and replaced by placebo.
Unsurprisingly, patients on fluoxetine (Lilly’s product) fared well, because this
drug has an active metabolite with a half-life of one to two weeks. Hence, little
change will occur during a five-day interruption.
Also unsurprisingly, patients on paroxetine were harmed because this drug has a
half-life of 21 hours. Even after missing just a single dose of paroxetine, a
statistically significant increase in adverse events was apparent, and as expected,
the severity of the symptoms worsened during the subsequent five days.
Since it was known beforehand that this would happen because of countless
clinical observations - even a similar study sponsored by Lilly 27 - I consider
Lilly’s trial grossly unethical. The abstinence symptoms after paroxetine
withdrawal were severe. 26,27 Patients experienced “statistically significantly
worsened severity in nausea, unusual dreams, tiredness or fatigue, irritability,
unstable or rapidly changing mood, difficulty concentrating, muscle aches,
feeling tense, chills, trouble sleeping, agitation and diarrhoea during placebo
substitution.” 26 In Lilly’s previous trial, 27 roughly one-third of the patients on
paroxetine or sertraline experienced worsened mood, irritability, agitation and
had an increase in the Hamilton score of at least eight, which is the difference
between being mildly depressed and severely depressed. 19
The reason I find this trial utterly unethical is because, for decades, it has been
known that the abstinence symptoms patients experienced on paroxetine - and
were expected to experience in Lilly’s trial - predispose to suicide, violence and
homicide. Many suicides and homicides have occurred when patients
experienced such symptoms. 1,5,22 In 2001, a jury found a drug company liable
for deaths caused by paroxetine: Donald Schell, aged 60, had been taking
paroxetine for just 48 hours when he shot and killed his wife, daughter,
granddaughter and himself. 1 In another court trial, an unpublished company
study came to light showing incidents of serious aggression in 80 patients, of
which 25 resulted in homicide on paroxetine. 1
I have described many other cases of suicide and homicide very likely caused by
paroxetine or other depression pills, including fluoxetine and sertraline, and also
the dirty tricks and the scientific dishonesty involved when drug companies and
leading psychiatrists try convincing us of the opposite: that these drugs protect
against suicide and other forms of violence. 1
Even the FDA, which has done almost everything it can to protect drug
companies marketing depression pills, 1 was forced to give in when it admitted,
at least indirectly, that depression pills can cause suicide at any age, in 2007: 28
“All patients being treated with antidepressants for any indication should be
monitored appropriately and observed closely for clinical worsening, suicidality,
and unusual changes in behaviour, especially during the initial few months of a
course of drug therapy, or at times of dose changes, either increases or decreases.
The following symptoms, anxiety, agitation, panic attacks, insomnia, irritability,
hostility, aggressiveness, impulsivity, akathisia (psychomotor restlessness),
hypomania, and mania, have been reported in adult and pediatric patients being
treated with antidepressants … Families and caregivers of patients should be
advised to look for the emergence of such symptoms on a day-to-day basis, since
changes may be abrupt.”
It seems that the FDA finally admitted that depression pills can cause madness at
all ages and that the drugs are very dangerous - otherwise daily monitoring
would not be needed. Such daily monitoring is, however, a fake fix. People
cannot be monitored every minute, and many have committed SSRI-induced
suicide within a few hours of seeming perfectly fine to their surroundings.
I reached the conclusion that, since the effect of psychiatric drugs is so poor and
the harms so overwhelming, they should not be used at all, or only in acute
situations for a few days - and always with patients’ acceptance and
understanding.
Neuroleptics
Neuroleptics are forced upon patients “for their own sake.” If not forced, few
would take them. I give many lectures and whenever I ask patients what they
would prefer next time they developed a psychosis if they felt they needed a
drug, they have said they prefer benzodiazepines - never neuroleptics. I have
heard chief psychiatrists and nurses deem patients delusional for ‘thinking’
neuroleptics to be dangerous after surfing the Internet! I fully understand why
some patients say they would rather be in prison than on a locked psychiatric
ward. The power imbalance is extreme in both places, and what you say or do
can be used against you, but in psychiatry, the punishment can be killer drugs.
That is not acceptable.
The laws about forced treatment of insane people are insane. In many countries,
a person considered insane can be committed to a psychiatric ward involuntarily
if the prospect of cure or substantial and significant improvement of the
condition would be significantly impaired otherwise. I have not come across any
drugs which can accomplish that. Treatment with psychiatric drugs must be
voluntary, as it is for all other drugs. 1
The other lawful reason for forcing drugs on people is if they present an obvious
and substantial danger to themselves or others. Neither does this rationale hold
water. Psychiatric drugs cause violence 1 and they do not protect against it unless
the patients are drugged to such an extent they become zombies.
Lithium
Lithium is a highly toxic metal used for bipolar disorder and other disorders.
Psychiatrists are proud of this drug, saying it prevents suicide. However, the
authors of a review in BMJ who came to this conclusion were cautious. 31 There
were only six suicides in their meta-analysis, all on placebo, and they noted that
the existence of only one or two moderately sized trials with neutral or negative
results could materially change their finding. Selective reporting could also be an
issue within the known trials. Only about half the suicides and other deaths that
occur in psychiatric drug trials are published. 32 Another problem is cold turkey
in the placebo group. Therefore, a Swedish psychiatrist and I did our own meta-
analysis excluding cold turkey trials. We did not find reliable evidence that
lithium decreases total mortality or suicides. 33
Depression pills
Research based on clinical study reports and other data from drug agencies has
shown that depression pills are far more dangerous than people know. 1,34 We
found they double the occurrence of events that can lead to suicide and violence
in healthy adult volunteers; 35 they increase aggression in children and
adolescents by a factor of 2-3 36 - an important finding considering the many
school shootings where the killers were on depression pills; and depression pills
increase the risk of suicide and violence by 4-5 times in middle-aged women
with stress urinary incontinence. 37 Also, twice as many women experienced a
core or potential psychotic event. 37
Since these drugs have purely symptomatic effects, one highly relevant outcome
is the patients’ conclusions weighing any benefits they have experienced against
the harms. They do that when deciding whether to continue the trials to the end
or drop out. Again, the clinical study reports were useful. Significantly more
patients dropped out on drug than on placebo. 38 That means, that if we have to
choose between a depression pill and a placebo, we should choose the placebo!
People preferred placebo even though some were harmed due to cold turkey
effects in the placebo group. That must mean that the drugs are pretty bad and
should be taken off the market.
We also wanted to study quality of life in the trials but were shocked to discover
that even the clinical study reports were grossly unreliable. 39 The degree of
selective reporting within these reports was unbelievable and drug regulators had
done absolutely nothing to request the missing data from the drug companies,
even though these reports are used in getting drugs approved. I never doubted
the drugs impact negatively on the quality of patients’ lives, e.g. half of all
patients have their sex life disturbed or made impossible after starting the drugs,
1 but I did not expect to find a cover-up in regulatory documents of this
magnitude.
Drugs for what psychiatrists call ADHD - which I consider a harmful social
construct and not a disease 1 - should not be used at all. They may cause children
to sit still in school, but that effect disappears quite quickly and, like any other
psychiatric drug, the long-term effects are harmful. 1 That was clearly
demonstrated in the large US MTA trial which reported results after 3, 6, 8 and
16 years. 40 -44 After 16 years, those who consistently took their pills were 5 cm
shorter than those who took very little 44 - and there were many other harms. I
have heard psychiatrists argue that methylphenidate protects against delinquency
and substance abuse. That is not true - if anything, quite the contrary. 42
Stimulant harms include tics, twitches and other behaviours consistent with
obsessive compulsive symptoms, all of which can become quite common. 15,45
Stimulants reduce overall spontaneous mental and behavioural activity, including
social interest, which causes apathy or indifference, and many children – more
than half in some studies – develop depression and compulsive, meaningless
behaviours. 6,46 Animal studies have confirmed this 46 and have documented
other harms, e.g. we found that the drugs impair reproduction even after the
animals were taken off them. 47 At school, this compulsive behaviour is often
misinterpreted as improvement even though the child may just obsessively copy
everything shown on the board without learning anything. Some children
develop mania or other psychoses, 6,48 and the harms of the drugs are often
mistaken for a worsening of the “disease,” which leads to additional diagnoses,
e.g. depression, OCD or bipolar - and additional drugs, leading to chronicity. 46
Antiepileptic drugs
The situation does not get any better for the remaining psychiatric drugs. Drugs
for epilepsy are commonly used, and like so many other drugs used in
psychiatry, their main effect is to suppress emotional responsiveness by numbing
and sedating people. 6 If you read package inserts, e.g. for gabapentin, you will
see that these drugs double the risk of suicide.
It is not surprising that doctors think antiepileptics work for mania, because
anything that knocks people down ‘works’ for mania. Drugs for epilepsy have
many harmful effects, for instance, 1 in 14 on gabapentin develops ataxia, a lack
of voluntary coordination of muscle movements. Psychiatrists call these horrible
drugs mood stabilizers, even though that is not what they do. In fact,
psychiatrists have never clarified the precise meaning of this term. 15 Who cares
if people are selling snake oil while calling it whiskey?
I have often encountered patients who are on lamotrigine. Only two positive
trials were published for this drug, while seven large, negative trials were not. 50
Two positive trials are all it takes for FDA approval and the agency regards the
others as failed trials, even though they clearly show a drug failure.
Another commonly used drug is pregabalin with its seductive trade name Lyrica.
Certainly, these drugs can cause euphoria, and they are popular among substance
abusers and even some patients as well. You can also get high on marijuana and
cocaine, but the police might come and get you, and we would not call these
substances mood stabilizers.
The amount of fraud in the clinical trials in this area is massive. 1 That means
you should not read anything because it cannot be believed. Unless you have
epilepsy, forget these drugs and, if you are on them, find help getting off them.
Psychotherapy
Almost all psychiatrists say psychiatric drugs are indispensable but that is not
correct. Using them is simply a bad habit. I have met with psychiatrists in
several countries who never use psychiatric drugs or another brain damaging
treatment: electroshock. 1 These psychiatrists handle even the most severely
disturbed patients with patience, empathy and psychotherapy. Good
psychotherapists get far better long-term results than those obtained with drugs,
which is not surprising, and I will explain why.
There are substantial issues when comparing psychotherapy with drugs. The
trials are not effectively blinded, neither for psychotherapy nor for drugs, and the
prevailing belief in the biomedical model would be expected to bias the
psychiatrists’ assessment in favour of drug effects over psychotherapy.
Therefore, trials that show that the effects of a drug and psychotherapy combined
are better than either treatment alone should be interpreted cautiously, and one
should also remember to only look at long-term results, i.e. a year or more.
Despite the various biases in the clinical trials, some hard-hitting facts about
psychotherapy are apparent. Cognitive behavioural therapy halves the risk of a
new suicide attempt in people acutely admitted after a suicide attempt. 20
Depression pills increase the risk of suicide and violence. It is insane to not drop
these pills for good and offer everyone psychotherapy instead. Not only is
cognitive behavioural psychotherapy effective, emotion regulation
psychotherapy and dialectical behaviour psychotherapy are also effective for
people who harm themselves. 58
Psychotherapy does not work for everyone and some therapists are not
competent or do not work well with some patients. Therefore, it can be
necessary to try more than one therapist. Yet, some people cannot be helped no
matter what we do, which is true also in other areas of healthcare. It is also
important to be aware that psychotherapy can be harmful. Forcibly recruited
child soldiers in Uganda have done remarkably well by avoidant coping. 62 If a
therapist insists on confronting these people with their encapsulated traumas, it
could backfire quite badly.
Let us return to the broken leg allegory I presented at the beginning of this
chapter. Broken legs heal by themselves. We can help with plaster casts and
screws, but they will heal even without our help. Psychiatry is similar. Just like
we need physical pain in order to avoid dangers, we need emotional pain to
guide us in life. 63 Acute problems like psychoses and depression are often
related to traumas and they tend to self-heal if we are a little patient. Through the
process of healing - whether assisted by psychotherapy or not - we learn
something important which can be useful if we get in trouble again. Such
experiences can even boost our self-confidence, whereas pills can prevent us
from learning anything because they numb our feelings and sometimes even our
thoughts.
Pills can also provide a false sense of security. Doctors may not think they need
to engage themselves as much because a patient is taking drugs, so it will
probably be okay (it will not because the drugs do not have any relevant effects).
63
Censorship
Leading psychiatrists and the drug industry have been so effective in shaping
myths about what psychiatric drugs do to people that creating space for a more
truthful perspective can be very difficult. In my experience when it comes to
censorship, self-censorship and political correctness, Sweden is particularly
difficult.
When a fine Swedish psychiatrist recently wrote a book about her decision not to
use drugs in her practice 63 - she had trained as a psychotherapist - a colleague
who read her manuscript worried the book might kill people. Another believed it
would be disastrous if people questioned their psychiatric diagnoses. I think it is
a wonderful book.
Whitaker recently published an article with the biblical title, Thou shall not
criticize our drugs, about two highly relevant letters criticizing a terribly flawed
paper on the long-term effects of neuroleptics. 66 The letters were rejected by the
editor of the American Journal of Psychiatry, which is owned by the American
Psychiatric Association - an organization solidly corrupted by industry money. 1
These letters can be read on Whitaker’s website.
The Copenhagen documentary film festival, CPH:DOC, the largest in the world,
showed a very moving Norwegian film, Cause of death: unknown, in 2017. It is
about the filmmaker’s sister who died very young after her psychiatrist had
overdosed her with olanzapine (Zyprexa), a neuroleptic drug that turned her into
a zombie. Her psychiatrist was so ignorant that he did not even know that
olanzapine could cause sudden death. I appeared in the film and the filmmaker
asked the organizers to put me on a panel discussing the film afterwards.
CPH:DOC announced the film with my name as the only panelist: Medicine or
manipulation? Film and debate about the psychiatric drug industry with Peter
Gøtzsche.
That announcement remained the same until seven days before the film was to
be screened. Then I was kicked off the panel under the pretense that the
organizers could not find a psychiatrist willing to debate with me. It turned out
that the Lundbeck Foundation had provided a major grant to the festival. The
Lundbeck Foundation looks like an independent fund. It is not. Its objective is to
support the business activities of Lundbeck, a Danish maker of psychiatric drugs
including neuroleptics. CPH:DOC never contacted me about it, even though I
know several psychiatrists who are willing to debate with me.
Apart from the film’s instructor, its producer and a previous patient who had
written a book about her experiences, the list of panel members was
embarrassing: Nikolai Brun, a chief of staff employed by the Danish Drug
Agency four months earlier after a long career in the drug industry, and
psychiatrist Maj Vinberg. Vinberg had financial conflicts of interest in relation to
Lundbeck and AstraZeneca. She is very positive towards psychiatric drugs but
seems to know little about them. Vinberg has also published utter nonsense about
depression being hereditary and observable in brain scans. Earlier that year, I had
responded to other nonsensical statements she had made in a Danish industry-
funded magazine where she had characterized the most thorough meta-analysis
of depression pills ever made 3 as “a smear campaign against antidepressants
drugs … doubtful populistic discussions … armchair gymnastics … performed
by a group of doctors, statisticians and medical students without special
knowledge about psychiatry and, thus, depressive disorders” (not true). This
meta-analysis was high-level science telling us that depression pills do not work
and are harmful. 3 But Vinberg was angry: Thou shalt not bring down the myths
of psychiatry without eliciting the Lord’s rage!
I wrote in the magazine 67 that I had published an article, The meeting was
sponsored by merchants of death,68 which included one of Vinberg’s industry
benefactors.
The panel debate was a total farce. After 25 very boring minutes, excepting the
filmmakers’ contributions, only five minutes were left. At that very moment a
former patient interrupted Nikolai Brun - who had talked endlessly - by
shouting: “Questions!” People in the audience - many of whom had lost loved
ones, killed by psychiatric drugs - had become increasingly angry because these
panelists only discussed amongst themselves and had quite clearly avoided
involving the audience. There was only time for three questions.
A woman asked, since neuroleptics killed people, why they had not been taken
off the market. Brun replied that he was not an expert on psychiatric drugs.
Nonetheless, he embarked on another endless talk, this time about cancer drugs
which was totally irrelevant. Yes, we know perfectly well that some people are
killed by cancer drugs, but more are saved. That is why we use them. If
psychiatric drugs have saved anyone, it would be a drop in the ocean compared
to how many they have killed. That is why they should be taken off the market.
I had enough and demanded: “Questions from the audience!” A young man said
that he had tried getting off his depression pills several times without success
and he had not received any help from doctors. The Nordic Cochrane Centre
later helped him to get off his drug.
The last question was posed by a Danish filmmaker, Anahi Testa Pedersen, who
had made a film about me and her own experiences as a psychiatric patient,
Diagnosing Psychiatry, which had world premiere in the same cinema seven
months later. Pedersen asked why I was taken off the panel since I could have
made a good contribution. A spokesperson for the festival replied that they had
asked “a lot of people,” but that no one wanted to debate with me. Pedersen
interrupted and named a psychiatrist who would have liked to come. The
spokesperson did not reply to this but instead said that since the film itself was
critical, there was no need for me; they needed someone to debate the film’s
messages.
In the middle of these endless excuses from the spokesperson, someone in the
audience shouted: “There is no debate!” The spokesperson replied that they
would invite me for “tomorrow’s debate,” which I of course did not accept
because I had been kicked off from the world premiere of the film.
Seconds before the allotted time ran out, I stood up and shouted (because I
doubted they would hand me the microphone): “I am actually here. I debate with
psychiatrists all over the world, yet I am not allowed to do this in my home
town.” That caused a big laughter and applause.
After the film festival, Pedersen wrote about the affair in a journalist magazine.
69 She pointed out that before I was removed, they had announced that sharp
focus would be aimed at the overconsumption of psychiatric drugs and at
whether drugs were the best treatment of psychiatric disorders. After my
removal, focus was moved to the relationships between doctors, patients and
industry – the subject of my award-winning book from in 2013. 70
CPH:DOX writes on its website: “We have many years of experience with
sponsorship agreements that cater to both individual enterprises and to the
festival. All collaborations are created in close dialogue with these individual
businesses and are based on common visions, challenges and opportunities." 69
In response to Pedersen’s article, Vinberg wrote that it was a pity that a debate,
which was supposed to be about improving the future treatment of people
suffering from a severe mental disorder in the form of schizophrenia, ended in a
rather indifferent debate about individuals (she meant me). 69 Her statement did
not agree with her evasive responses during the panel debate.
Three years later, she was back on paroxetine at a lower dosage. Because of
suicidal expressions after three months, she was seen by a hospital psychiatrist
who advised continued use of paroxetine. That is malpractice. A person who
becomes suicidal after three months on a depression pill should be tapered off
the drug as quickly as possible.
The other expert witness was Professor Anton J.M. Loonen. Although he had
worked as a physician and clinical pharmacologist responsible for long-term,
intensive psychiatric care of patients with serious psychiatric disorders, in his
expert report, he did not comment about whether Versluis had been treated
according to good professional standards. He wrote that in the months leading up
to the homicides, Versluis constantly had paroxetine side effects: dizziness,
anxiety, gloominess, and mood swings. She cried a great deal; she had not been
able to sit still and kept busying herself without finding peace; and she was
constantly fidgeting and wiggling.
Loonen did not explain in his report what that all meant even though it is
obvious: These are strong warning signals that drug-induced suicide or homicide
might occur. 1,5,6
During this time, Versluis fainted and fell down stairs. She told two people about
nightmares where she slit her children’s throats. Two days prior to the
homicides, she reported to her ‘supervisor’ that she was ill, and she told several
people that she was not feeling well. She also went to her family doctor (who
had prescribed paroxetine) with her complaints and visited her company doctor
who dismissed her. Finally, she contacted her psychologist who did not have
time for her.
It appears to me that Versluis did everything she possibly could, yet the people
charged with her care were indifferent to her ominous symptoms. I concluded to
the court that “serious professional malpractice was a causal factor that was
partially or wholly responsible for the two homicides.” I found it deeply
concerning that Loonen said absolutely nothing in his long report about his
colleagues’ appallingly poor conduct.
Versluis left a suicide note and tried to kill herself after killing her two children.
When arrested, she was not responsive, did not answer questions or orders.
Three days after her arrest, she was seen by a forensic psychiatrist who noted
that, intellectually and rationally, she was not functioning well, and in the areas
of feeling, emotion and affection, numbness was apparent - a kind of twilight
state with a suspended wish for non-existence. Versluis indicated that she could
only remember fragments and did not fully regain consciousness until she was in
the cell.
Loonen noted that Versluis was still not herself three days after the homicides
and acknowledged that depression pills can cause reduced consciousness in
various ways.
Loonen noted that on the day of the homicides Versluis took overdoses of both
paroxetine and midazolam (a benzodiazepine - another drug that can cause
violence). He stated that she must have used paroxetine irregularly prior to the
homicides because her blood level of paroxetine was far too high. That could be
related to the fact that Versluis was an intermediate metabolizer (i.e. her body’s
capacity for breaking down and excreting paroxetine was decreased).
Versluis was accused of having deliberately and premeditatedly taken the lives
of her children. That is totally absurd. If she had killed her children in an LSD
psychosis, I doubt the court would have entertained such an accusation.
Furthermore, I noted in my report to the court that my research group had just
finished a systematic review of double-blind, placebo-controlled trials of
depression pills given to healthy adult volunteers, which found that the drugs
doubled the occurrence of harms that predispose to suicidality and violence. 35
Loonen stated in his report that the risk of serious violence (suicide and
homicide) caused by depression pills must be considered extremely small.
However, it is irrelevant for judgments about cause-effect relationships how
rarely events occur. The risk of airplane crashes is also extremely small but,
when they happen, we do try to find out why.
At some point in the proceedings, Loonen suddenly asked the court for
permission to hand out a 4-page note he had written, which was granted. As it
was in Dutch, I could not read it, but someone translated it for me during lunch
break. In it, Loonen attempted to counter my criticism of his own report to the
court and concluded by saying, “I am of the opinion that this gentleman goes far
beyond the pale in his report and, thus, successfully obstructs the course of
justice. I hope the fairness of the trial against Mrs. Versluis will not sustain
damage.” Outrageous remarks considering I had done everything to ensure she
received due process and that he, himself, had done quite the opposite.
On the first page, Loonen wrote that I had become a controversial figure by
making several assertions about the scientific stature of psychiatry - statements
that went beyond my field of expertise. He called this tragic and false, and stated
that the Cochrane Collaboration had distanced itself from the content of my book
on psychiatry. 1 Loonen suspected I suffered from a mental disorder that made
me seriously disinhibited and advised that I should be examined by a doctor in
order to protect myself from myself.
I told the court it was grossly unfair - both to Versluis and to me - that Loonen
presented a document to the court I had not seen beforehand and therefore could
not comment on. I asked the court to disregard it, which the court did. Later, I
complained over Loonen’s unethical conduct towards a colleague to several of
the institutions where he worked and to the Dutch Medical Association, which
turned me down with the excuse that I was not a Dutch doctor. In every instance,
I was told it was none of their affair, or that I should complain elsewhere. The
University of Groningen where Loonen was employed ignored me for two years.
It took six emails before they reacted. I was informed that, during a meeting the
Dean had arranged, Loonen was told that his conduct was inappropriate and that
he must prevent the university from suffering possible damage because of his
behaviour.
Two weeks after the proceedings, the prosecutor asked for a 14-year jail sentence
and a hospital order for compulsory treatment. I replied to Versluis’s lawyer that
absolutely nothing would ‘work’ for Versluis other than keeping her away from
psychiatric drugs.
Four months after the court proceedings, I went to Holland again to lecture about
psychiatry as an invited speaker at an international scientific meeting in Leiden.
71 Loonen tried to prevent me from speaking. He wrote to the organizer of the
meeting referring to the court proceedings and claiming that I, for personal
reasons, had violated the requirements of confidentiality as an expert witness by
making public Loonen’s reports to the court. That was not true. I had shown the
defamatory note he suddenly presented to the court to a journalist, which I was
entitled to do because there was nothing confidential about it. Interestingly,
another speaker, Allen Frances - once regarded as the most powerful psychiatrist
in the United States - said during his talk that I had provided a tremendous
service to psychiatry.
The case was appealed to the Dutch Supreme Court. Versluis’s lawyer wanted
me to participate but that was rejected by the court, arguing that I could not
provide unbiased research into the case because I had already presented my
views. Interesting. Even if you do your best to be unbiased, the mere act of doing
disqualifies you.
Versluis should be set free. It is meaningless to keep her in jail. I also believe her
court case constitutes a miscarriage of justice.
9 Physical pain
What should you do if you experience pain somewhere in your body? Most
importantly: That happens to all of us and it usually goes away rather quickly.
Your attitude to pain is therefore crucial (see also the section about back pain in
Chapter 2).
We should rarely do anything about pain, but that is not how the world operates.
I often see mothers giving analgesics to their children because of headaches,
fever (see Chapter 5) or something else. In my household, we have never done
so.
I have looked up the effect of paracetamol (acetaminophen) and do not find the
results impressive enough to want to take it. I have tried it in the past but never
noticed any effect.
NSAIDs
I would not dream of taking aspirin for pain, because this drug belongs to the
drug class that is misleadingly called non-steroidal, anti-inflammatory drugs
(NSAIDs). These drugs are harmful, and aspirin might cause bleeding among
other things.
The fraud in research and marketing of NSAIDs is formidable. 1 The biggest lie
is probably the one that has given rise to the name of these drugs: anti-
inflammatory. The story behind this is when newly synthesized cortisone was
first given to patients with rheumatoid arthritis in 1948, the effect was so striking
that some people believed a cure had been discovered. 2 The initial enthusiasm
evaporated quickly, however, when the drug’s serious harms came to light.
However, I have shown in my research that these drugs are not anti-
inflammatory. 1 They merely reduce pain and fever like paracetamol but are far
more dangerous and expensive. They delay wound healing and it is therefore
particularly bad practice to use them for sports injuries, also because pain is an
important signal that has helped us survive throughout evolution. When
something hurts after an injury, it is a warning about letting that body part rest
until it heals. If the signal is blunted with pain killers, it could make matters
worse and cause acute problems to become chronic.
Not all pain goes away quickly, and if you do not know what caused it, you
should try to find out. The hope is to find a cause and a treatment that works.
Not a pain killer but a treatment that can remove the cause.
Chronic pain is different. Many people become dependent on opiates and many
are killed by them, whether abused or not. For doctors, chronic pain patients can
be very difficult to deal with because they can be chronically dissatisfied no
matter what you do for them. It is important to realize that when nothing seems
to work, psychological factors are often at play. Therefore, rather than getting
absorbed in hopeless regimes of ever-changing polypharmacy, we should focus
much more on trying to persuade patients that pain, to a considerable extent, is a
matter of how you choose to react to it.
Antiepileptic drugs
Pain is such a lucrative market that all sorts of drugs are being sold as pain
relievers. Like in psychiatry, virtually everything that has side effects, which all
drugs have, seem to ‘work’ for pain. Drugs for epilepsy and depression are much
used despite their toxicity. I often wonder how much of the measured effect is
bias and how little is true effect, or whether there is any true effect at all, but I do
not have time to pursue all the questions I have about drugs. I will leave it to
you, now that you know how to find the evidence. But be cautious. There is a
great deal of fraud in these trials, 1 which means that published trial reports and
reviews of trials cannot be trusted.
That does not look particularly attractive, and Pfizer’s prescribing information
for US physicians is frightening. 4 It consists of 70 pages but there is a short
summary at the top. Under Adverse Reactions, it states that the adverse
reactions, which occur in at least 5% of the patients and which are twice as
common on drug than on placebo, are dizziness, somnolence, dry mouth,
oedema, blurred vision, weight gain and abnormal thinking (primarily difficulty
with concentration/attention). The summary warns that pregabalin may cause
“foetal harm” during pregnancy (which usually means malformations) and
breastfeeding is not recommended. Under WARNINGS AND PRECAUTIONS
are listed angioedema (swelling of the throat, head and neck) that can be life-
threatening, hypersensitivity reactions, increased seizure frequency in patients
with seizure disorders if Lyrica (pregabalin) is rapidly discontinued, and
increased risk of suicidal thoughts or behaviour, peripheral oedema. Lyrica may
cause dizziness and somnolence that impairs patients' abilities to drive or operate
machinery.
Would you take this drug if you had neuropathic pain or fibromyalgia? I would
not, no matter how severe my pain was. Dizziness and somnolence are common
with antiepileptic drugs and the death risk is not negligible: You might die in a
traffic accident or fall and break your hip, in which case about 20% will be dead
within the next year. There are many ways to kill people with drugs, and in most
cases such deaths are not recognized as being drug-induced deaths. That means
we do not learn anything from the many deaths which therefore just continue to
pile up.
You might also consider the amount of fraud that has been documented for
another antiepileptic drug, gabapentin (Neurontin), which Pfizer sells for pain. 1
Manipulations with the trials involved selective statistical analyses, selective
reporting of outcomes that happened to show a positive effect, inappropriate
exclusion or inclusion of patients in the analyses, multiple publication of
desirable results, differential citation of Pfizer results, and spin to make negative
results appear positive. The bias was already introduced at the design stage, e.g.
high doses were used that led to unblinding and biased reporting of subjective
outcomes. Pfizer even recognized that unblinding due to adverse events could
result in corruption of the study’s validity.
Opiates
Opiates are yet another horror story. A letter published in New England Journal
of Medicine in 1980 has been very harmful. It was written by two people from
the Boston Collaborative Drug Surveillance Program, and its title was
informative: Addiction rare in patients treated with narcotics. 6 There were only
five sentences in the letter:
It is extremely rare that the editors of this journal admit their mistakes 1 and they
did not do so on this occasion either. On the journal’s website, there is a note
from the editor, published 37 years later:
“For reasons of public health, readers should be aware that this letter has been
‘heavily and uncritically cited’ as evidence that addiction is rare with opioid
therapy.” Retraction Watch wrote about the affair the next day. 7
The journal did not comment on the merits of the letter but put the blame
elsewhere: The problem was how it had been used by others. Researchers from
Canada wrote that the letter had been heavily and uncritically cited (608 times)
and that this had contributed to the North American opioid crisis by helping to
shape a narrative that allayed prescribers’ concerns about the risk of addiction
associated with long-term opioid therapy. 8
A former editor of Anesthesia & Analgesia called for an apology from the
authors of the 1980 letter and the editors of New England Journal of Medicine. 7
“Both should be deeply ashamed to have published such data without providing
adequate scientific documentation of the methodology, results, and study
limitations. Both should apologize for the incredible damage, including
destroyed lives, communities, and thousands of deaths traced to this
irresponsible Letter to the Editor … so far below acceptable reporting standards
as to nearly represent misconduct.”
We have seen similar fraud with tramadol, which also became widely used. The
initial ad in the Danish Medical Journal introduced the drug in this way:
“Finally, a highly effective pain reliever with a low dependency risk.” 9 The drug
was said to be minimally addictive. Danish television asked the Danish drug
agency which studies had caused the agency to accept Grünenthal’s claim of a
low addiction risk. The agency handed over eight studies which were scrutinized
by three experts who all concluded that none of the studies provided evidence
that the patients rarely became addicted. Some of the studies did not look at
addiction at all or included drug abusers or people who had only taken the drug
for a couple of weeks. There was also a study of four monkeys, one of which
died.
Confronted with this, the drug agency said that they also based their judgment on
other data. However, the agency was asked to provide exactly those data on
which it based its judgment. And we have seen no data that supports the
agency’s wishful thinking.
There are no quick fixes with opiates. They are all harmful - with and without
prescription. So, what did the drug agency do to help the 300,000 Danes (5% of
the population) who take tramadol or to rectify its own faults? Nothing. It did
not change its totally unfounded belief that the risk of dependence is low 10 but
instead passed the buck to those who had believed in the propaganda by
sharpening the reporting obligation for doctors when they saw cases of
dependence. That was a fake fix and highly unprofessional as well. The agency
reversed the burden of proof in effect. They let Grünenthal get away with the
incredible assertion that their opiate did not cause addiction - which all opiates
do - without having a shred of proof that this was correct.
A Danish TV documentary, Smerter til salg (Pain for sale) from December
2017, showed how totally corrupt the pain area is. Grünenthal is the great
corruptor. It is everywhere, supporting numerous meetings, doctors, patient
organizations and even guidelines about how to treat pain. Even the head of the
Danish drug agency has worked at Grünenthal, the producer of thalidomide,
which fiercely denied that this drug could cause birth defects in the 1960’s.
Nothing has changed. Grünenthal has now launched a successor to tramadol –
tapentadol - which is 20 times more expensive, again claiming that addiction is
nothing to worry about, again without having any science to back up this
preposterous claim.
As the treatment effect diminishes with time, larger doses are needed.
Glucosamine
A full-page advertisement in a major newspaper promised an effect of Revadol,
suggesting that you can become revalidated, i.e. cured. We were told that clinical
studies have shown it lessens pain after four weeks and improves joint mobility
in patients with osteoarthritis. Is this too good to be true? Certainly.
Glucosamine is naturally present in our bones and is one of the building blocks
of cartilage. We produce it ourselves in our bodies. Therefore, it does not make
sense that supplements containing glucosamine would work for people with
osteoarthritis. Is there any evidence that people develop osteoarthritis because
they produce too little glucosamine? No, and if there had been, surely the ads
would have told us. According to Wikipedia, it is one of the most common non-
vitamin, non-mineral, dietary supplements used by adults in the United States.
The review is from 2005 and common sense tells us this can hardly be true. So,
we must try a little harder and go beyond the Cochrane review. If you google
glucosamine systematic review, the first entry is a review in BMJ from 2010. 12
One of the authors is one of my previous PhD students, so I expected this review
to be done well, which it was.
Some of the authors of the BMJ review had previously shown that small trials of
glucosamine are much too positive. 13 Therefore, they only included randomized
trials with more than 200 patients. No meaningful effects of glucosamine were
found, and industry independent trials showed smaller effects than commercially
funded trials. The authors concluded that glucosamine should not be used.
Bravo! No mumbo jumbo about talking to your doctor about whether
glucosamine is right for you. But of course, the authors are from Europe, not
from the Unites States, and you do not need a prescription for glucosamine.
10 Treatment of cancer
We hear a great deal about progress against cancer, now called a chronic disease
even though most people, by far, still die from their cancer despite being
presented with convincing survival statistics. Yet little progress is actually being
made, which is not the impression you get from newspapers and TV that often
quite uncritically propagate highly misleading information from cancer charities.
1
The propaganda is massive, and I shall therefore explain what is wrong with the
type of data we see most often. It requires a little attention to look behind the
facade and dissect the inflated messages, because there are several ways to
measure progress - all of which have weaknesses.
One of the best things we can do is to look at the annual age-adjusted mortality
of individual cancers. Mortality needs to be age-adjusted. Since we are getting
increasingly older, more of us will die from cancer, no matter what we do.
The problem with this method is it is hard to know what people die from,
especially when autopsies are no longer common. When a person has been
diagnosed with a cancer, there is a risk the diagnosis will also be considered the
cause of death if that person dies in an emaciated condition. But the cause of
death could be another cancer or unrecognized heart disease.
The opposite can also happen. If you believe a patient is well - without
recurrence - after treatment, you may think something else killed the patient.
However, what we see most often is a period of survival after the diagnosis has
been made - 5-year survival, for instance. Unlike age-adjusted cancer mortality
where bias, which is rarely extensive, can go both ways, bias in this case almost
always leads to overestimations of the results of screening for and treatment of
cancer. The bias can easily be so large that interventions having no effects appear
to be quite effective.
If the diagnosis is made earlier than previously, we will see an increase in 5-year
survival even when the early diagnosis does not improve survival, and where the
age-adjusted mortality of the cancer is therefore unchanged. Patients do not live
longer; they live longer with the knowledge that they have cancer because the
clock started earlier. Therefore, they are harmed by this type of early diagnosis.
We diagnose some cancers earlier than before because patients and doctors have
become more aware of cancer symptoms. For breast cancer, for example, the
average tumour size in Denmark was 33 mm in 1978-1979, but only 24 mm ten
years later. 2 This decrease had nothing to do with screening for breast cancer,
because it occurred before screening was introduced.
In 2016, a journalist wrote that, after being in last place in cancer survival,
cancer treatment in Denmark was now at the same level as in our neighbouring
countries. 4 The argument was that 5-year survival for breast cancer had risen
from 12% to 18% in 20 years. But we now had screening for breast cancer in
Denmark, which leads to 33% overdiagnosis. 5 That means many healthy women
who would never have received this diagnosis in their lifetime if they had not
gone to screening will get a diagnosis of breast cancer. As none of them would
have died from the disease, this will improve 5-year survival, of course.
Twenty years ago, we only had screening for 20% of the country. For
simplicity’s sake, if we assume that the 5-year survival rate of 18% comes from
a population that is not screened, and 12% comes from a screened population,
the calculation is easy. Without screening, 18% of 100 women with breast cancer
die over five years, i.e. 18 women. With screening, 12% of 133 women die (the
same 100 plus 33 healthy, overdiagnosed women), which equals 16 women or
almost the same. That is not to say there has been no progress in the treatment of
breast cancer, but the calculation shows that 5-year survival after a breast cancer
diagnosis is highly misleading. The true progress must be far less than the
difference between 18% and 12%.
If cancers were always the same, that would be an outstanding progress in the
treatment of malignant melanoma, but that is not the case. The explanation is
that many more diagnoses are being made because people are more likely to get
their brown spots examined. Almost all of these additional cancers are harmless.
6,7
The best thing we can do to find out whether a treatment of cancer has any
benefit is to do a randomized trial. If we do a randomized trial, there is no
problem in using 5-year survival because everyone has cancer to begin with, and
the randomization ensures that the two groups are comparable for prognostic
factors.
So, what do the randomized trials tell us? A 2004 meta-analysis of 250,000 adult
cancer patients treated with chemotherapy in randomized trials showed an effect
on 5-year survival for testicular cancer (40%), Hodgkin's disease (37%), cervical
cancer (12%), ovarian cancer (9%) and lymphoma (5%). 8 That is reassuring but
these cancers represented less than 10% of all cancer cases. In the remaining
patients, 5-year survival increased by less than 2.5%, which corresponded to
only three months. New drugs for solid cancers approved by the European
Medicines Agency increased survival by only one month compared to other
regimes. 8
Breast cancer was not among those cancers where a worthwhile effect of
chemotherapy was shown. Yet, that is not what people think. We constantly hear
about breast cancer, and people think screening works - which it does not (see
Chapter 7) - and that chemotherapy is effective - which it is not. I, too, fell
victim to the propaganda. I knew that chemotherapy increased survival and
thought it was a substantial effect, and I even recommended it to a patient
worried about serious harms. That was before I wrote this book. When I looked
up the evidence, I got shocked.
If you google chemotherapy breast cancer, the first entry goes to the American
Cancer Society, which begs you to DONATE in white letters on a red
background. You should not donate because the society is obscenely wealthy and
spends a lot of money on its own people, just like the cancer society does in my
country. By 1989, the society’s cash reserves were worth more than $700
million; 74% of its budget was spent on “operating expenses,” including about
60% for generous salaries, pensions, executive benefits and overhead. 1
Furthermore, the society has partners that include drug companies like Pfizer,
Bristol-Myers Squibb, Abbvie, Merck, Quest Diagnostics, AstraZeneca, Abbott,
Eli Lilly, and Genentech. Some of these companies earn exorbitant amounts of
money selling chemotherapy at extremely inflated prices that do not reflect
research and developments costs. 9 Even Morgan Stanley, which played a major
role in the global financial crisis in 2009, is one of the partners. What a group of
playmates.
The American Cancer Society once announced that early detection of breast
cancer results in a cure “nearly one hundred percent of the time.” 11 That was an
error of “nearly one hundred percent,” since mammography screening does not
lead to cures.
The society says nothing about the effects of chemotherapy, only when chemo
should be used, and despite providing a long list of serious harms, it omits the
statistics regarding their frequencies. The text starts out by saying that, “Chemo
drugs can cause side effects.” Can? Has anybody ever heard about a patient who
did not get harmed by chemo? No. There is no free ride.
If you add cochrane to the Google search, the fourth record takes you to a page
that says there are 17 Cochrane reviews of chemo for breast cancer, divided
according to whether the cancer was advanced or not. The American Cancer
Society noted that polychemotherapy is often used and the first Cochrane review
found that the addition of one or more chemotherapy drugs to a regimen caused
greater shrinkage of tumours seen with imaging, but increased toxicity. 12 There
was insufficient evidence to determine an effect on overall survival or length of
disease progression. The hazard ratio for survival (similar to the relative risk)
was about one, 0.96 (95% confidence interval 0.87 to 1.07, P = 0.47), and time
to progression was also unchanged, 0.93 (0.81 to 1.07, P = 0.31).
We always prefer absolute risks to relative risks (e.g. a risk ratio or hazard ratio),
and a large meta-analysis from 2005 provides this. 13 This study is about early
breast cancer, which means that the cancer and any affected lymph nodes can be
surgically removed and includes both chemotherapy and hormonal therapy. It
runs for more than 31 pages in The Lancet, which would take many hours to read
and digest. But that is not necessary. A graph shows that, for women aged 50 to
69 years who received polychemotherapy, the breast cancer mortality is 47.4%
after 15 years, compared to 50.4% in women who did not get multiple drugs.
Not much of a difference, but it is reassuring that half of the women with early
breast cancer had not died from breast cancer after 15 years.
Yet, that does not mean that half the women are still alive after 15 years. Some
died from other causes, including the chemotherapy they received, which is why
breast cancer mortality is a flawed outcome. Thus, the most important outcome
in cancer trials is always total mortality. We do not know whether
polychemotherapy reduces total mortality because the 31-page paper does not
tell us. The readers are referred to figures 1, 6 and 8 in a web appendix not
included in the paper.
But my troubles were not over. There were three PDF files. As the first one was
called Annex-Figures 1-13, this should be the one I was looking for. But it was
not what it was supposed to be. There were 249 pages of graphs and often more
than one graph on each page, with no meaningful legends to help me find what I
was after. I could not find any figures 1, 6 and 8. The first graph showed annual
event rates, but there was no information whether these events were total
mortality, breast cancer mortality, recurrence of the cancer or something else.
Another PDF told me that the information I was looking for was to be found on
another website, outside the journal’s control! The third document - 142 pages
long - contained some other information.
I looked out the window and swore loudly but did not want to give up. I started
browsing the many hundreds of graphs. Nowhere was anything called figure 1, 6
or 8. But on page 17, I found the graph on breast cancer mortality I had also seen
in the paper, with a 3% difference in breast cancer mortality after 15 years
(50.4% versus 47.4%). The next graph was actually labelled “Any death,” which
was 55.7% versus 53.6%, i.e. a difference of 2.1%. I knew it! Of course, total
mortality was higher than for breast cancer alone and - of course - the mortality
benefit was lower because some women were killed by the chemo.
Why were there no data on the only unbiased outcome - total mortality - in the
31-page Lancet paper? And why were these data so well-hidden that only people
as stubborn as me could find them?
This story illustrates what has been documented many times before: academia
can be just as biased as the drug industry and just as ‘skilled’ at hiding the most
important facts.
If that woman with breast cancer asked me today, I would tell her that I would
not recommend polychemotherapy, and probably not an individual
chemotherapeutic drug either, considering the meta-analysis of treatment of
various cancers mentioned above. 8 Taxanes do have some effect compared to
other chemotherapies, which a Cochrane review shows, 14 but the question is
whether these small effects make it worthwhile to get chemo.
It would take some time to find out the effects of single agent chemotherapies
because there are so many of them. You would also need to learn some basic
issues, like the difference between adjuvant therapy and neoadjuvant therapy
(which means chemo before surgery). There are also many forms of breast
cancer. Therefore, the easiest way forward will be to ask your doctor what the
precise effect is compared to no treatment. The doctor should be able to respond.
People - including most doctors - often say that a small average benefit can be
worthwhile because some patients benefit more than others. “Perhaps I will be
one of the lucky ones who adds 6-12 months to my life, not the 1-3-month
average.” Sometimes patients refer to other people who lived many years after
polychemotherapy.
That is a false hope. Some patients live especially long because cancer is highly
variable, with highly varying growth rates. 1 Some women are therefore
predestined to live much longer than others. It has nothing to do with the
treatment. We can only make rational decisions if we base them on the average
life extension obtained in randomized trials.
In Denmark, prominent doctors have declared publicly that they would abstain
from life-prolonging chemotherapy if they got lethal cancer, 9 and few
oncologists and nurses are willing to accept the chemo their patients endure for
minimal benefit. 9,17 I wonder why we do not offer patients the same privileges
that we enjoy as health professionals. A woman, only 39 years old, who recently
died of breast cancer, said after four courses of chemo, "If this is my last spring,
I'd like to put myself in the middle of it instead of having to go to hospital all the
time." 16 It was her last spring.
There is something badly wrong with the way we approach incurable cancer
(which almost all cancers are), and I will end this chapter with two recent stories
from my nearest family which illustrate how absurd it can be to fight a battle you
cannot win. 16
Obituaries often say: “He lost the battle against cancer.” I would prefer we left
out the war rhetoric and said something positive like: “He had a good life.”
My two relatives battled until the very last moment. One, a 67-year-old man was
diagnosed with incurable stomach cancer with metastases to the kidney and the
liver. As far as I can tell, absolutely nothing could be reasonably done, yet the
patient underwent many diagnostic tests, which, due to their invasive nature,
aggravated his condition. Several types of chemo were tried, and at one point the
patient and his wife were told that he would be offered life-prolonging treatment.
They both perceived this message very positively - like a four-year extension of
life. The reality was it was extremely unlikely any life extension would be
obtained; in fact, it would be more likely the chemo would kill him. Yet this
false hope led to a series of additional chemo regimens that pestered the last six
months of his life. He did not experience one single tolerable day and was
constantly plagued by the harmful effects of the chemo. That was not dignified -
not a good death.
Adding to all this misery, we have totally spineless drug regulators who approve
new cancer drugs without having a clue whether they are better or worse than
those we already have. 9,18 This broken system has resulted in huge expenditures
on cancer drugs with certain toxicity but uncertain benefit. Even when
randomized trials have been performed and marginal advantages have been
found, these trivial differences may disappear when the drugs are used in real
life on patients suffering from co-morbidity. 18
11 The digestive tract
An overlooked ileus
Early one morning three years ago, I got a pain in the lower abdomen that I did
not bother about. Then suddenly, an extremely strong pain ascended to my
mouth which, within seconds, filled with saliva with a pronounced bloody taste.
I rushed to the toilet in extremely poor condition thinking I might die. I have had
pain caused by kidney stones - described in textbooks as the most intense pain
you can experience - but this was worse, and I immediately thought it must be
ileus. My wife, who is also a doctor, came home an hour later and we listened to
my stomach with a stethoscope. No bowel sounds. We both had strong
suspicions of ileus.
The pain was not present all the time and, in the afternoon, I ran 5 km in the
woods alone - which was quite irresponsible, yet illustrates the type of optimistic
person I am. Unlike other runs, I did not pass gas during my run or the rest of the
day. No stools the whole day. Then I got equally violent attacks in the evening as
in the morning and we were convinced I had ileus.
Two hours later, a doctor declared that I did not have ileus because I had bowel
sounds, my abdomen was soft, and I appeared unaffected. I repeated that I
suffered insanely strong intermittent pains, but he would not listen. He had
decided it was not ileus, even though I had not passed gas or stools all day. Yet
in my file, he did not provide any guesses about what I suffered from. I
apologized for canceling the blood tests, and he gave me the business card of a
coordinator I could contact if I did not get better.
The next day, I still had pain but passed some loose stools.
On the third day, I experienced strong abdominal pain during the night and
excruciating attacks of pain with saliva the following morning. I drove to the
surgical department at Hillerød Hospital and told the staff I was convinced I had
ileus and that I was in very bad shape.
My pain attacks were so strong that I was close to fainting. I contacted the staff
several times and asked to be seen by a doctor. I saw several doctors enter my 8-
person room during the morning and overheard a conversation about a man who
was to be transferred to another hospital. I wondered why the doctors took care
of lesser issues before seeing a patient with acute abdominal pain that strongly
indicated ileus. They continued with their minor tasks the entire morning. I felt I
was being punished for canceling the blood tests two days earlier.
I found a chair in the hallway and sat down to demonstrate my existence. I saw
more doctors going to and fro, and I told the nurses that I did not understand
why the doctors would not attend a patient with acute abdominal pain. They
knew I was a doctor but just said that doctors made their own priorities.
Five hours after arrival, my desperation had turned to fury. I called my wife and
we agreed the situation was untenable and that I needed to go to Rigshospitalet.
She called a senior colleague there but he was busy at that time.
I stopped a doctor on his way to a room with other doctors. I hid my rage and
kindly said that they had to see me because I was in a bad condition and
probably had ileus. "A doctor will soon come," said the doctor before walking
away. The doctors knew I was a doctor - my punishment continued.
Half an hour later, a doctor finally arrived. She was convinced I had
gastroenteritis even though I emphasized that extremely violent attacks of pain
were incompatible with that diagnosis. I explained that I had some experience
with gastrointestinal surgical patients and also with infections from my previous
work at hospitals. She wondered why the first doctor gave me a business card to
the coordinator; she could not help me because she was taking care of cancer
patients.
I was told at Hillerød Hospital that I should see my own doctor if I did not
improve. My doctor suspected gallstones and wanted to send me back to
Hillerød for an ultrasound examination. I flatly refused. Under no circumstances
would I go back to a place where they do not take patient complaints seriously. I
will never ever go to that hospital again. Then he sent me to Herlev Hospital
where a doctor agreed to do an ultrasound the following morning, even though
we agreed that gallstones do not cause watery diarrhoea which I had by that
point. I was still convinced I had ileus.
Several times, I remarked how very crippling that operation could be and asked
if less might suffice, because nothing was wrong with my intestine.
I went online and read articles about invagination in adults. The recommendation
everywhere is to remove the intestine in its entirety, which has to do with the
blood supply. Far more than the invaginated part needs to be removed. I was
forewarned that I might wake up with a pouch on my belly for the stools.
The following morning, half an hour before the orderly came, I passed normal
stools for the first time in a week. If I had not been a doctor, I would have been
operated on, but I knew it meant my condition was improving. I therefore left
my bed and told the nurses and Boel about it. The porter came but was asked to
wait because the doctors would discuss my case at their morning conference. A
little later, Boel told me the radiologists had spotted a 5 cm tumour in the
transverse colon, which appeared to be a lipoma causing my ileus. They would
try to remove it through the anus with a coloscope so I could keep my intestine.
The discussion at the morning conference had been lively. Some maintained that
my colon should be taken out, while others wanted to be conservative. I should
have participated in the conference. It was my life and colon they had discussed,
and I was also a colleague. I would immediately have said that, since they were
in doubt, they should be conservative.
There was so much oedema in the intestinal wall that they dared not remove the
tumour immediately. Furthermore, the department did not have a sling big
enough to be passed around the tumour to burn its stem, and they did not want to
cut into it as it might be malignant.
A miracle happened at the very last moment. The pain was gone and the palpable
tumour in my abdomen was gone, which meant that the invagination had
straightened itself out. I was dead hungry and had a real meal for the first time in
a week. The doctors would get a bigger sling from the manufacturer, but the
intestinal wall would need to go back to normal before they could use it. I was
very worried I might suffer a relapse and the colon would have to be removed
anyway. I suggested the textbooks should be rewritten: You should always make
a colonoscopy before deciding to remove a large portion of a healthy colon. Boel
agreed.
They gave me an appointment for tumour removal four weeks later but that was
too long to wait. Immediately after that, I was chairing an important international
meeting in Copenhagen which I had arranged and, if something went wrong, I
would not be able to do so. Therefore, I asked for an appointment three weeks
later.
They could not get the sling around the tumour and instead put two rubber bands
around the stem so it would die and be passed with the stools after 2-5 days.
The pain came back the next day and gradually became worse. I had been told
by one of the doctors that he believed that, if I got ileus again, they could blow
my intestine back the opposite way with a coloscope. That was what I thought
myself. In the evening before I was scheduled for the removal of my colon, I
talked to a senior surgeon and asked if they could not straighten out my colon
instead of removing it. He said no, but maybe they could: Invagination is a very
rare condition in adults and since surgeons have no experience with it, they
choose the most drastic solution because ileus is life-threatening.
It was a fight against the clock. Four days later was the worst. I drank a lot and
ate only soup hoping it would make it easier for the stools to get past the tumour.
I made it just in time. Six days after the rubber bands were put in place, the
tumour came out. It was strange to see a nasty, repulsive piece of yourself that
was no longer part of your body. I washed it and cut through it. It was white
inside but more varied with bleedings on the periphery, and there was also blood
on the surface. I was worried it might be malignant after all, even though a
biopsy had ruled that out four weeks previously. I knew microscopies can
overlook a cancer, and that large colon tumours in people my age were almost
always malignant.
The patient was a journalist and had been so shocked about what he experienced
during his hospitalization that he made secret recordings on his iPhone. Thus, he
could reveal that the doctors changed their explanations to cover up for their
mistakes. 3 The senior surgeon told the journalist in a so-called dialogue-
interview after the diagnosis, that the risk of biliary tract infection increased with
the amount of time needed to reach a diagnosis. Later, the deputy director of the
hospital denied that was the case in a direct interview. And later again, the senior
surgeon changed his diagnosis.
"What was said between the doctor and me completely diverged with what the
management said when I came with an open camera. That shows that they are
trying to adapt their explanations," said the journalist.
The case was reported to the Patient Safety Board, which criticized Hvidovre
Hospital for being too slow initiating relevant investigations. 1 The journalist
received an apology from the hospital director, not for overlooking a life-
threatening infection, only for minor things such as consultations in the
hallways, the lack of adequate cleaning and the multitude of doctors. 1 With a 40
degree Celsius fever and staphylococcal infection in his blood, the journalist
staggered up from his hospital bed and documented that old stools still remained
on the edge of the toilet seat, and that the urine was still on the floor, even
though he had filmed the cleaning assistant leaving the toilet after having ended
her work less than a minute earlier. 3 Unfortunately, that is typical of hospital
administrations. They do not save money on far too expensive medications; they
reduce staff, terminating the most defenseless first - the cleaners.
I explained that ileus requires an urgent operation and that it has a mortality rate
of about 16%: "The longer you have ileus, the worse it gets. The intestine
becomes vulnerable, and if bacteria slip through the intestinal wall and infect the
abdominal cavity, one third of the patients die. So, I was in a very dangerous
condition. I felt very lonely and knew I might die because I've seen many people
die on gastrointestinal surgery and medicine wards. It is not amusing to be in that
situation and, on top of that, feel a lack of medical expertise."
Filing a complaint virtually never leads anywhere but I did so for the sake of
future patients. First, I complained to Hillerød Hospital. I found it remarkable
that, even though I had persistently informed about the incredibly strong pain,
that pain was not described in my record, which indicated quite mild pain.
Nothing was stated about the severity and the doctor indicated pain resembling
"heartburn." At no point in time did I ever mention any reflux or heartburn, so
that was imagined and a colossal understatement of my troubles. Furthermore, I
noted:
"In all the departments where I have worked, we always took the nurses
seriously, and if a nurse said that a patient was in poor condition and should be
seen by a doctor, we always did that, of course. The fact that none of the nurses
asked a doctor to see me - or if they did the doctors ignored their request - is a
testimony of a sick department culture that could be a danger to patients."
"Throughout this entire course of six days, I ate almost nothing. Some days
nothing, other days one piece of crispbread in a single day. That extreme loss of
appetite is also not characteristic of gastroenteritis."
"My very strong and unusual symptoms were not taken seriously at any time,
and they do not even appear in my record. What patients tell their doctors is of
utmost importance and if you listen carefully, you can often make a diagnosis
solely by listening, instead of suppressing any part of the narrative that doesn’t
fit with the tentative diagnosis you have created in your head."
"It's totally unacceptable to let a patient wait for five hours, especially when that
patient is a professional and shows up with an acute abdomen, his wife is also a
professional, and they both think he has ileus. I simply do not understand how
this can happen, and plenty of doctors were at work that Sunday. I have never
experienced anything like this throughout my years as a doctor at University
Hospitals in Copenhagen. It should not be possible. There is a sick culture in the
department which must be radically changed. Allow me to remind you that ileus
is a life-threatening condition."
The journalist with the gallstones had also stated that the whole system is sick. 2
The hospital did not address my criticism. I received an elusive answer stating
that no doubt could be raised about the quality of the treatment I had been
offered!
Patients are very patient and can forgive a great deal if they only get apologies.
But administrators do not understand that. They never apologize for anything
unless they are under severe pressure, as in the case with the journalist. Not a
trace of an apology was found in the letter I received. That is why people get
angry.
Next, I complained to the Patient Safety Board. It took a year and a half to get a
response. The board unequivocally cleared Hillerød Hospital of any wrongdoing.
It was only then that I realized the serious errors in my records, because the
hospital sent them along with its report to the board.
For instance, the doctor and I allegedly did not find "any great suspicion of a
fulminant ileus condition, and hence we agree" that I could be sent home. That
was not true. At no point in time did I dispose of my suspicion of ileus, and
certainly not during my first visit to the hospital.
The worst part was a statement from Henrik Stig Jørgensen, the chief surgeon at
the department. He was very curious as to why a senior surgeon at Rigshospitalet
(the National Hospital) recommended an acute CT scan after only speaking to
me on the phone and without an examination, and also how "a doctor at
Rigshospitalet can make a diagnosis over the phone, which a doctor couldn’t
make at a physical examination” at Hillerød Hospital.
These remarks smell of professional jealousy. You should not think you are
important just because you work at Rigshospitalet, the country's finest hospital.
Further, contrary to Jørgensen’s claim, the surgeon at Rigshospitalet did not
make a diagnosis but recommended a highly relevant diagnostic test. Jørgensen
also ignored the fact that my wife and I are doctors, that we had made the
physical examination Jørgensen wanted, and that we had agreed I most likely
had ileus. Finally, Jørgensen ignored the fact that virtually nothing in this world
is constant and, therefore, may have looked differently when I arrived at Hillerød
Hospital.
That was a lie. My acute condition was not a priority. Furthermore, I could not
recognize Jørgensen’s remark that, “It is the senior surgeon’s clear perception
that they agreed there was no sign of serious illness." The surgeon wrote in my
records that she thought I have gastroenteritis. But even though I might have
admitted that this could be a very remote possibility, I considered my illness
serious and I was still convinced that I had ileus. Technically, it was not ileus but
subileus, because that condition sometimes improves with the passage of stools.
But that is not relevant for this narrative. If I could have decided myself, I would
have ordered an acute CT scan but, despite my experience, I was at the mercy of
other people. Ironically, one of the first diagnoses I ever made as a medical
student was ileus in an old lady who complained of stomach pain. I made that
tentative diagnosis with my hands and stethoscope. The chief physician was so
impressed, he offered me a job when I had passed my medical exams.
Jørgensen ended his letter by saying that the department’s leadership in no way
believes that the department had a sick culture. Evaluating yourself is not
credible and in fact, Jørgensen demonstrated the opposite in his letter. It takes a
sick culture to trample on a patient who is already lying down and afraid of the
consequences of the serious illness he had diagnosed in himself but which the
department ignored. And it is blatant contempt for patients that the head of the
department deliberately chose to refer to the victim as “Professor, chief
physician, DrMedSci and MSc Peter Christian Gøtzsche.”
Doctors must never modify what the patient says to better fit with what the
doctors believe their patients to be suffering from, and they must never write
symptoms that do not exist in the patient records, which happened in my case. If
that happens in science, it is called fraud.
We cannot trust patient records. They have been changed when physicians got in
trouble many times, even though that kind of fraud is punishable by law.
Therefore, it can be a good idea to ask for the doctor’s notes early on, so that you
can request amendments if they are incorrect or do not reflect your condition
well. In my case, the false information about heartburn was repeated by the
Patient Safety Board when it cleared Hillerød Hospital for any wrongdoing.
Patients can request a copy of their records, but they should also have the right to
approve what is being stated in their records.
Most were Greeks. The economy is still poor in Greece, so these are not society's
losers - you meet the winners at a luxury hotel. They know overweight increases
the risk of complications and premature death. I assume they have some self-
discipline because they have managed their lives well, and they choose to
become obese because it gives them an increase in quality of life that outweighs
the increased health risk.
Being obese is not a disease, yet obesity increases the risk for hypertension,
heart disease and diabetes. If obese people visit their doctors for some other
reason, doctors might perform additional investigations and prescribe three or
more drugs to lower blood pressure, cholesterol and blood sugar.
In Denmark, these drugs are subsidized. Yet is it reasonable that taxpayers pay
for them? Some people become overweight because they have difficult lives and
comfort themselves by eating too much; others eat unrestrainedly without
thinking about it; and still others are obese because they take psychotropic drugs.
Yet most obese people have no excuses - they have simply chosen to eat more
than needed.
Should we subsidize drugs that only treat risk factors, not diseases? These drugs
lower the risk a little of untoward things happening, and they cause a great deal
of harm. We all have risk factors for all sorts of issues, so the drug industry
diligently and endlessly turns the screws. We should also consider that the drug
industry, assisted by corrupt doctors on industry payrolls, is continually lowering
the bar for what is considered normal – the result is selling sickness to healthy
people. We should raise that bar a little and only subsidize drugs when the risks
are considerable, for instance, when patients have a markedly elevated blood
pressure or hereditary hypercholesterolaemia, and not if it is a matter of a chosen
lifestyle.
When people lose weight, many of them can stop taking drugs. All doctors know
that weight loss is a highly effective intervention for many issues, e.g. not only
for hypertension and diabetes, but even pain due to osteoarthritis. The main
reason why we do not take this more seriously is because it is difficult to
maintain weight loss. Further, it takes time and effort to get there. It is far easier
and more lucrative for doctors to hand out slimming pills to overweight people.
But do not take them - they are far too dangerous (see below).
In 2017, I lectured at another weekend course in the United States where the
organizer has also taken an interest in nutrition. I think most of the 200
participants were vegans. The food served was vegan, just like when I visited
McDougall. Some people described only eating plants as similar to eating dirt
but, in time, they got used to it. It was difficult for me to get through my two US
weekends on vegan food. When a participant told me that, “I’ve been a vegan for
40 years,” I could not help responding, “And I’ve eaten everything for 67!” One
morning, I told the entire gathering how much I enjoyed my two fried eggs,
bacon and a sausage for breakfast - everything forbidden for vegans. I also told
them they would need to show me randomized trials demonstrating that
veganism extended life by at least 5 or 10 years, if I were to ever consider giving
up eating what I liked.
I heard stories about considerable weight losses on a vegan diet which I found
plausible because of how uninspiring it is to only eat plants. I have also heard
about people being cured of long-lasting inflammatory bowel disease. I find it
likely that a change in diet may induce changes in gut bacteria that might affect
various diseases, but I urged participants to publish their findings and to do
research on these issues (in particular randomized trials). One of my friends had
inflammatory bowel disease for 35 years, yet he improved over time and now
has long periods - up to a couple of years - without symptoms. If he had changed
to a vegan diet when he was destined to improve, he might have been convinced
it was the diet that did it and not just nature’s spontaneous healing.
Vegans do not get the micronutrients they need. A lack of vitamin B12 is most
commonly mentioned but more than that is lacking. You can find out by
googling vegan supplements recommendations.
I fail to understand why so many vegans resort to excesses. Why is the beer
tasteless in vegan restaurants (if they serve beer at all)? Because the alcohol has
been removed through a chemical process. That is not natural and has nothing to
do with being vegan. And why is only ice water offered, both for lunch and
dinner? Ice water is for polar bears to swim in. I went to the front desk to get a
beer for lunch, but I could not get one, even though I was staying at the Hilton,
because the bar opened at 5 p.m. I then asked where in town I could buy a beer
and was transported in the hotel bus to a German bar where I bought a bottle of
dark ale, which made it easier for me to get through the vegan lunch. I always
have a beer for lunch on weekends.
Dieting is a big industry, and the advice is often confusing and conflicting. So,
what would you do if you wanted to lose 6 kilos? I decided to make no changes
to my diet other than eating less and exercising more. My approach was
scientific and very simple. You can easily find out online what your basic
metabolism is; what the calorie contents are for various foods and alcohol; how
many calories you use when exercising; and the number of calories in fatty
tissue. That is all you need for a spreadsheet. When you read about calories, it is
actually referring to kilocalories, even though people call them calories. I
weighed my food every day, exercised most days, and the spreadsheet told me
how to obtain the daily weight loss of 200 g I had planned.
Weight varies with your hydration level, but you can still get a fairly accurate
idea of how well you are doing. You are your own merciless judge when you get
on the scale each morning - succeeding becomes a kind of sport.
Certain food items I did not perceive as fattening have a surprisingly high calorie
count, e.g. sweet biscuits which I stopped buying to avoid the temptation. When
my wife baked a cake, I took a piece half as large as usual and enjoyed that small
piece a little extra. If I felt hungry enough in the evening that my stomach hurt
and cried out for food, I did not destroy my whole scheme by going bananas. I
simply took a thin slice of bread with cheese - not that many calories.
Three weeks later, after losing 5 kilos, it got difficult. I could not understand
why but found out online why it is difficult to continue losing weight. When you
get slimmer, your basic metabolism drops and you spend fewer calories when
exercising. If you exercise a lot, your muscle weight will increase. So, you will
need to adjust the values on your spreadsheet. Then you have to get very
determined. My weight loss slowed, and it took another four weeks to lose 3 kg
more. I had not been that slim in 30 years.
You have to be constantly on guard. A year later, I had regained 4 kilos and
needed another round, this time without the spreadsheet because I had learned
what I needed to do.
It is difficult, but this is the best way to make progress. If you avoid overeating,
which for the most part is just a bad habit, you can very likely maintain your
quality of life without ever needing drugs. Yet as we get older, it is surprising
how little food we need, even if we run many kilometers in the forest every day.
Instead of offering drugs, we should offer free courses on better - and less –
nutrition. If that does not help, and the patients want to take drugs for their
obesity-induced risk factors, they should pay themselves. These drugs reduce the
quality of life for many people. They can make you tired and cause muscle pain;
they reduce sex drive, create impotence and many other conditions people do not
often associate with the medications they take but believe to be due to advancing
age.
Stating these facts is kind of taboo. I am not on Facebook, but when I wrote
about these issues in a newspaper, 5 I was told hundreds of hateful comments
appeared on Facebook in response. And when I was a guest at a golden wedding
anniversary, a colleague immediately verbally attacked me saying he hoped my
wife drove our car. Why? Because if I crashed, I should have to pay for the
hospital costs myself! Convincing him my article could not be interpreted like
that was impossible. If we fall ill, of course we should be treated without cost -
at least in my country - whether we are overweight or not. And if a smoker gets
lung cancer, we treat the cancer for free. It is incredible how emotions can
prevent people from thinking clearly.
A big culprit in the obesity epidemic is sugar. In the 1960’s, the US Sugar
Research Foundation wanted to "refute" concerns about sugar's role in heart
disease and shift the blame to fatty foods. The foundation sponsored research by
Harvard scientists who published their review in the New England Journal of
Medicine in 1967 without disclosing sugar industry funding. 6 The foundation
‘cherry-picked’ studies for the researchers to look at, and then they applied
different standards to different studies. They very critically looked at research
implicating sugar as a cause of heart disease and ignored problems in studies that
found dangers in fatty foods.
The Harvard researchers then turned to studies that examined the risks of fat,
which included the same kind of epidemiological studies they had dismissed
when it came to sugar. They cited few study characteristics and no quantitative
results and concluded that cutting out fat was "no doubt" the best dietary
intervention to prevent coronary heart disease.
Harvard University is in Boston, which is also home to the New England Journal
of Medicine. Once again, that medical journal was very harmful to public health.
In the 1970’s, the US Sugar Association convinced the FDA to issue a statement
that sugar was not hazardous to health. 7 The chair of the FDA’s committee was
also the chair of that association. And the corruption continues to this day. In
2015, the New York Times obtained emails revealing Coca-Cola's cozy
relationships with sponsored researchers who conducted studies intent on
trivializing the effects of sugary drinks on obesity. 6 The Associated Press
obtained emails showing how a candy trade association funded and influenced
studies to show that children who eat sweets have healthier body weights than
those who do not. 6 The sugar industry’s tactics are very similar to the organized
denial the tobacco industry used, and heart and cancer associations were
corrupted by sugar industry money.
In Denmark, Arne Astrup, the most renowned expert on nutrition, did a trial
funded in 1990-94 but not published until 2002. 8 The giant Danish sugar
producer, Danisco Sugar, was one of the funders, and the trial showed that
sugary drinks cause weight gain. In 1997, Astrup wrote a leaflet on behalf of
Danisco where he claimed that sugar is not converted into fat, even though it had
been common biochemical knowledge - for decades - that sugar can convert into
fat. 9 Astrup has also been a consultant for Coca-Cola and was the chair of the
Danish Council for Nutrition, a governmental agency. In 2015, it came out that
Astrup was on the governing board of the Global Energy Balance Network, an
international research network that has received many millions in support from
Coca-Cola, 10 one of the most harmful companies in the world. This network
claims that sugary drinks and fast food are not the main reasons for overweight!
Soft drinks have never been allowed in our home, only freshly pressed fruit
juice. Our children never missed them because we did not allow them to get
addicted to Coca-Cola and other soft drinks. Most of the time, we also avoided
buying candy and chocolate.
Sugar consists of equal amounts of glucose and fructose, and both substances
can be converted into fat. When a slim young man began eating the rich, sugary
diet many people eat, including breakfast cereals, yoghurt and ice tea, he gained
8.5 kg in two months - even though he had not eaten more calories than usual
and did not touch soft drinks, ice cream or chocolate. 7 A calorie is not just a
calorie; it depends on where it comes from. When he switched back to his
normal diet, the kilos vanished from his belly, but he did have to endure a period
with abstinence symptoms.
To increase sales, the food industry has conducted experiments to find out what
the optimal amount of sweetness should be for sugary drinks and food items. 7
Coca-Cola has been very effective at selling their products to the world’s most
rural and impoverished areas. In 2008, the company said that the Northern
Territory in Australia was the best-selling region in the world per inhabitant. 7
Many aboriginals there die young from diabetes and kidney failure. Coca-Cola
must have shortened the lives of millions of people.
Cochrane reviews about diets abound. If you search for diet in the Cochrane
Library, you get a list of 280 Cochrane reviews which can be quickly browsed
since most - by far - deal with specific diseases and not with prevention.
A TV documentary called The World’s Best Diet explored some of the world’s
most varied diets. The Marshall Islands were at the bottom: the highest rate of
diabetes associated deaths in the world and one of the most overweight
populations as well. Their inhabitants eat mostly canned foods from the US, and
even eat canned vegetables, because they are cheaper than fresh produce.
Surprisingly, they do not eat much fish, preferring turkey tails from the states
containing 73% fat.
Mexico suffers immensely from the free trade agreement the country made with
the US. The population consumes a great deal of soft drinks and is highly obese.
The US is placed far down this ranking list, of course.
The top end of the list is equally interesting. Here we find the Nordic countries,
France, Italy and Spain, for example. The small village of Campodimele in Italy
boasts an average lifespan of 95 years! The inhabitants have their own small
vegetable gardens, they make use of olive oil, eat chicken and indulge in very
little beef. France is particularly interesting because its inhabitants defy the usual
dietary advice. Fatty cheeses, fat poultry and beef are enjoyed, yet they still have
a long life expectancy.
Exercise is good for so many things that there is no doubt that we should all
exercise. Before one of my meetings at the Danish Ministry of Health, the guard
at the entrance pointed out the elevator to me. I asked him, “Isn’t this the
Ministry of Health?” I told the guard that elevators are only for handicapped
people and deliveries. I always use the stairs, even up to the 14th floor at my
hospital - two steps at a time. I will consider myself very old when I cannot do
this any longer.
Yet, there are harms. When you get old, you do not pick up your feet as much
when running, which means you might stumble on overlooked stones and tree
roots. It is also easier to fall off your bike when a dog suddenly runs across your
path. I have decided to take that risk, because I find it extremely boring to run on
a treadmill in fitness centres. If you run 8 km (5 miles) in the forest, the
landscape changes all the time, and even though you are familiar with it, it does
not feel like a long way. Eight kilometers on a treadmill feels like a half
marathon - and you do not get to hear any birds or meet people, dogs and horses.
If exercise becomes part of your life, you might choose to accept the risk of
falling because you like what you are doing - which increases your quality of
life.
What about advice coming from our national boards of health? You should be
skeptical about such advice, too. First, institutions always think in utilitarian
terms: “If the whole population did so and so, we would expect to save so and so
many lives.” But you should think of yourself. Second, very little reliable
evidence is at hand about the effects of the suggested dietary changes because
very few randomized trials have been done. Here is an example:
In 2001, the Danish National Board of Health launched the Six a Day campaign.
I thought it meant that we all should eat six pieces of fruit every day, which I
found impossible. Some days, I did not even eat one simply because I forgot. It
turned out vegetables also counted, so you could eat, for instance, three apples,
one large tomato, one large carrot and a banana. Still quite a challenge.
I am not obsessive with what I eat, and I also thought that our authorities could
not possibly have any firm evidence backing up these recommendations. Our
board of health has also advised men to limit their alcohol intake to a maximum
of three drinks a day, and only two for women.
Your first question when you hear of such advice should always be: Was this
advice based on randomized trials? Obviously not, because it would not be
feasible to randomize people to varying levels of fruit and alcohol intake for
years at a time.
Observational studies are fraught with difficulties. People who eat little fruit and
vegetables, or drink more than others, cannot be compared to vegetarians and
teetotalers. They differ from them in all sorts of ways that could influence their
longevity. If we are to rely on observational evidence, high quality research is
required and the subsequent decrease in mortality must be substantial.
Otherwise, any such decrease can be explained by bias. But just what does
substantial mean? Many respected epidemiologists have published erroneous
results stating that, because of how easy it is to be fooled, any less than stunning
results are almost impossible to believe. 16 Some have stated that even a
threefold risk increase is not persuasive, and persuasiveness is only achieved if
the lower limit of the 95% confidence level falls above a threefold increased risk
(which means that we are 95% certain that the true risk is increased at least
threefold).
Let us find a few examples that might clarify this. I searched for doubles the risk
in the titles of abstracts on PubMed. The most recent article was: Alcohol intake
more than doubles the risk of early cardiovascular events in young hypertensive
smokers. 17 Sounds pretty scary. Even if you are not young, not hypertensive and
do not smoke, you might nevertheless consider giving up alcohol. You might
deduce that if alcohol is so harmful to people with these characteristics, it should
also be expected to be harmful to people like yourself.
This is the type of study that continually makes headlines in the news, and which
is constantly contradicted by other research a little later, creating great confusion
and leading some people to distrust any dietary advice at all.
What was the real risk? In their abstract, the authors write that, in a multivariable
model also including follow-up changes in blood pressure and body weight, the
hazard ratio was 1.48 (95% confidence interval 1.20 to 1.83) for smoking and
1.82 (1.05 to 3.15) for alcohol use.
Then I got highly skeptical and dismissed the findings entirely. The authors
found that the risk of cardiovascular events from drinking alcohol was greater
than the risk from smoking - which I find hard to believe. Furthermore, the lower
limit of the 95% confidence interval came very close to 1.00, which means zero
risk increase.
Another study on PubMed was not about diets, but I want to mention it because
it is fairly typical of a great deal of medical research literature. The study
examined whether excessive infant crying predicted emotional and behavioural
problems at age of 5-6 years. 18 Excessive infant crying was associated with a
twofold increased risk of overall problem behaviour, conduct problems,
hyperactivity, and mood problems. The odds ratios (which are very similar to the
risk ratios when event rates are low) varied between 1.75 (95 % CI 1.09 to 2.81)
and 2.12 (1.30 to 3.46).
When I started reading the abstract, my first thought was, “Of course, there will
be an increased risk of problems later in life when infants cry excessively. It is
not unreasonable to expect that babies who are difficult as infants could also be
difficult five years later; it could be in their genes. Or perhaps it was the mother
who was difficult, both times.”
Furthermore, it can be very stressful for the mothers of crying babies. These
mothers cannot be expected to be neutral observers five years later. Therefore, I
looked up the entire paper and found out that the study relied on these mothers’
evaluations of their children. Already at this point, I would have normally
skipped reading. And even with these biased evaluations, the odds ratios were
not at all impressive.
Yet, I was curious about what the authors wrote in their conclusion, because I
would not conclude a thing based on a study like this. They wrote that special
care for mothers with a high burden of care for their excessively crying infants
can be a feasible strategy for possible prevention of mood and behavioural
problems in their children later in life. That is really far-fetched, yet typical of all
the many do-gooders in healthcare. It makes you look good to suggest helping
people with difficulties, nonetheless, everything comes with a cost, and we
should not suggest interventions if we do not know they will work.
Getting back to the diets, the search for subtle links between diet, lifestyle or
environmental factors and disease is an endless source of fear offering little
certainty. 16 I do not know on what basis the Six a Day campaign was launched.
Yet by googling it (in Danish), and allowing Google to suggest various options
while I was typing, I quickly found a large meta-analysis of 95 studies from
2017. 19 It reported that, for fruits and vegetables combined, the relative risk for
all-cause mortality was 0.90 (95% CI 0.87 to 0.93) if the intake was 200 g per
day (corresponding to two apples). It also noted that reductions in risk were
observed up to 800 g/day for which the reduction in mortality was estimated to
be 31%.
The authors concluded: “An estimated 5.6 and 7.8 million premature deaths
worldwide in 2013 may be attributable to a fruit and vegetable intake below 500
and 800 g/day, respectively, if the observed associations are causal.”
But we do not know if the observed associations are causal or just bias. What
was reported was a 31% reduction in mortality if you eat eight apples a day -
which I would find really stressful. If you turn this around: 0.69 to 1/0.69 = 1.45
is nowhere near a lower limit of the 95% confidence level falling above a
threefold increased risk if you do not eat eight apples a day.
Coffee
Coffee must be very healthy since it has been subjected to numerous
observational studies and has survived these attacks despite the unavoidable
biases in such research. For example, if you are stressed, you might drink more
coffee and be at greater risk for cardiovascular disease; and more smokers are
also found among coffee drinkers. Even if you have adjusted for this, you might
not have adjusted enough and, therefore, it would seem that coffee causes the
same illnesses as smoking.
If you google coffee is healthy, you will find many interesting studies. A large
study reported that coffee increases the risk of death, but that was before it had
been adjusted for smoking. 20 Actually, the more cups we drink a day, the more
we will lower our mortality risk. Yet these differences are small, e.g. a 10%
reduction in mortality for six cups. Thus, you might argue that we should not pay
attention to this finding - and I agree. But the interesting thing about coffee is
that there are so many studies, with very varied designs and therefore not the
same biases, that have shown that coffee doesn’t kill you. A study presented at a
cardiology congress in 2017 found that those who consumed at least four cups of
coffee per day had a 64% lower risk of all-cause mortality than those who never
or almost never consumed coffee (adjusted hazard ratio 0.36 (0.19 to 0.70). 21
I have never understood why so many Americans ask for decaffeinated coffee
when it seems that coffee cannot harm you. In a recent, huge study of 58,397
deaths, the dose-response relationship was similar for caffeinated and
decaffeinated coffee, and already at 2-3 cups a day, the mortality was 18% lower
than for non-drinkers. 22
Enjoy your coffee! That is why we drink it. We do not drink it because we think
it will make us live longer.
Slimming pills
Don’t ask your doctor whether they are right for you. Just don’t take them. There
is no quick fix for overweight. Most slimming pills were taken off the market
after killing many people. Sometimes these deaths were preceded by periods of
horrible suffering where people felt they were being slowly suffocated or that
they were drowning. 12 A couple of years ago, we reviewed the only pill that was
still on the market in my country - it did not look good, either. 23 We examined
the clinical study reports of orlistat which we had obtained from the European
regulator, EMA. We identified important disparities in the reporting of adverse
events between protocols, clinical study reports, and published papers which had
systematically understated adverse events. This drug is far more bothersome
than what seems to be the case if published trial reports are studied.
One of the drugs that was taken off the market was praised in an editorial in New
England Journal of Medicine. 12,24 Not a word was mentioned in the editorial
about the fact that its two authors were paid consultants for the companies that
sold such drugs. They said that the risk of pulmonary hypertension was small
and outweighed by the benefits of the drug. However, the benefit was a mere 3%
weight loss, as stated by the company, e.g. from 100 kg to 97 kg. 12 Furthermore,
many patients dropped out of the trials and the conventional statistical method
companies employ is to carry forward the last recorded weight until the end of
the trial. Since much of the weight people lose in the beginning comes back later,
that method is flawed. In our own study of a slimming pill, we showed that the
last observation carried forward yielded a weight loss of 6.4 kg above placebo
while baseline carried forward showed a benefit of only 1.5 kg. 25
There was no doubt about the diagnosis. My sudden bouts of loud snoring
sounded like the roar of a ferocious animal which not only woke up my wife but
even myself. I was sometimes exhausted and sweaty during the day, and
experienced irresistible urges to sleep at the most inconvenient times, such as
when driving a car or during dinner parties. The urge to sleep could be so intense
that I had to leave a party to take a nap, excusing myself with not feeling well,
which was true and sounded better than saying I needed a nap.
Throughout my adult life, I have fallen asleep suddenly at times, and in 2009, I
gave in and visited an ear, nose, and throat specialist. He handed out some
equipment to monitor my sleeping pattern and I discovered I had periods when I
did not breathe for an entire minute. The specialist suggested removing the uvula
and possibly other tissues, but I would not - under any circumstances - accept
this surgery which was not only irreversible but could even be harmful. I told
him no data existed from randomized trials showing that such surgery worked.
2,3 The surgeon then suggested using continuous positive airway pressure
(CPAP), which I accepted not knowing that its benefits in mild to moderate sleep
apnoea were inconclusive. 3 He referred me to a sleep centre, and I asked him to
send a copy of my file containing the sleep recordings to the centre so they
would be there when I arrived.
When the sleep specialist wanted to make more sleep recordings, I told him that
new recordings were unnecessary - the diagnosis was indisputable. It turned out
he had not received the recordings. I insisted he requested them rather than
subjecting me to superfluous testing.
The missing recordings were not important. The specialist unpacked a CPAP set
and explained how to use it. He mentioned that I could get a special permit
allowing me to carry the apparatus as hand luggage when I was flying. I
explained that I did not intend on carrying it with me, because I did not see
myself as a desperately ill patient and did not want the pity of onlookers.
The specialist explained that, after being turned on, it took 20 minutes for the
apparatus to start working, allowing people to fall asleep. He also said that if the
inhalation pressure was too high, I could adjust it, but he did not tell me how.
Back home, I unpacked the CPAP apparatus with great trepidation, feeling badly
about my new role as an intensive care patient. I usually fall asleep immediately
but, lying there with my face mask, I just waited until the apparatus started
working. It inflated me like a balloon. It was very unpleasant, and after a while,
my throat dried out. I consulted the instruction manual, which was
approximately a hundred pages, but I could not find any description of one of the
most essential functions: how to reduce the pressure. I gave up, dismantled the
machine, and fell asleep.
The next evening, my wife convinced me to give it another try, yet the same
sequence of events unfolded.
On the third day, I read the instruction manual more carefully and found a page
which curiously, considering its dire text, was not located at the beginning of the
manual. Under no circumstances should anyone use this apparatus without
reading that crucial page! Well, how clever of you to hide it then. Reading on, I
discovered that there was no guarantee that the apparatus would not kill me. If it
malfunctioned, I might rebreathe the same air and die peacefully without
triggering any alarms, and without any intensive care nurse rushing to my
rescue. Startled by this, I searched PubMed and the Internet but found nothing
about the risk of this lethal complication.
The inconvenience of my condition did not justify any risk of dying from the
treatment. I consulted the research literature before seeing the surgeon again,
because I knew some observational studies indicated that sleep apnoea increased
the risk of cardiac disease. Yet so many things in life increase that risk, and
observational studies are often misleading. Prolonged oxygen deficiency is
definitely not good for the heart, but I did not include that in my private
decision-making analysis.
At my next visit to the sleep centre, the specialist asked me how it went. I told
him about my experiences and reflections - and then returned the equipment. He
had received the recordings from the surgeon and, to my surprise, he said that,
since my sleep apnoea was mild, he would not have recommended the apparatus
in the first place! He continued to surprise me because he seemed to base his
judgment on these recordings rather than on my symptoms - which were fairly
pronounced. He also said that CPAP was only considered to work at certain ages
and was not recommended after age 70. Approaching 60, I replied that I saw no
reason to use this terrible apparatus which might even kill me.
When I told my story to my two friends, one of them remarked that most patients
cannot tolerate using CPAP for sleep apnoea. He sent me a paper that reported
that only 2 of 35 patients used CPAP for seven hours for at least 70% of the
nights. 4 That would have been nice to know beforehand. Furthermore, if I had
known about a BMJ review that questioned the effect of CPAP in mild to
moderate sleep apnoea, I would not have consulted a doctor in the first place. 3
My short guest visit as a patient was alarming. Worst of all, I lost my autonomy
to the doctors who told me what to do. I dropped the patient role, because I am
not ‘patient’ enough to consult doctors unless I am really ill. When I wrote the
article eight years ago, what struck me most was the fact that a doctor like me,
who did research and was used to searching for evidence, was very privileged in
comparison to the vast majority of patients who cannot do so, and therefore, just
have to hope their doctors know best. That is grossly unfair. I have tried to make
up for this injustice by writing this book.
This story, and many others similar to the ones I described above, demonstrate
that too much of what takes place in healthcare can be characterized as knee-jerk
reactions: The doctor has a hammer and sees a nail, and hammers it. What if it is
actually a screw? The doctor still uses his hammer because he never learned to
use a screw driver.
For his part, the specialist wrote that obstructive sleep apnoea occurs in up to
24% of men and 9% of women. 5 It seems to me that anything becomes common
when left to specialists to decide. My specialist also wrote that surgery could be
used in some cases, but I could not find any evidence in support of this.
I suffer much less from sleep apnoea and from cardiac problems today than I did
some years ago. That is an important take-home message. Not everything goes
downhill when you get older. And mind you, I did not take any drugs or had any
surgery performed, both of which I declined. My friend with inflammatory
bowel disease also got much better in time - without any treatment. And even
schizophrenia – and many other ailments - can disappear or improve
dramatically.
Non-diseases
In 2014, at a meeting in California, a woman asked me my opinion of “low
testosterone.” I asked her to explain what she meant, because I had not heard of
any such disorder - great laughter ensued around the table. It turned out that her
husband was impotent, and his wife hoped testosterone tablets would straighten
him out (so to speak). It was a big issue in the United States, yet I had never
heard of anyone talking or writing about it in Denmark.
There are loads of ‘non-diseases.’ It is not enough for the drug industry to sell
pills to those who are ill, and even those who risk falling ill - which are all of us.
The industry also invents many non-diseases which always seem to be serious, at
least to some people. Here are a few other examples.
Psychiatry is full of non-diseases, and ADHD is one of them. 6 This is not to say
that people do not have problems, but the issues are best treated with
psychotherapy and other psychosocial interventions - without being labeled
diseases and without drugs.
Citalopram, a depression pill, has been tested for compulsive shopping disorder.
Good Morning America told TV viewers that this new disorder could affect as
many as 20 million Americans, 90% of which were women. 7,8 Escitalopram
from Lundbeck, which contains the same active substance as in citalopram, has
been tested for hot flashes during menopause; 9 fluoxetine is approved in the US
for premenstrual dysphoric disorder; 6 and depression pills are used for many
other non-diseases, including marital problems, bullying at work, and stress and
worry about an upcoming exam.
Pre-hypertension is the same kind of bluff. If you lower the blood pressure, you
lower the blood pressure - so what?
What about birth, the most dangerous of all events? Birth causes a 100% risk of
dying from the day we are born, so we are all pre-dying. What should we do
about that?
Old age
We especially suffer from pre-dying when we get old. But old age is a non-
disease, so please leave old people alone. If we are lucky enough to get this far
in life, and we can still stand on our feet and take care of ourselves, it is time for
doctors and other do-gooders to stop interfering with our lives. Young and old,
we treasure our independence. In Greek mythology, it was considered good
fortune when a young man was killed in a battle at the height of his
achievements, because if he had survived, his life could only go downhill.
Perhaps a bit exaggerated, but you know what I mean. I would rather die
somewhat early with my boots on, than dying in a nursing home - demented and
wearing diapers. Who would not? To be alive is about living, not about being a
living dead.
The other day I talked to an American colleague whose 84-year old father was
still attending annual health checks. I told her it was pointless and asked if she
knew that going to health checks was harmful? She did, but as is often the case,
the old man listened more to his doctor than to his own daughter, even though
she was well informed. In my country, the Danish National Board of Health
started a campaign to remind women about getting smears to prevent cervical
cancer. The campaign is aimed at every woman up to 100 years old.
What does getting old mean today? It means taking many pills because our
clinical guidelines are written with tunnel vision, peering at one problem at a
time and not taking the overall picture into account. What does it mean to be
treated for a multitude of risk factors and minor ailments with countless drugs,
many of which do not work? It means an increased risk of death - the more drugs
you take, the greater the risk. Most drugs affect brain functions, and when old
people fall and break their hip, one-fifth will die within a year. Do not forget
what I wrote at the beginning of this book: Our drugs are the third leading cause
of death. Therefore, we need to do trials which compare taking many drugs
simultaneously with taking none.
Leaving old people alone is difficult for doctors - and that seems to be getting
worse. In Korea, for instance, the proportion of older patients receiving
chemotherapy in the last month of life increased from 26% in 2000, to 33% in
2005 and to 44% in 2010. 11
Dementia drugs
Don’t take them. They don’t work. 6 The trivial effects that have been noted in
industry-sponsored, placebo-controlled trials on some rating scales could easily
have been caused by a lack of blinding, because the drugs have conspicuous side
effects. Furthermore, even if the measured effects were true, they were too small
to be clinically relevant. And the drugs cause many harms. Also, using common
sense is not forbidden. How likely is it that a drug will slow down degenerative
processes in our brains? Close to zero. How likely is it that the trials are biased?
Close to 100%. How likely is it then that the small effects measured in these
trials are just bias? Extremely likely.
I told him about natural variation. It is a purely statistical phenomenon that not
all patients will have the same value on a rating scale after treatment, and if the
trial is repeated in the same patients, the next time, other patients will seem to
respond better.
Imagine your old car is causing problems and you take it to a mechanic. He says
he will try to fix it with some new method but adds that he cannot guarantee he
will succeed. You ask what the results of this method have been on other cars.
He explains that, on average, it did not work but sometimes it seemed to work a
little better for some cars than for others. I think you would look at him in
disbelief and drive away with your defective car. After all, your car is not totally
dead yet and can still transport you where you want to go.
Unfortunately, doctors know very little about statistics and think very little about
statistical variation - which cannot be taken to mean that their treatments work.
Instead, they emphasize their own clinical experience even though it is greatly
misleading. They start patients on drugs that do not work - like dementia drugs
or psychiatric drugs, for instance - to see how that goes. They think they can
differentiate between patients responding to a drug and patients who do not. This
is impossible because of the natural variation in the intensity of the disease and
because they have nothing to compare with. Would the patients have fared better
without treatment? No one knows. That is why we must treat patients based on
what the most reliable trials tell us about benefits and harms after we have taken
all sorts of biases into account. That means that no one should be treated with
dementia drugs - absolutely no one. 6
One of the most popular dementia drugs is donepezil (Aricept). The most
common adverse effects of donepezil are nausea, diarrhoea, disrupted sleep,
vomiting, muscle cramps, tiredness, and not wanting to eat. 6 Not exactly what
we would want for an old person who might already have problems with
sleeping, feeling tired, and lack of appetite. The list of frequent harms - called
side effects in Pfizer’s product information for Aricept - is very long.
Hypotension and syncope occurs in more than 1% and, as already noted, when
old people fall and break their hips, many of them die. What a drug.
A large Canadian study showed that people who took drugs against dementia
almost doubled their risk of hospital admission due to fainting compared to those
who did not take drugs. And they broke their hips more frequently and even had
pacemakers inserted more often. 12 Astonishingly, more than half the patients
who were admitted to hospital with a pulse that was too slow (bradycardia) were
re-treated with the same type of drug after discharge. Overvalued clinical
experience certainly did not have the expected and desired effects in these cases.
Doctors should see themselves as psychosocial workers and not as pill pushers.
A great deal can be done - without pills - to help our demented people and other
patients. 6
We take so many pills that every one of us could be in daily treatment from the
cradle to the grave. The older you get and the more you see a doctor, the more
pills you get. Doctors are not good at ‘un-prescribing’ pills again and, if a
specialist physician ends your course of drug treatments during hospitalization,
your family doctor will often reinstate every one of them again. Conversely, if
your doctor takes away a pill that a specialist put you on, the specialist will often
resume that treatment upon your next visit, even though your doctor knows you
far better and, therefore, often knows what would be best for you. 13
If you are old, try to keep the doctors away. You are much more likely to need
social workers. And if you are on drugs, try tapering off slowly, one by one. Of
course, there are exceptions, when you really do need the drugs you are on, yet
most often you are better without them. If the issue is not you, but rather your
old mother or father, then try to help them. They might feel years younger if they
get off some of their drugs.
A slow withdrawal is often necessary to avoid abstinence symptoms because
your body adapts to the drugs you take. You may then find out that your fatigue,
impotence, muscle pain or memory loss are not signs of getting old, they are just
the adverse effects of drugs. You should also remember that medicine, beneficial
to you when you are young, might be harmful when you grow older because old
people tolerate drugs more poorly than young people. And perhaps your body
has healed itself so much that you no longer need drugs.
Most patients respect authority and would not dream of cutting back on their
medications without their doctor’s go ahead. However, since doctors far too
often advise people to continue with medications indefinitely, some patients
decide to do otherwise. It can be risky to take action on your own, but it is far
riskier for us to continue our current, massively high drug consumption. Many of
those who die could very well have avoided the medication that killed them.
I get many emails from patients who describe what happened to them when they
decided to taper off the drugs themselves. Here is an example, in a truncated
version. 13
"My 67-year-old life was characterized by physical and mental health problems
to such an extent that I asked myself if it was worth living at all. Your book
meant that I have skipped my cholesterol medicine, half of my antihypertensive
medicine and a few other preparations. I have changed my diet somewhat and I
have switched from a doctor who did not bother listening to the side effects to
one who has no involvement with any drug manufacturers. My blood pressure is
even lower than before I started treatment.”
“Most surprisingly, not only did I reduce the number of physical problems, but
even the psychological problems I had accepted I had to live with totally
disappeared a few days after I dropped my cholesterol drug. Even my short-term
memory returned. I had been unable to remember anything. And my cholesterol
count is actually fine and even within the goals set by cholesterol hysterics.”
“I now tell my story far and wide, even to doctors who often either raise their
eyebrows and quickly change the subject or are simply too obvious with their
indifference. I have also met with hospital doctors who have reaffirmed some of
my experiences. But I still get angry at the pharmaceutical industry and its
lackeys."
13 Alternative medicine is no
alternative
When I give public lectures explaining how dangerous many of our drugs are,
and how many lives they take, I am often asked: “What is the alternative?”
Quite often, we should let nature take its course because our bodies and minds
have a great capacity for self-healing. In other cases, we might prefer a non-drug
intervention that has documented effects, for instance, psychotherapy for mental
health issues.
Many patients and some doctors are attracted by the irrationality of alternative
medicine, which I assume is related to the propensity human beings have for
religious beliefs. Alternative medicine is so popular that the editors of the
textbook of internal medicine (also called general medicine), used by medical
students in Denmark, decided a chapter was needed about it, even though
allowing such an addition to be included in a fine textbook like that is very
strange. They asked me to write it, not because I had demonstrated any interest
in the subject, but because they knew I had the skills to go through the literature
critically.
I looked for evidence of beneficial effects of the most commonly used treatments
and came up empty-handed. None of the evidence I found was so convincing
that I would recommend the treatments. 1 Furthermore, as I will explain in the
following, alternative medicine is not harmless.
To a large extent, drug development builds on natural products. For example, the
first effective drug against cancer, paclitaxel (Taxol), was extracted from the
bark of the Pacific yew tree, and quinine, the first effective drug against
falciparum malaria used in Europe came from the bark of the South American
cinchona tree. Malaria was introduced to the Americas by Europeans, and the
Quechua peoples of Peru, Bolivia, and Ecuador discovered that a bark they
already used for shivering of other causes, even worked for malaria. Artemisia,
an extract from sagebrush, is also effective against falciparum malaria. That
treatment has been used by the Chinese for over a thousand years, yet it needs to
be stated that the Chinese have used many other herbal remedies, and this was
the only one out of almost 200 which proved to be effective when it was
scientifically investigated.
When I sit at a dinner table with people I have not met before, I try not to reveal
that I am a doctor, because I have experienced that the conversation may then go
astray and become quite strenuous. Sometimes I cannot escape listening to my
tablemate’s long, convoluted medical histories because they are seeking my
opinion. Yet taking on the role of a doctor for strangers is usually a bad idea
because I do not know the details of their medical histories. Sometimes people
get very agitated when I gently tell them that I am not interested in discussing
alternative medicine. It is like telling a religious fanatic that I do not believe in
any gods and do not wish to discuss it.
Once my tablemate was immensely tenacious and just would not accept my
excuse that I knew too little about Chinese herbs to say anything of value about
them. I tried to start a conversation with someone else at the table, but the man
would not let me go. There was no empathy whatsoever, let alone any modicum
of politeness.
He ultimately played his trump card: “Don’t you agree that Chinese herbs must
be good for people, because the Chinese have used them for thousands of
years?” I responded: “They also used bamboo as a building material for
thousands of years. If I was an engineer, would you then have told me to use
bamboo to build road bridges because the Chinese used it for thousands of
years?” For the rest of the evening, he did not look my direction.
The explanations about causality that alternative therapists use to support their
claims of positive effects are often speculative and have no connection with
reality. In 1964, American magician James Randi promised a reward of one
million dollars to anyone who, under agreed-upon, controlled circumstances,
could prove pseudoscientific postulates, for instance, the alleged mechanism of
action for the effects of reflexology, homeopathy, acupuncture and chiropractic
healing (apart from the effects on back and joint pain). Over a thousand people
have tried, yet all have failed, and the challenge was terminated in 2015.
About a quarter of all Danes contact an alternative practitioner every year, 1 and
many more buy alternative products like supplements and herbal medicine in
pharmacies and elsewhere. The most popular treatments are those involving
bodily contact – which is easily understood from an evolutionary perspective.
Apes and monkeys spend quite some time grooming each other which is
important for the social cohesion and for maintaining the hierarchy, and we
humans probably miss that kind of physical proximity. In addition, some
alternative therapists are good listeners and they tell their customers how unique
they are.
The most frequent reasons for seeking alternative treatment are mild symptoms
or disorders, a desire for increased well-being or the prevention of diseases. A
wish to be actively involved while avoiding medication harms also plays a role.
Some people have realized their doctors cannot cure them and are desperate to
try anything, making them vulnerable to exploitation by all sorts of quacks and
fraudsters. Unfortunately, some of those practitioners who exploit people’s fear
of dying are doctors using quack remedies, such as large doses of vitamins for
AIDS.
A term that has found its way to contemporary language is “the worried well,”
which implies that although you feel healthy and are in good shape, there could
be some unknown issues in your body that you should get checked. That is a
terribly bad idea. Alternative practitioners cannot make proper diagnoses and, if
they do, they are highly likely to be speculative, wrong and lacking any
scientific basis. People are often told that something is wrong with their energy
systems; that they have a lack of certain minerals or vitamins; that they are being
poisoned with all sorts of substances and therefore need special treatments like
intestinal cleansing; or that they need peculiar diets.
Today, we know so much about the human body, and its physiology and
pathophysiology, that alternative therapists cannot be excused for speaking
mumbo jumbo to their clients. No “cleansing” is needed, because the liver and
kidneys take care of toxic substances, and there is no good evidence that dental
fillings with amalgam leads to health problems, or that some people suffer from
multiple chemical sensitivities.
A common argument for using alternative medicine is that it cannot hurt. Apart
from the fact that we treat people because we hope to help them, not because we
hope we will not harm them, the argument is wrong for several more reasons.
Third, patients are often exposed to curious regimens with strict injunctions
about what to eat and drink, or they are treated with mineral mixtures or large
doses of vitamins, even though such regimens can be dangerous. As already
noted, a review of the placebo-controlled trials of antioxidants showed that beta-
carotene and vitamin E increase mortality. 6 We need vitamins and essential
minerals, e.g. zinc and copper, to make our enzymes work, but if we get too
much, we might die. The human body is far more complicated than alternative
practitioners would like us to think and it is well adapted to the environment.
Fourth, many alternative practitioners advise against vaccines even though there
is no doubt that their beneficial effects far outweigh the harms. A 2002 survey
showed that 31 of 77 homeopaths and 3 out of 16 chiropractors advised against
giving one-year old infants a vaccination for measles, mumps, and rubella
(MMR). 7 Since they knew they were participating in a research study, their
advice in their daily practices might be even worse.
Alternative medicine fraud is not only about secretly and illegally adding
substances with known pharmacological effects, listed ingredients may also be
missing: In 2015, four US retailers were accused of selling fraudulent dietary
supplements that were, in many cases, contaminated with unlisted ingredients. 8
Authorities had run tests on popular store-brands of herbal supplements at
Walmart, Walgreens, Target and GNC and found that approximately four out of
five of the products contained none of the herbs listed on their labels. In many
cases, the supplements contained little more than cheap fillers like rice and
house plants, or substances that could be hazardous to people with food allergies.
In pills labelled ginkgo biloba, the agency found only rice, asparagus and spruce,
an ornamental plant commonly used for Christmas decorations. At Target, the
agency tested six herbal products, of which three - including ginkgo biloba, St.
John’s wort and valerian root - tested negative for the herbs listed on their labels.
The pills contained powdered rice, beans, peas and wild carrots.
Chiropractors and like-minded people often take X-rays of the spine and then
declare that they can see what is wrong, typically minor subluxations. Such
statements should be disbelieved. Many scientific studies have been carried out
comparing X-ray films with clinical symptoms, and the correlation between the
two is close to zero. That poor correlation is also found elsewhere in the
skeleton. X-rays of the hip or knee joints can look terrible, with almost no
cartilage, in patients without pain, whereas other patients, who have very little
visible damage to their joints, may suffer a great deal from osteoarthritis pain. Of
course, there are exceptions to this general rule, one of which is osteoporosis
with compression fractures in the spine, but that is not a condition which is
treatable with manual therapy.
Many randomized trials have been carried out, but since the effect measures are
subjective, the fact that these trials cannot be blinded is a major problem.
Detected modest effects on pain could be due to reporting bias, because both
therapists and patients want to believe that manipulation works. Spinal
manipulative therapy is widely practiced, but a Cochrane review of 20 trials that
had studied acute low back pain did not find any effects. 9 Manipulation was no
more effective than sham manipulation, inert interventions or when it was
combined with another intervention, and it also appeared to be no better than
other recommended therapies.
The effect of spinal manipulative therapy on chronic low back pain is similarly
disappointing. A Cochrane review of 26 trials found small, statistically
significant but clinically irrelevant, short-term effects on pain relief and
functional status compared to other interventions. 10 Data were particularly
sparse for recovery, return-to-work, quality of life, and costs of care. The effect
on functional status was measured by a variety of scores based on many
individual components and a small improvement in such a score doesn’t tell us
whether the patients have actually been helped, only that it is not very likely.
Other deaths have occurred after the patients received chiropractic therapy, and
many hundreds of serious complications have been described in the research
literature. 13
People seem to love manual therapies. In the course of a year, about one-fifth of
all Danes had manual therapy or massage at least once. 1 Whenever I have a
musculoskeletal problem on the tennis court, be it a tennis elbow, a sprain, acute
low back pain or knee pain, I get the same advice from my fellow players: “You
should see a chiropractor.” Even when I tell them I was a rheumatologist for 18
months and know what I am talking about, they keep insisting I should see a
chiropractor. The only thing that works for a tennis elbow is rest. It takes time to
heal and, in the meantime, you can swing your racket like a golfer - using your
body instead of your arm - which lessens the strain on your elbow considerably.
It might even improve your game.
Massage
Many Cochrane reviews of massage have been made, but the trials are small and
of questionable quality.
Massage for promoting mental and physical health in infants has been studied in
34 trials, but the results do not support the use of massage. 16 The trials are of
poor quality and many of them do not address the biological plausibility of the
outcomes being measured, or the mechanisms by which change might be
achieved.
A review of 15 trials of massage for mechanical neck disorders graded the trials
low or very low with respect to methodology, and no recommendations for
practice could be made. 17
Another Cochrane review of 25 trials reported on the effects of massage on low
back pain and functional outcomes. 18 Large effects were reported but only in the
short-term follow-up, and the outcome assessors had not been blinded. Thus, the
authors had very little confidence that massage works for low back pain. I agree.
Sometimes we believe in treatment effects even if we don’t know the mechanism
of action, but in such cases, the trials need to be of high quality and provide
rather consistent results from trial to trial. When this is not the case and the
outcome assessors have not been blinded and furthermore, when it appears
unlikely that an intervention could work, we should be very skeptical. I cannot
see any rationale for using massage for low back pain.
The only thing that is certain about massage is that it hurts. And yet we are
supposed to be grateful. Healthcare can be very strange at times. Massage of
sore trigger points is very common, yet there are no indications that this painful
treatment helps.
Reflexology
Reflexology has roots in traditional Chinese medicine and is based on the idea
that massage of special zones on the soles of the feet can bring healing to sick
organs. However, no one has demonstrated the existence of topographic links
between the soles of the foot and the internal organs, or from hands or ears,
which are also sometimes massaged. Few trials have been carried out and they
are small and biased. There is no proven effect of reflexology on disease and it
would not be expected. Reflexology has to do with well-being, not with curing
or alleviating disease.
Acupuncture
Over a thousand randomized trials of acupuncture have been carried out, but the
vast majority are of very poor quality. Trials performed in China have more
positive results than other trials, and an overview of 49 Chinese trials in stroke
showed that the more patients in the trial, the smaller the effect. 20 The bias was
extreme and very rarely is this well-known bias as pronounced as it is for
acupuncture. Another review of acupuncture trials published in Chinese journals
found that 99.8% of 840 trials reported positive results for the primary
outccomes. 21
One of the authors of the stroke overview 20 quizzed Chinese colleagues why the
results were always positive. The uniform answer he received was that it would
be very offensive for Chinese researchers to conceive a study which does not
confirm the views held by their peers. 22 In other words, acupuncture research in
China is conducted to confirm prior assumptions that acupuncture is effective.
His conclusion was that acupuncture trials from China – which constitute most
of the trials - cannot be trusted and should be discarded outright.
By 2017, not less than 47 Cochrane reviews had the word ‘acupuncture’ in the
title, and they are really colourful. Many of the reviews are about diseases for
which we would not expect needle pricks to have any effect, e.g. schizophrenia,
artificial insemination, induction of labour, autism, myopia, glaucoma,
depression, insomnia, ADHD, stroke, epilepsy, traumatic brain injury, ischaemic
encephalopathy in neonates, stress urinary incontinence, menopausal hot flushes,
uterine fibroids, asthma, mumps, cocaine abuse, Bell's paresis, vascular
dementia, smoking cessation, restless legs and irritable bowel syndrome.
Considering the poor quality of the trials, a high likelihood exists that the
positive findings are fraudulent or false positives. On this background, it is
remarkable that very few positive effects were reported for any of these diseases.
I don’t find it worthwhile to comment on the individual reviews.
I cannot see any clothes on this emperor. Yet it is easy to be fooled. "I was once
invited to a conference in Florence. I wanted very much to see the famous Uffizi
with the Renaissance paintings, but unfortunately, I had acquired a fierce pain in
my back. At a dinner for the invited speakers, I happened to sit next to an
acupuncturist who was kind enough to offer me a free treatment. The next day
the pain in my back was gone and I could visit the Uffizi without any problems.
What makes the story interesting is that I declined the acupuncturist's offer. If I
had accepted, I would probably have had a more positive impression of
acupuncture today." 24
Acupuncture can be dangerous. During just one year, Danish authorities learned
about four cases, including two children, where the needles had punctured the
lungs, and one of the patients died. 25
Several Cochrane reviews of the effects of touch are available. The underlying
concept is that sickness and disease arise from imbalances in a so-called vital
energy field. The effect of touch is believed to occur by exerting energy to
restore, energize and balance energy field disturbances using hands-on or hands-
off techniques. 27
A Cochrane review reported some effect on pain but the review was withdrawn,
officially because it was outdated, 27 yet that might have to do with inadequate
methods. Blinding of the outcome assessor is crucial when pain is the outcome –
but the review included trials that were not blinded and did not address this
issue.
Other Cochrane reviews did not find support for the idea that touch could have
an effect on anxiety or depression. 28,29
Intercessory prayer
The authors of the Cochrane review apparently did not discover that a suspicion
of fraud had been raised against a large trial included in their review, and that the
largest ‘trial’ was meant to amuse rather than present scientific evidence.
It is also hard to accept that a god would help Peter in bed A, because someone,
after randomization, was asked to pray for him, but not the less fortunate Paul in
bed B. The authors contradict themselves when they say that their review
focuses on people, "setting time aside to communicate with God," as the review
is not about divine intervention. They are also inconsistent when they note that,
"If understanding of God is as limited as the Holy Literature suggests (1
Corinthians 13:12), the consequences of divine intervention may be considerably
more subtle than could be measured in the crude results of a trial." If that was a
real concern, the authors should not have undertaken the review, because their
reservation means that people who do trials of prayer cannot rely on what they
observe.
Another statement also belongs to the realm of mysticism. The authors write
that, "An omnipotent God would make concealment of allocation (of the
participants to prayer or no prayer) impossible and may be noncompliant with
the limitations of a randomized trial (Psalm 106:14,15, Job 42:2)." Since such a
god could interfere with the experimental setup, it is difficult to understand why
the authors excluded trials in which the treatment allocation was not concealed,
and why they bothered to discuss the level of concealment in the trials they
included.
The largest trial was published in BMJ's Christmas issue and was meant to
amuse, in line with the tradition of this special issue, because the trial evaluated
the effect of prayer taking place 4-10 years after the patients had either left the
hospital alive or had died from their bloodstream infection. Thus, the trial
evaluated the effect of retroactive intercessory prayer using historical data and
its author argued that we cannot assume, "that God is limited by a linear time."
The authors of the Cochrane review did not mention anywhere that the patients
were randomized many years after their outcomes had occurred and did not
discuss the likelihood that time can go backwards, and that prayer can wake the
dead.
The author of the retrospective prayer study subsequently noted that "if the pre-
trial probability is infinitesimally low, the results of the trial will not really
change it, and the trial should not be performed. This, to my mind, turns the
article into a non-study." 32 The non-study ‘found’ a nonsignificant reduction in
death for those prayed for (relative risk 0.93, 95% confidence interval 0.84 to
1.03), but since it carried 75% of the weight in the meta-analysis in the Cochrane
review, it led to a statistically significant effect of prayer.
Two years later, also in the Christmas issue, people with an interest in alternative
medicine, prayer and healing tried to explain why the results of the retroactive
study could be true using arguments from quantum theory. 33 They seemed to
take their own arguments seriously, even though they were total nonsense which
a physicist demonstrated a year later - again in the Christmas issue. 34 Down-to-
earth, it should not be too difficult to realize that prayer cannot make dead
patients come to life again. Furthermore, all the randomization did was to divide
both the living and the dead into two groups that were then compared
statistically. That is also meaningless because we already knew that any
differences between the two groups were random.
The amusements and surprises did not stop there. Another trial originally had
three authors, but the senior author subsequently withdrew his authorship. On
PubMed, there is reference to an erratum in the journal, 35 but our university
library has informed us that the page that should describe the withdrawn
authorship does not exist in the journal. Therefore, we asked the editors of the
Journal of Reproductive Medicine whether the PubMed citation is wrong or
whether the erratum was not published in the Journal. We did not receive any
reply despite repeated requests but were not the only ones who were ignored.
Clarifications addressed to the authors and editors from scientists and journalists
were not answered either, and not a single critical letter was published in the
journal. 36,37 The trial was carried out at Columbia University in New York City
and a news release from the university stated that the senior author led the trial.
However, the vice president noted that the senior author first learned of the trial
from the first author six to twelve months after it was completed. 36 One of the
two remaining authors, lawyer Daniel Wirth (mentioned above), was sent to
prison after 20 years of continuous criminal, fraudulent activities, 36,37 and the
other author provided incorrect and misleading statements about the research
38,39 after being challenged by the editor to provide explanations when the
scandal broke loose three years after the trial was published.
Wirth organized the study which reported a significantly higher pregnancy rate
in the prayer group (50% versus 26%, P = 0.001) after in-vitro fertilization at a
Korean hospital. The prayer was long-distance - carried out in USA, Canada and
Australia. Those who prayed were Christians, as opposed to the Korean patients.
Another curiosity is that the Catholic Church condemns in-vitro fertilization.
Therefore, it would have been equally reasonable to conclude that the responsive
god was not very well represented by the Pope as to conclude that one should
pray for those seeking in-vitro fertilization.
Scientific misconduct seems to have been involved in a third trial 34,40 which
was originally included in the Cochrane review, yet is now excluded, not
because of suspected misconduct but because the intervention was distance
healing and not prayer.
The authors of the Cochrane review also contributed further to the amusements,
albeit not deliberately so. They included a study reporting an increased risk of
surgical complications due to prayer but only if the patients were aware that
people prayed for them. Instead of discussing the plausibility of this finding, the
authors concluded that people intervening with prayer should be "cautious about
informing the recipient," when it comes to surgery, and that managers and
policymakers may wish to exercise some caution about "praying at the bedside
of those who are about to have a surgical operation."
When discussing the effect of prayer on the "clinical state," the Cochrane authors
argued that the lack of effect might be because the participants only received
prayer for 14 days. Their inclination to theological reasoning leads to a
tautology: "A caring God may not wish to prolong suffering, so death therefore
might be a positive outcome of prayer." This is a perfect immunization of the
hypothesis that makes trials of prayer meaningless. If people survive, it is good
for them, and if they die, it is also good for them. The authors’ reasoning is based
on the assumption of an omnipotent and all-knowing god. But if that were true?
Why should we then try to influence our fate when such a god already knows
what is best for us?
The review was updated in 2009 after our criticism of it and the authors have
changed their conclusion. They originally wrote that, “The evidence presented so
far is interesting enough to justify further study into the human aspects of the
effects of prayer.” But now they write: “We are not convinced that further trials
of this intervention should be undertaken and would prefer to see any resources
available for such a trial used to investigate other questions in health care.”
However, they still include the study of retroactive prayer, justifying it with the
most mysterious arguments. They call it a “relevant study,” “not in jest,” but “a
rather serious paper.” They say further that, “retrospective prayer is practised by
some people,” and that the study was double-blind since those praying did not
know the outcome for any of the patients. Well, perhaps they did not, but as the
outcome was already known for all patients, it is wrong to give a study like this
bonus points for being “double-blind.” The authors of the Cochrane view
perverted the research methodological principles without even being aware that
they had made themselves laughable.
On the possibility of waking the dead through prayer, they say: “Retrospective
prayer may be considered theologically controversial, but we are not concerned
with theology. Our aim is to review the empirical evidence for the efficacy of
prayer as a treatment for ill-health rather than to consider questions of
metaphysics. We judge ourselves bound to analyse the results of any trial that
fits our original criteria (including our initial definition of prayer) and which is
methodologically well constructed. Having set our protocol we are convinced
that it would be unscientific to modify it to exclude a study that fits our criteria
for inclusion.”
This is dogmatic cook-book ‘science’ in its worst form. People are obliged to
think even with a protocol. Otherwise, it is not science.
The review authors claim they have found no evidence that the study was a jest.
That is not correct. The author of the retroactive prayer study explained that it
was a jest, 32 and we pointed out in our comments on the review that we got the
same answer when we wrote to the author.
Finally, the review authors say that, “We are keen that all studies meeting the
clearly stated inclusion criteria should be reported (even if later stated to have
been “written in jest”), rather than being kept hidden and perpetuating
publication bias.” That argument is nonsense. They could formally include the
jest according to their inclusion criteria, but review authors are free to not
include unreliable studies in their meta-analyses; in fact, this is recommended for
Cochrane reviews.
It is a scandal for the Cochrane Collaboration that this ridiculous review has not
been withdrawn a long time ago.
Craniosacral therapy
A craniosacral therapist website describes that treatment is based on a rhythm,
the so-called ‘craniosacral pulse,’ which can be felt throughout the body.
However, such a pulse has never been found in studies of human physiology.
Light touches are thought to relieve tension and blockages, especially around the
head (the skull and its sutures), spine and pelvis. By searching the Cochrane
Library on craniosacral, I found only one review. 41 It is about interventions for
preventing and treating low-back and pelvic pain during pregnancy, and it
includes one trial. There were 123 patients in the analyses and the therapy
improved pelvic pain in the morning and functional disability (P = 0.02 for both
outcomes). However, this result is very uncertain because the patients were not
blinded. The authors described this trial as being at low risk of bias because
there was an independent observer who measured the pain without knowledge of
group assignment. But that is wrong. Pain is a subjective feeling that can only be
assessed by the patient - and the patients were not blinded. Furthermore, the
difference in morning pain was only 8 mm on a 100 mm pain scale, which is
lacking clinical relevance and can easily be nothing more than bias in a
nonblinded trial. The effect on functional disability was also small, and there
were no significant differences between groups in evening pain or days off
work/sick leave.
Homeopathy
Homeopathy was created by Samuel Hahnemann, a German physician who,
more than 200 years ago, ceased to work as a medical practitioner because he
realized that many treatments of his time were harmful. He noted that quinine
induces the same symptoms as malaria and drew the wrong conclusion that
patients should be treated with medications which, in healthy people, produce
the same symptoms as the disease. He ‘solved’ the problem with the toxicity of
quinine by diluting the solution a great number of times.
It is easy to work out what these dilutions lead to. 44 Hahnemann created the
centesimal or "C scale,” diluting a substance by a factor of 100 at each stage. A
2C dilution requires a substance to be diluted to one part in one hundred, and
then some of that is diluted by the same amount. That works out to one part of
the original substance in 10,000 parts of the solution. A 12C dilution means that
the original material is diluted by a factor of 10 12. Already at this stage, the
dilution is at a level that, starting with a substance contained in a small 1 cl
syringe, it corresponds to dissolving that substance in the all the world's oceanic
water. And then we are not even half-way through this because Hahnemann
advocated 30C dilutions for most purposes, i.e. a dilution by a factor of 10 60. If
we do that, it corresponds to dissolving the substance in a cube of water with
sides far larger than the distance from the Earth to the nearest galaxy, the
Andromeda Galaxy 2.5 million light years away.
For homeopaths, that is not a problem. A solution that is more diluted than
another is described as having a higher potency, and homeopaths consider more
diluted substances to be stronger and deeper-acting.
Four years later, another group of researchers looked at the same 89 trials and
found very different results. 46 When the results were depicted graphically, they
were highly asymmetrical. The treatment effects were much larger in small
studies and in studies with inadequate blinding of outcome assessment, and they
were also larger in trials published in languages other than English. There was
no effect in the largest trials of homeopathy, which were double-blind and had
adequate concealment of the randomization (i.e. it was not possible to cheat by
deliberately assigning patients with good prognoses to homeopathy and others to
placebo homeopathy).
The European Union also contributed to the folly in the most remarkable way. In
2011, the European Parliament’s agriculture committee agreed to spend two
million Euros on investigating whether cattle, sheep and pigs could benefit from
homeopathy. 47 Critics pointed out that animals cannot benefit from a placebo
effect because they will not understand they have been given a treatment.
There are reports of homeopaths convincing their customers not to take malaria
prophylaxis when traveling in infested areas. In 2006, the BBC visited Britain’s
biggest manufacturer of homeopathic remedies with a hidden camera. 48 The
journalist said she planned to go to Malawi – a high risk area – yet the shop only
suggested garlic, oil of citronella and vitamins rather than a trip to the doctor.
That was all they recommended for malaria. The adviser also told the journalist
that the homeopathic compounds would protect her: “They make it so your
energy doesn’t have a malaria-shaped hole in it so the malarial mosquitos won’t
come along and fill that in.” Sheer gobbledygook, yet not atypical of the way
many practitioners of alternative medicine ‘explain’ things.
The BBC also revealed that some homeopathic pharmacies claimed their
products could treat malaria in lieu of anti-malarial drugs. 49 Homeopathic
pharmacy websites show many products with indications, e.g. for influenza, and
there are homeopathic replacements for vaccines against measles, mumps and
rubella, and homeopathic pills for hepatitis, tuberculosis and typhoid. 49
Sometimes it is the other way around - homeopathic remedies can contain too
much of a substance. In the United States, several babies died probably because
the biggest manufacturer of homeopathic remedies put too much of the deadly
nightshade, Atropa belladonna, in its “teething” tablets. 50
In 2017, a seven-year old Italian boy died from an ear infection that had spread
to his brain. 51 The family’s homeopath - who played doctor - discouraged the
mother from giving the child antibiotics even though his condition worsened
over a couple of weeks, and the homeopath noted that the homeopathic treatment
should continue. In contrast, a doctor on call advised that the child should
immediately go to hospital. Even when the infection spread and the boy was in
critical condition, the family refused to give him antibiotics. When his parents
finally called for an ambulance, it was too late. The boy went into coma and died
three days later. The parents have been charged with manslaughter.
Homeopathy has never been popular in the Nordic countries. In 2003, only 1%
of Danes took a homeopathic remedy, 52 whereas in France, 36% took such a
remedy in 1992. 53 In the UK, homeopathic hospitals are functioning within the
National Health Service, and in several European countries, homeopathy can be
studied at universities. 53 In 1998, homeopathy was the most frequently used
alternative therapy in 5 out of 14 surveyed countries in Europe, and in Germany,
approximately 6,000 medical doctors had formal qualifications in homeopathy.
53
The European Commission has estimated that adverse reactions kill about
200,000 EU citizens annually at a cost of €79 billion. 3 Many of these deaths are
avoidable. Our for-profit system encourages overprescribing and many patients
could have fared well without the drug that killed them, e.g. an NSAID or a
psychoactive drug. 4,5 Meanwhile, many important health problems do not
receive the attention they deserve, e.g. addressing antimicrobial resistance.
The main problem is that the current system is based on patents and monopolies,
allowing companies to set prices as they want. This system is unethical because
people can die if they cannot afford to buy the drugs they need. It is also
inefficient because research knowledge, for instance, about toxicology and failed
projects is not shared. Furthermore, an international agreement on intellectual
property rights prohibits generic drug manufacturers from using the clinical trial
data that brand manufacturers submits to drug regulators. 6 That means many
superfluous and therefore, unethical trials need to be carried out in order to
provide citizens with cheaper drugs when the patents on expensive drugs have
expired.
I have proposed a radically different approach in which the current drive for
profit maximization via patents is replaced by a public interest driven system
that is not-for-profit. 1
The drug industry spends only 1-2% of gross revenues, net of taxpayer subsidies,
on basic research into discovering new molecules. 7 Most of the basic knowledge
for developing treatment advances comes from publicly funded laboratories and
institutions. 8-10 In its drive to maximize profits, the industry tends to focus on
drugs treating chronic conditions affecting many people, often by patenting
minor variations of existing drugs with no added therapeutic value. However,
that is rarely a hindrance to selling them in large volumes at prices that can be 10
to 20 times more expensive than off-patent drugs. 4,5 In order to achieve this, the
industry spends much more on marketing than on research and development. 4
The current system provides little incentives for studying and developing less
expensive non-drug interventions even though they are often preferable. Some of
our largest expenditures are for drugs treating type 2 diabetes, hypertension and
high cholesterol – all largely attributable to poor diet and lack of exercise.
Another example is the high and increasing rate of usage of psychiatric drugs,
even though psychiatric disorders are better treated with psychotherapy.
The industry’s justification for patents and high drug pricing is that patents are
needed to recoup the great costs of drug development and thus, ensure a needed
supply of new drugs. Some 15 years ago, the industry narrative put the cost of
developing a new drug at about $1 billion US, 10,11 while independent analysts
arrived at a figure that was only 10% of this. 12 Currently, the Drugs for
Neglected Diseases Initiative (DNDi) estimates that it can develop a new drug
for $110-170 million including theoretical costs of failed projects. 13 In reality,
the prices of drugs do not reflect research and development costs, but rather what
heavily subsidized ‘markets’ are willing to pay.
In recent years, drug agencies have gradually relaxed their standards for
approval, and drug companies pay fees which provide them leverage in the
regulatory system. 14 The drug industry contributes 83% of the entire budget for
the European Medicines Agency (EMA), 15 even though anybody receiving 83%
of their salary from the industry would not be permitted to serve on any drug
evaluation committee. With less rigorous regulatory standards, more drugs than
previously have been withdrawn from the market or have received serious safety
warnings. 4,16,17
Regulatory requirements are particularly low for cancer, and many hugely
expensive cancer drugs have been approved without conducting a single
randomized trial, 18,19 and with only surrogate outcomes, e.g. disease-free
survival instead of longer life. New cancer drugs are generally no better than
existing ones, 4 or they increase survival by a mere month or two. 20,21
The standards for approval continue to fall, most recently illustrated by EMA’s
introduction of adaptive pathways allowing drugs to be approved based only on
observational data. 22
Some patient groups support industry demands for faster approvals which will
provide them rapid access to the latest drugs. However, it is a myth that the
current system benefits patients and, since most patient advocacy groups accept
drug company funding, they should generally not speak publicly on behalf of
patients. 23,24
Substantial initial funding will be needed in the transition phase to any new
system for developing the necessary public infrastructure and to pay for public
drug development. Several models already exist, one of which is taxation. The
Italian drug agency requires drug companies to contribute 5% of their
promotional expenses beyond salaries – which has created a large fund used
partly for independent clinical research. 26,27 Spain has a similar initiative. 27 A
sales tax would create much greater income, but most importantly, the new
system will avoid the huge wastes we currently endure. It has been estimated
that the savings generated by the new system would be 5-10 times greater than
the amount the drug industry currently spends on research and development. 6
Therefore, vastly more public money is being poured into the current system
than what would be needed in future.
In the not-for-profit model, the price of drugs will be set low enough – the
manufacturing cost plus a small margin - that even third world countries could
afford to buy drugs. That would improve the health of their citizens and increase
international trade and prosperity.
Once fully implemented, this new system will have abolished the patenting of
drugs and devices. In the transition period, all regulations that impede the
introduction of generic medicines and biosimilars to the market should be
removed, and new patents for minor changes, e.g. the removal of the inactive
part in a stereoisomer, should not be allowed. The bar should be raised
substantially for launching lawsuits against generic competitors with claims of
patent violations, and the time limits shortened for resolving law suits and for
patent exclusivity. Companies that launch frivolous law suits should be subject
to stiff penalties because the mere threat of such law suits often stifles innovation
in start-up companies. 6
In the transitional phase when drug companies still have new drugs under
development, they could be offered a buy-out of their patent, a ‘prize,’
commensurate with the benefits and harms of the drug, as documented in
publicly conducted trials with relevant comparators and outcomes. The use of a
prize system is consistent with proposals from the World Health Organization
and the EU Council, which have called for needs-driven innovation and for
further exploration of innovation models that de-link the postulated cost of
research and development from the price of medicines. 28 Similar thoughts were
expressed in a US Senate Bill. 11
Countries should collaborate with companies on price negotiations and use their
powers to refuse reimbursing overpriced drugs and to impose pricing practices
that take into account public research investments and the fact that
pharmaceuticals are public goods (as opposed to the industry’s currently
promoted value-based pricing approach, which puts a monetary value on life as
an excuse for high drug prices).
To engage the public in the profound changes of the new system, a program of
education and myth busting needs to be undertaken to combat the widespread
erroneous beliefs that sustain the current system.
Like any drug development endeavour, the new system will have to manage the
risk of aborted projects. Taxpayers may view such failures as a poor use of their
money unless they understand the realities of scientific research, including the
rarity of breakthroughs. Continuous education of the public and politicians with
trustworthy and transparent figures for the costs of research and development
will be needed.
In the new system, drug agencies are fully publicly funded and are far more
focused on drug harms. Trials submitted for obtaining marketing authorization
should be large enough, and run for sufficient lengths of time, to capture rare but
lethal harms, particularly because promises of post-marketing studies are often
not fulfilled. 26,32
Currently, drugs are allowed to be brought to the market based on effects that are
not clinically relevant. The effect of neuroleptics and depression pills are, for
instance, far below the thresholds psychiatrists have found to constitute a
minimally, clinically relevant effect. 5
Drugs should not be approved based on surrogate outcomes (e.g. blood glucose
rather than complications to diabetes) except when they are validated to correlate
with patient relevant outcomes, which is very rarely the case. 33-35 So-called non-
inferiority and equivalence trials are also usually misleading 36-38 and should
rarely be accepted. The norm should be to establish benefit in usual (superiority)
trials compared to the best available interventions, instead of allowing
companies to arrive at misleading conclusions like: “Our drug was at least as
effective as the old one,” when what they really showed was: “We are 95%
certain that our hugely expensive new drug doesn’t kill 50% more patients than
the old, cheap drug.” 36
Publicly conducted drug trials will ensure that new drugs are being compared
with old, cheap drugs under fair circumstances, and even with non-drug
interventions. They will also be vastly less expensive than drug industry
conducted trials. The European Society of Cardiology has estimated that
university centres can perform drug trials for about one-tenth to one-twentieth
the cost of industry trials where numerous for-profit middlemen tack on hefty
surcharges. 40
To improve the usefulness of trials for patients, draft trial protocols will be
publicly available on a website that allows patients and others to comment on
them. All information related to the trials will be publicly accessible, from the
pre-planned outcomes to the raw, anonymized patient data, allowing others to
conduct their own analyses. The trial reports will be published in open access
journals or on the web so that everyone, including patients volunteering for the
trials, can access them free of charge. Preclinical studies (e.g. animal toxicology
studies), including the raw data, will also be made available.
Politicians often see the drug industry as a growth factor contributing to job
opportunities, trade balance and the knowledge economy - a perception the
industry promotes. In 2013, according to the European pharmaceutical industry
association, the industry directly employed more than 690,000 people in Europe
and generated three to four times more jobs indirectly. 41 However, many of
these jobs are in sales and legal departments - ultimately paid for by all of us
through higher drug prices. Furthermore, intensive marketing causes many
unnecessary deaths and is harmful for our national economies.
A better future requires political will, yet we have no other option than to
radically change the current system. Our political leaders are slowly waking up,
and as a result, two workshops were held in Amsterdam in 2016 with the
participation of 30 experts and stakeholders from Europe and North America,
including patient representatives, industry leaders, academics, regulators, payers,
and government representatives. At these workshops, four scenarios were
developed, which were presented to a wider audience in the context of the Dutch
EU presidency in June 2016. 43,44 The idea was to develop creative scenarios
with novel, more sustainable ways to ensure patient access to safe and effective
drugs, while maintaining strong incentives for innovation. I participated in the
workshops and suggested that we abandon patents in healthcare and make drug
development a public enterprise.
This multitude of trials was also used for another purpose. The length of time
between the first symptom and initiation of treatment varied between trials, and
when the mortality was correlated to the treatment delay, there was a clear, linear
relationship showing that streptokinase and similar drugs decrease mortality only
if started within the first 16 hours. 46 If started later, they increase mortality.
Patients must be respected as equal partners in clinical trials. If you are asked to
participate in a trial, you should ask for copies of all documents important to
your considerations. The principle of informed consent is crucial to the ethical
conduct of experiments on people, and if you are not properly informed, you do
not know to what you are consenting.
We explained that clinical study reports and protocols are based on well-known
principles that can be applied to any drug trial; that clinical study reports
describe the clinical effects of drugs; and that nothing in EMA’s guidelines for
preparation of such reports indicates that any information included in them can
be considered trade secrets. Trial protocols are always sent to the clinical
investigators and it is unlikely that companies would have included any
information that could be of commercial value (such as a description of the drug
synthesis). We also noted that the clinical study reports and trial protocols
represent the last phase of drug development, which are preceded by many years
of preclinical development, and that other companies could hardly use them as a
basis for developing similar drugs.
If people have nothing to hide, they should hide nothing, and if they hide
something, you should not participate in the trial.
You should decline if you are not allowed to get a copy of the full trial protocol
and all other relevant documents related to the trial, such as the investigators’
brochure, separate financial or publication agreements that are not part of the
protocol, and the case report forms that will be used to register the outcomes of
the trial. These are very important for finding out if the investigators are
genuinely interested in recording the harms of the drug. You should also verify
for yourself that the results will be published, no matter what they show, and that
the raw data (in anonymized form) will be made publicly available. This will
allow others to analyze the data and see whether they agree with the corporate
sponsors.
He remained unhappy and contacted me for advice. I suggested he asked his trial
physician for the trial protocol and the investigators’ brochure with the results
from the animal studies. The company responded that, “It is GSK policy not to
hand out a study protocol to non-site staff or persons as long as the
corresponding clinical trial is being conducted, because the protocol may contain
sensitive information from a commercial perspective.” However, the person who
requested to see the documents was not just anybody. Since he was a patient, he
was a partner in the trial. When the company refused to provide the requested
information, the vet became suspicious there was something wrong with
darapladib.
Since GSK did not provide the information the patient wanted, I requested it
from the research ethics committee and the Danish National Board of Health.
Both authorities asked the contract research organization (CRO) that carried out
the trial on GSK’s behalf, and GSK whether the requested material contained
confidential information that needed to be redacted. The CRO said that was not
the case, even though every page of the documents was labelled “confidential.” I
received the toxicology section from the investigators’ brochure and the trial
protocol from GSK in the United States, and also from the Danish National
Board of Health.
It emerged that one reason the participant was not given the results of the long-
term animal studies he asked for in 2010 was that these studies began in tandem
with the human trials, and therefore, it was unlikely that GSK would have been
able to communicate the results.
It also turned out that some of the animals had developed rare cancers in their
small intestines, which caused GSK concern, yet which they did not clearly
address when writing to the investigators. In fact, GSK’s information was hard to
interpret. GSK also downplayed the risk in its updated patient information.
However, the data were clear enough in the investigators’ brochure, and I found
these data worrying, as did also the vet who wrote to me, “Had I known these
facts, I would not have signed up as a participant in a clinical trial. That the rules
probably were followed does not change the fact that they failed to inform me
there were no long-term studies in animals at the time when I first asked to see
the results. That is both immoral and unacceptable.” He felt that he and 13,000
other people had served as human guinea pigs and that he had been misled when
he was told that GSK would not provide the data.
GSK conducted another study on darapladip. I find it difficult to accept that drug
regulators in over 30 countries allowed GSK to enroll over 28,000 patients
before the results of the carcinogenicity studies in animals were communicated
to them. That should not happen.
This story contributes to the existing concerns that the social contract between
patients and corporate sponsors of drug trials is broken. 4 The patients were
unaware that long-term animal studies had not been completed. And the patient
consent form for the trial stated that the research data were owned by GSK.
Patients should not sign such a form, and doctors should not participate in such
trials.
15 Pregnancy and childbirth
Not so long ago, many interventions were used during pregnancy and childbirth
that were harmful. But in the 1980’s, the founder of the Cochrane Collaboration,
obstetrician Sir Iain Chalmers, set out to get that house in order. He and two
other obstetricians collected the randomized trials in this area and convinced
many of their colleagues that they should review the trials, find out what
worked, and what did not. They published their reviews and meta-analyses in a
database and wrote a huge, two-volume book. 1 They also published a short
guidebook 2 which had sold more than 40,000 copies already 25 years ago.
Today, it can be downloaded for free.
The two oldest Cochrane groups, the Pregnancy and Childbirth Group and the
Neonatal Group, have published hundreds of Cochrane reviews. If you are
pregnant, you might find it worthwhile to browse the titles of the Cochrane
reviews about interventions during pregnancy and childbirth, and if you
experience difficulties with your new-born child, you could browse the reviews
about the neonatal period.
The guidebook was only five years old when my wife got pregnant with our first
child. I felt my role was to protect her against interventions that did not work or
were harmful. Therefore, whenever a midwife or a doctor wanted to intervene, I
consulted the book. After I had gently demonstrated several times to our
midwife that what she wanted to do was not evidence-based, she became so
interested that she bought the guidebook.
In preparation for childbirth, a small group of women had group sessions with a
midwife. One of the women gave birth to a mentally handicapped child, because
the staff had not been sufficiently on guard considering that she had previously
had a Caesarean section. Her uterus burst causing damage to the child’s brain. If
you already had a Caesarean, it is safest to have another one.
People need to remind their doctors and other staff again and again of crucial
issues in order to reduce the risk of adversities.
Organizations that are idealistic from the outset and aim at providing the most
reliable evidence and suggesting the most cost-effective interventions, are in
great danger of being shut down by short-sighted or corrupt politicians. 3 The
existence of such organizations is far too threatening for all sorts of special
interests.
The Cochrane Neonatal Group, based in Canada, is a sad example that no one is
safe. It lost public funding in 2017, which is painfully short-sighted. It costs only
about 250,000 Euros a year to run a Cochrane review group. The whole world
benefits tremendously from the output of this group and conversely, Canada
benefits tremendously from the many Cochrane reviews produced outside
Canada.
It costs more than 15 million Euros annually to run the entire Cochrane
Collaboration, and Cochrane has a reserve capital that can be used as transitional
funding when a group loses funding. However, loss of funding is disruptive and
there are no guarantees the group will survive.
That does not seem to be a viable way for Cochrane reviews to move forward.
Expecting people to donate many months of spare time and then requiring them
to pay for their generous contributions is not reasonable. Furthermore, getting
funding for Cochrane reviews is far more difficult than for so-called original
research (most of which is not the slightest bit original or useful).
In 2017, I was elected to the Cochrane Governing Board and wrote a short
article about my experiences at two board meetings. 5 We had just been to South
Africa and I pointed out what a great moral leader Nelson Mandela was, not only
for his country, but for all of us. 6
My vision for the Cochrane Collaboration was that it became the world’s moral
leader in healthcare. Cochrane must incessantly point out the need for a better
world where clinical trials are not designed, carried out and analyzed by those
who have direct financial interests in the outcomes, and where the data are no
longer manipulated or hidden if they do not please sponsors. 3
Unfortunately, Cochrane is far from being a moral leader and has not advocated
for more reliable data despite being encouraged to do so.
“If your life has any value, it is that is has no value. Those who can die, live
freely.”
I hope this book has encouraged you to use fewer drugs, and if you are on drugs,
to try to get off some or all of them. Perhaps you will run a slightly increased
risk without your drugs. But your life will likely improve. And many more of us
will survive longer if we take fewer drugs, whether we get them on prescription
or buy them illegally on the street.
About the Author
Professor Peter C. Gøtzsche graduated with a Master of Science in biology and
chemistry in 1974 and as a physician 1984. He is a specialist in internal
medicine, worked in the pharmaceutical industry, 1975-1983 and at hospitals in
Copenhagen, 1984-95. In 1993, in collaboration with approximately 80 other
professionals, Professor Gøtzsche helped start the Cochrane Collaboration - with
the founder, Sir Iain Chalmers - and established the Nordic Cochrane Centre the
same year. He became the Professor of Clinical Research Design and Analysis in
2010 at the University of Copenhagen and was elected to the Cochrane
Governing Board in 2017.
In March 2019, Gøtzsche opened the Institute for Scientific Freedom of which
he is the director: scientificfreedom.dk.
Gøtzsche has published more than 70 papers in “the big five” ( BMJ, Lancet,
JAMA, Annals of Internal Medicine and New England Journal of Medicine ) and
his scientific works have been cited 30,000 times.
Gøtzsche PC. Deadly Medicines and Organised Crime: How big pharma has
corrupted health care. London: Radcliffe Publishing; 2013 (Winner, British
Medical Association’s Annual Book Award in the category Basis of Medicine in
2014). Translated into several languages including Danish, see
deadlymedicines.dk.
3 Information sources
5 Infections
7 More on screening
8 Emotional pain
9 Physical pain
10 Treatment of cancer
12 Other issues
16 Epilogue
1 Introduction
1 Weingart SN, Wilson RM, Gibberd RW, et al. Epidemiology of medical error.
BMJ 2000;320:774–7.
6 van der Hooft CS, Sturkenboom MC, van Grootheest K, et al. Adverse drug
reaction-related hospitalisations: a nationwide study in The Netherlands. Drug
Saf 2006;29:161-8.
7 Landrigan CP, Parry GJ, Bones CB, et al. Temporal trends in rates of patient
harm resulting from medical care. N Engl J Med 2010;363:2124-34.
10 Makary MA, Daniel M. Medical error - the third leading cause of death in the
US. BMJ 2016;353:i2139.
11 Gøtzsche PC. Deadly medicines and organised crime: How big pharma has
corrupted health care. London: Radcliffe Publishing; 2013.
16 Zahl PH, Gøtzsche PC, Maehlen J. Natural history of breast cancers detected
in the Swedish mammography screening programme: a cohort study. Lancet
Oncol 2011;12:1118-24.
18 Getz L, Sigurdsson JA, Hetlevik I, et al. Estimating the high risk group for
cardiovascular disease in the Norwegian HUNT 2 population according to the
2003 European guidelines: modelling study. BMJ 2005;331:551.
24 Yarnall KS, Pollak KI, Østbye T, et al. Primary care: is there enough time for
prevention? Am J Public Health 2003;93:635-41.
2 Gøtzsche PC. Deadly medicines and organised crime: How big pharma has
corrupted health care. London: Radcliffe Publishing; 2013.
7 Gøtzsche PC. Believability of relative risks and odds ratios in abstracts: cross-
sectional study. BMJ 2006;333:231-4.
8 Ioannidis JP. Why most published research findings are false. PLoS Med
2005;2:e124.
11 Gøtzsche PC, Jørgensen KJ. Screening for breast cancer with mammography.
Cochrane Database Sys Rev 2013;6:CD001877.
13 Krogsbøll LT, Jørgensen KJ, Grønhøj Larsen C, et al. General health checks
in adults for reducing morbidity and mortality from disease. Cochrane Database
Sys Rev 2012;10:CD009009.
18 Gøtzsche PC, Johansen HK. House dust mite control measures for asthma.
Cochrane Database Syst Rev 2008;2:CD001187.
19 Gøtzsche PC, Johansen HK. House dust mite control measures for asthma:
systematic review. Allergy 2008;63:646–59.
20 National Heart, Lung, and Blood Institute; National Asthma Education and
Prevention Program. Expert panel report 3: guidelines for the diagnosis and
management of asthma. Washington, DC: US Department of Health, 2007.
http://www.nhlbi.nih.gov/guidelines/7asthma/asthgdln.htm.
21 Gøtzsche PC. Asthma guidelines on house dust mites are not evidence-based.
Lancet 2007;370:2100–1.
22 Gøtzsche PC, Johansen HK. Authors’ reply on ‘House dust mite control
measures for asthma’. Allergy 2009;64:190.
24 Bacharier LB, Boner A, Carlsen KH, et al. Diagnosis and treatment of asthma
in childhood: a PRACTALL consensus report. Allergy 2008;63:5–34.
26 Schmidt LM, Gøtzsche PC. Of mites and men: reference bias in narrative
review articles; a systematic review. J Fam Pract 2005;54:334–8.
28 Hallas HE. House-dust mites in our homes are a contamination from outdoor
sources. Medical Hypotheses 2010;74:777–9.
44 Koretz RL, Lin KW, Ioannidis JP, et al. Is widespread screening for hepatitis
C justified? BMJ 2015;350:g7809.
49 Kew KM, Seniukovich A. Inhaled steroids and risk of pneumonia for chronic
obstructive pulmonary disease. Cochrane Database Syst Rev 2014;3:CD010115.
54 Nannini LJ, Poole P, Milan SJ, et al. Combined corticosteroid and long-acting
beta 2-agonist in one inhaler versus placebo for chronic obstructive pulmonary
disease. Cochrane Database Syst Rev 2013;11:CD003794.
3 Information sources
1 Gøtzsche PC. Deadly medicines and organised crime: How big pharma has
corrupted health care. London: Radcliffe Publishing; 2013.
4https://en.wikipedia.org/wiki/Wikipedia:Size_comparisons.
9 Gøtzsche PC, Johansen HK. House dust mite control measures for asthma.
Cochrane Database Syst Rev 2008;2:CD001187.
16 Bakalar N. No extra benefits are seen in stents for coronary artery disease.
New York Times 2012; 27 Feb.
21 Dodd SR, Lancaster GA, Craig JV, et al. In a systematic review, infrared ear
thermometry for fever diagnosis in children finds poor sensitivity. J Clin
Epidemiol 2006;59:354–7.
22 Krogsbøll LT, Jørgensen KJ, Gøtzsche PC. Screening with urinary dipsticks
for reducing morbidity and mortality. Cochrane Database Syst Rev
2015;1:CD010007.
5 Infections
1 Burcharth J, Pommergaard HC, Alamili M, et al. [One in five surgeons do not
wash hands after visiting a toilet - an ethnographic field study]. Ugeskr Læger
2014;176:V66434.
2 Stranden AL. Pas på: Her er turistlandene med flest resistente bakterier.
Videnskab.dk 2017; 18 July.
3 Deer B. How the case against the MMR vaccine was fixed. BMJ
2011;342:c5347.
10 Aaby P, Ravn H, Benn CS. The WHO review of the possible nonspecific
effects of diphtheria-tetanus-pertussis vaccine. Pediatr Infect Dis J
2016;35:1247-57.
19 Chandler RE, Juhlin K, Fransson J, et al. Current safety concems with human
papillomavirus vaccine: a cluster analysis of reports in Vigibase (R). Drug Saf
2016; 16 Sept.
33 McCartney M. New York University sacks professor for refusing flu shot.
BMJ 2017;357:j1975.
35 lacobucci G. NHS staff who refuse flu vaccine this winter will have to give
reasons. BMJ 2017;359:j4766.
37 Lenzer J. Centers for Disease Control and Prevention: protecting the private
good? BMJ 2015;350:h2362.
40 Karlowski TR, Chalmers TC, Frenkel LD, et al. Ascorbic acid for the
common cold. A prophylactic and therapeutic trial. JAMA 1975;231:1038-42.
44 Sharfstein JM, North M, Serwint JR. Over the counter but no longer under
the radar – pediatric cough and cold medications. N Engl J Med 2007;357:2321–
4.
45 Mann D. FDA pulls 500 cold medicines from the market. 2001; 2 March.
http://www.webmd.com/cold-and-flu/news/20110302/fda-pulls-500-cold-
medicines-from-market#1.
47 Evans SS, Repasky EA, Fisher DT. Fever and the thermal regulation of
immunity: the immune system feels the heat. Nat Rev Immunol 2015;15:335-49.
2 Taylor F, Huffman MD, Macedo AF, et al. Statins for the primary prevention
of cardiovascular disease. Cochrane Database Syst Rev 2013;1:CD004816.
3 Gøtzsche PC. Deadly medicines and organised crime: How big pharma has
corrupted health care. London: Radcliffe Publishing; 2013.
4 Taylor F, Ward K, Moore THM, et al. Statins for the primary prevention of
cardiovascular disease. Cochrane Database Syst Rev 2011;1:CD004816.
3 Welch HG. Should I be tested for cancer? Maybe not and here’s why.
Berkeley: University of California Press; 2004.
5 Singer N. Forty years’ war in push for cancer screening, limited benefits. New
York Times 2009; 16 July.
6 Lee JH, Shin SW. Overdiagnosis and screening for thyroid cancer in Korea.
Lancet 2014;384:1848.
10 Marmot MG, Altman DG, Cameron DA, et al. The benefits and harms of
breast cancer screening: an independent review. Lancet 2012;380:1778-86.
12 Gøtzsche PC, Jørgensen KJ, Zahl PH. Breast screening: why estimates differ
by a factor of 20-25. J Med Screen 2010;17:158-9.
13 Gøtzsche PC, Jørgensen KJ. Screening for breast cancer with mammography.
Cochrane Database Sys Rev 2013;6:CD001877.
26 Krogsbøll LT, Jørgensen KJ, Gøtzsche PC. General health checks in adults for
reducing morbidity and mortality from disease. Cochrane Database Syst Rev
2019;1:CD009009.
27 Krogsbøll LT, Jørgensen KJ, Grønhøj Larsen C, et al. General health checks
in adults for reducing morbidity and mortality from disease: Cochrane
systematic review and meta-analysis. BMJ 2012;345:e7191.
28 Gøtzsche PC. Deadly medicines and organised crime: How big pharma has
corrupted health care. London: Radcliffe Publishing; 2013.
30 Andersen TK. 10 forslag til mere sundhed for pengene. Mandag Morgen
2011; 21 Feb.
31 Gould M. Expert panel will assess cost effectiveness of health checks. BMJ
2013;347:f5222.
32 Gøtzsche PC. "I don't want the truth, I want something I can tell Parliament!"
BMJ 2013;347:f5222 (rapid reponse).
33 Krogsbøll LT, Jørgensen KJ, Gøtzsche PC. Universal health checks should be
abandoned. BMJ 2013;347:f5227.
35 Price C. NHS Health Checks programme stalling amid poor uptake and
critical MPs' report. Pulse 2014; 28 Feb.
36 Grønhøj Larsen C, Jørgensen KJ, Gøtzsche PC. Regular health checks: cross-
sectional survey. PLoS One 2012;7:e33694.
39 Siu AL and the US Preventive Services Task Force (USPSTF). Screening for
depression in adults: US Preventive Services Task Force Recommendation
Statement. JAMA 2016;315:380-7.
42 Lee D. Google will ask: 'Are you depressed?' BBC 2017; 24 Aug.
http://www.bbc.com/news/technology-41034618.
8 Emotional pain
1 Gøtzsche PC. Deadly psychiatry and organised denial. Copenhagen: People’s
Press; 2015.
9 Danborg PB, Gøtzsche PC. Benefits and harms of neuroleptic drugs in drug-
naïve patients with psychosis: systematic review. Submitted for publication.
13 Whitaker R. Mad in America: Bad Science, Bad Medicine, and the Enduring
Mistreatment of the Mentally Ill. Cambridge: Perseus Books Group; 2002.
14 Leucht S, Kane JM, Etschel E, et al. Linking the PANSS, BPRS, and CGI:
clinical implications. Neuropsychopharmacology 2006;31:2318-25.
16 Khin NA, Chen YF, Yang Y, et al. Exploratory analyses of efficacy data from
schizophrenia trials in support of new drug applications submitted to the US
Food and Drug Administration. J Clin Psychiatry 2012;73:856–64.
17 Leucht S, Fennema H, Engel R, et al. What does the HAMD mean? J Affect
Disord 2013;148:243-8.
19 Fournier JC, DeRubeis RJ, Hollon SD, et al. Antidepressant drug effects and
depression severity: a patient-level meta-analysis. JAMA 2010;303:47–53.
20 Gøtzsche PC, Gøtzsche PK. Cognitive behavioural therapy halves the risk of
repeated suicide attempts: systematic review. J R Soc Med 2017;110:404-10.
22 Healy D. Let them eat Prozac. New York: New York University Press; 2004.
34 Gøtzsche PC. Antidepressants increase the risk of suicide and violence at all
ages. Mad in America 2016; 16 Nov.
https://www.madinamerica.com/2016/11/antidepressants-increase-risk-suicide-
violence-ages/.
41 Jensen PS, Arnold LE, Swanson JM, et al. 3-year follow-up of the NIMH
MTA study. J Am Acad Child Adolesc Psychiatry 2007;46:989-1002.
42 Molina BS, Flory K, Hinshaw SP, et al. Delinquent behaviour and emerging
substance use in the MTA at 36 months: prevalence, course, and treatment
effects. J Am Acad Child Adolesc Psychiatry 2007;46:1028-40.
43 Molina BS, Hinshaw SP, Swanson JM, et al. The MTA at 8 years: prospective
follow-up of children treated for combined-type ADHD in a multisite study. J
Am Acad Child Adolesc Psychiatry 2009;48:484-500.
44 Swanson JM, Arnold LE, Molina BSG, et al. Young adult outcomes in the
follow-up of the multimodal treatment study of attention-deficit/hyperactivity
disorder: symptom persistence, source discrepancy, and height suppression. J
Child Psychol Psychiatry 2017;58:663-78.
45 Borcherding BG, Keysor CS, Rapoport JL, et al. Motor/vocal tics and
compulsive behaviours on stimulant drugs: is there a common vulnerability?
Psychiatry Res 1990;33:83-94.
46 Breggin PR. The rights af children and parents in regard to children receiving
psychiatric diagnoses and drugs. Children & Society 2014;28:231-41.
50 Ghaemi SN. The failure to know what isn’t known: negative publication bias
with lamotrigine and a glimpse inside peer review. Evid Based Ment Health
2009;12:65-8.
51 Gøtzsche PC. Chemical or psychological psychotherapy? Mad in America
2017; 29 Jan. https://www.madinamerica.com/2017/01/chemical-psychological-
psychotherapy/.
52 Krupnick JL, Sotsky SM, Simmens S, et al. The role of the therapeutic
alliance in psychotherapy and pharmacotherapy outcome: Findings in the
National Institute of Mental Health Treatment of Depression Collaborative
Research Program. J Consult Clin Psychol 1996;64:532–9.
56 Cuijpers P, Hollon SD, van Straten A, et al. Does cognitive behaviour therapy
have an enduring effect that is superior to keeping patients on continuation
pharmacotherapy? A meta-analysis. BMJ Open 2013;26;3(4).
59 Morrison AP, Turkington D, Pyle M, et al. Cognitive therapy for people with
schizophrenia spectrum disorders not taking antipsychotic drugs: a single-blind
randomised controlled trial. Lancet 2014;383:1395-403.
66 Whitaker R. Thou shall not criticize our drugs. Mad in America 2017; 22
Sept. https://www.madinamerica.com/2017/09/thou-shall-not-criticize-our-
drugs/.
69 Pedersen AT. Debat: Vi har ret til at undre os. Journalisten 2017; 8 May.
70 Gøtzsche PC. Deadly medicines and organised crime: How big pharma has
corrupted health care. London: Radcliffe Publishing; 2013.
9 Physical pain
1 Gøtzsche PC. Deadly medicines and organised crime: How big pharma has
corrupted health care. London: Radcliffe Publishing; 2013.
2 Hench PS, Kendall EC, Slocumb CH, et al. The effect of a hormone of the
adrenal cortex (17-hydroxy-11-dehydrocorticosterone; compound E) and of
pituitary adrenocorticotropic hormone on rheumatoid arthritis. Proc Staff Meet
Mayo Clin 1949;24:181–97.
3 Moore RA, Straube S, Wiffen PJ, et al. Pregabalin for acute and chronic pain
in adults. Cochrane Database Syst Rev 2009;3:CD007076.
5 Wise J. Gabapentinoids should not be used for chronic low back pain, meta-
analysis concludes. BMJ 2017;358:j3870.
7 McCook A. NEJM issues unusual warning for readers about 1980 letter on
opioid addiction. Retraction Watch 2017; 2 June.
http://retractionwatch.com/2017/06/02/nejm-issues-unusual-warning-readers-
1980-letter-opioid-addiction /.
8 Leung PTM, Macdonald EM, Stanbrook MB, et al. A 1980 letter on the risk of
opioid addiction. N Engl J Med 2017;376:2194-5.
9 Schmidt AL, Rasmussen LI. Overlæge: Vi vil gerne tro på den magiske
medicin. Politiken 2017; 15 June.
10 Treatment of cancer
1 Gøtzsche PC. Mammography screening: truth, lies and controversy. London:
Radcliffe Publishing; 2012.
2 Rostgaard K, Vaeth M, Rootzén H, et al. Why did the breast cancer lymph
node status distribution improve in Denmark in the premammography screening
period of 1978–1994? Acta Oncol 2010;49:313 –21 .
9 Gøtzsche PC. Deadly medicines and organised crime: How big pharma has
corrupted health care. London: Radcliffe Publishing; 2013.
16 Gøtzsche PC. Hun fik den sidste kemo på vej til kapellet. Politiken 2017; 2
July.
5 Gøtzsche PC. Din overvægt er en selvskabt plage - derfor skal du selv betale.
Berlingske 2017; 12 Aug.
8 Raben A, Vasilaras TH, Møller AC, et al. Sucrose compared with artificial
sweeteners: different effects on ad libitum food intake and body weight after 10
wk of supplementation in overweight subjects. Am J Clin Nutr 2002;76:721-9.
11 Dinu M, Abbate R, Gensini GF, et al. Vegetarian, vegan diets and multiple
health outcomes: a systematic review with meta-analysis of observational
studies. Crit Rev Food Sci Nutr 2017;57:3640-9.
13 Gøtzsche PC. Deadly medicines and organised crime: How big pharma has
corrupted health care. London: Radcliffe Publishing; 2013.
17 Palatini P, Fania C, Mos L, et al. Alcohol intake more than doubles the risk of
early cardiovascular events in young hypertensive smokers. Am J Med
2017;130:967-74.
18 Smarius LJ, Strieder TG, Loomans EM, et al. Excessive infant crying doubles
the risk of mood and behavioural problems at age 5: evidence for mediation by
maternal characteristics. Eur Child Adolesc Psychiatry 2017;26:293-302.
19 Aune D, Giovannucci E, Boffetta P, et al. Fruit and vegetable intake and the
risk of cardiovascular disease, total cancer and all-cause mortality—a systematic
review and dose-response meta-analysis of prospective studies. Int J Epidemiol
2017;46:1029-56.
20 Freedman ND, Park Y, Abnet CC. Association of coffee drinking with total
and cause-specific mortality. N Engl J Med 2012;366:1891-904.
3 Elshaug AG, Moss JR, Hiller JE, et al. Upper airway surgery should not be
first line treatment for obstructive sleep apnoea in adults. BMJ 2008;336:44-5.
4 Kribbs NB, Pack AI, Kline LR, et al. Objective measurement of patterns of
nasal CPAP use by patients with obstructive sleep apnoea. Am Rev Respir Dis
1993;147:887-95.
7 Gøtzsche PC. Deadly medicines and organised crime: How big pharma has
corrupted health care. London: Radcliffe Publishing; 2013.
8 Petersen M. Our daily meds. New York: Sarah Crichton Books; 2008.
9 Larson JC, Ensrud KE, Reed SD, et al. Efficacy of escitalopram for hot flashes
in healthy menopausal women: a randomized controlled trial. JAMA
2011;305:267–74.
10 Montori VM, Isley WL, Guyatt GH. Waking up from the DREAM of
preventing diabetes with drugs. BMJ 2007;334:882–4.
11 Lee HS, Chun KH, Moon D, et al. Trends in receiving chemotherapy for
advanced cancer patients at the end of life. BMC Palliat Care 2015;14:4.
4 Ramsay HM, Goddard W, Gill S, et al. Herbal creams used for atopic eczema
in Birmingham, UK illegally contain potent corticosteroids. Arch Dis Child
2003;88:1056-7.
9 Rubinstein SM, Terwee CB, Assendelft WJJ, et al. Spinal manipulative therapy
for acute low-back pain. Cochrane Database Syst Rev 2012;9:CD008880.
11 Gross A, Langevin P, Burnie SJ, et al. Manipulation and mobilisation for neck
pain contrasted against an inactive control or another active treatment. Cochrane
Database Syst Rev 2015;9:CD004249.
12 Bertino RE, Talkad AV, DeSanto JR, et al. Chiropractic manipulation of the
neck and cervical artery dissection. Ann Intern Med 2012;157:150-2.
14 Hondras MA, Linde K, Jones AP. Manual therapy for asthma. Cochrane
Database Syst Rev 2005;2:CD001002.
15 Beckmann MM, Stock OM. Antenatal perineal massage for reducing perineal
trauma. Cochrane Database Syst Rev 2013;4:CD005123.
17 Patel KC, Gross A, Graham N, et al. Massage for mechanical neck disorders.
Cochrane Database Syst Rev 2012;9:CD004871.
19 Loew LM, Brosseau L, Tugwell P, et al. Deep transverse friction massage for
treating lateral elbow or lateral knee tendinitis. Cochrane Database Syst Rev
2014;11:CD003528.
22 Ernst E. And this is why we might as well forget about Chinese acupuncture
trials. 2014; 21 May. http://edzardernst.com/2014/05/and-this-is-why-we-might-
as-well-forget-about-chinese-acupuncture-trials /.
23 Madsen MV, Gøtzsche PC, Hróbjartsson A. Acupuncture treatment for pain:
systematic review of randomised clinical trials with acupuncture, placebo
acupuncture, and no acupuncture groups. BMJ 2008;338:330-3.
25 Sørensen TK, Pihl M. ”Vi må advare om de risici, der er, for at en nål
punkterer en eller begge lunger”. Jyllands-Posten 2017; 30 Sept.
27 So PS, Jiang Y, Qin Y. Touch therapies for pain relief in adults. Cochrane
Database Syst Rev 2008;4:CD006535. WITHDRAWN.
29 Joyce J, Herbison GP. Reiki for depression and anxiety. Cochrane Database
Syst Rev 2015;4:CD006833.
34 Bishop JP, Stenger VJ. Retroactive prayer: lots of history, not much mystery,
and no science. BMJ 2004;329:1444-6.
38 Flamm BL. Prayer and the success of IVF. J Reprod Med 2005;50:71.
41 Liddle SD, Pennick V. Interventions for preventing and treating low-back and
pelvic pain during pregnancy. Cochrane Database Syst Rev 2015;9:CD001139.
4 Gøtzsche PC. Deadly medicines and organised crime: How big pharma has
corrupted health care. London: Radcliffe Publishing; 2013.
7 Light DW, Lexchin J. Foreign free riders and the high price of US medicines.
BMJ 2005;331:958-60.
9 Stevens AJ, Jensen JJ, Wyller K, et al. The role of public-sector research in the
discovery of drugs and vaccines. N Engl J Med 2011;364:535–41.
10 Goozner M. The $800 Million Pill: the truth behind the cost of new drugs.
Berkeley: University of California Press; 2005.
12 Relman AS, Angell M. America’s other drug problem: how the drug industry
distorts medicine and politics. The New Republic 2002; Dec 16:27–41.
13 Maxmen A. Busting the billion-dollar myth: how to slash the cost of drug
development. Nature 2016;536:388-90.
24 Batt S, Fugh-Berman A. EpiPen furor: patient groups take money, stay mum.
Hastings Center 2016; Aug 29. http://www.thehastingscenter.org/epipen-furor-
patient-groups-take-money-stay-mum/.
25 Light DW, Maturo, AF. Good pharma: the public-health model of the Mario
Negri Institute. Basingstoke: Palgrave Macmillan; 2015.
35 Kim C, Prasad V. Cancer drugs approved on the basis of a surrogate end point
and subsequent overall survival: an analysis of 5 years of US Food and Drug
Administration approvals. JAMA Intern Med 2015;175:1992-4.
40 Bassand J-P, Martin J, Rydén L, et al. The need for resources for clinical
research: The European Society of Cardiology calls for European, international
collaboration. Lancet. 2002;360:1866–9.
41 The pharmaceutical industry in figures. EFPIA (European Fedration of
Pharmaceutical Industries and Associations) 2014.
http://www.efpia.eu/uploads/Figures_2014_Final.pdf.
48 Gøtzsche PC. "Human guinea pig" asks for animal studies. BMJ
2014;349:g6714.
5 Gøtzsche PC. What is it like being on the Governing Board? 2017; 27 Sept.
http://community.cochrane.org/news/what-it-being-governing-board.
6 Mandela N. Long walk to freedom. Boston: Little, Brown & Co; 1994.
ISBN: 978-87-7036-430-0
1. edition
www.peoplespress.dk