0% found this document useful (0 votes)
36 views

Hospital at night

Como se vive el hosila durante la noche y como las largas horas de trabajo afectan a los residenctes y a los acientes.

Uploaded by

Gariana Marrido
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
36 views

Hospital at night

Como se vive el hosila durante la noche y como las largas horas de trabajo afectan a los residenctes y a los acientes.

Uploaded by

Gariana Marrido
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 10

Hamilton-Fairley et al.

BMC Medical Education 2014, 14(Suppl 1):S17


http://www.biomedcentral.com/1472-6920/14/S1/S17

REVIEW Open Access

Hospital at night: an organizational design that


provides safer care at night
Diana Hamilton-Fairley1,3, John Coakley2, Fiona Moss3*
From Duty Hours: Solutions Across Borders
Quebec City, Canada. 23 September 2011

Abstract
The reduction in the working hours of doctors represents a challenge to the delivery of medical care to acutely
sick patients 24 hours a day. Increasing the number of doctors to support multiple specialty rosters is not the
solution for economic or organizational reasons. This paper outlines an alternative, economically viable
multidisciplinary solution that has been shown to improve patient outcomes and provides organizational
consistency. The change requires strong clinical leadership, with organizational commitment to both cultural and
structural change. Careful attention to ensuring the teams possess the appropriate competencies, implementing a
reliable process to identify the sickest patients and escalate their care, and structuring rotas efficiently are essential
features of success.

Introduction reduction in total hours worked over a postgraduate


“Medicine used to be simple, ineffective, and relatively (residency) surgical training program from 30,000 to
safe. Now it is complex, effective, and potentially dan- 8,000 [4].
gerous [1].” The task of reducing doctors’ working hours should
The quality of the working lives of junior doctors in the be considered against the background of continuing sig-
United Kingdom has been a matter of media and public nificant advances in medical technology and changes in
concern for many years [2]. In 1993, most doctors in the the delivery of health care. Twenty-first century inter-
country were working over 84 hours per week, and some ventions are much more effective than those available in
in acute admitting specialties were working more than the 1980s; much more care is delivered in ambulatory
100 hours. Shift lengths varied from 32 hours to over 90 settings or requires a shorter time in hospital. Working
hours with no guaranteed breaks [3]. Now, under the patterns must also be based on knowledge of research
European Working Time Regulation (EWTR), British about the impact of working at night on practitioners’
doctors-in-training work an average (over 26 weeks) of health and patient safety. In this paper we briefly outline
48 hours per week. Reductions in the working hours of some of the evidence that should inform decisions
doctors-in-training have been challenging to implement, about doctors’ working patterns and how to support
and continue to be an emotionally charged issue. Con- safe care for patients during the night. We then describe
cerns expressed publicly and privately in medical and the experience of two hospitals – both part of the UK
health care circles include the negative impact on Hospital at Night (H@N)/Taking Care 24/7 project –
patients of a perceived reduction in continuity of care. that have taken a “whole systems” approach to organiz-
The “craft” specialities question the feasibility of training ing care at night [5]. The experiences of these hospitals
doctors to become competent independent practitioners show that it is possible to reduce hours while providing
in a shorter working week. Reducing the hours worked good training and improving patient safety.
from 84 to 48 per week has contributed to an estimated

* Correspondence: [email protected]
3
London Deanery, London, UK
Full list of author information is available at the end of the article
© 2014 Hamilton-Fairley et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative
Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and
reproduction in any medium, provided the original work is properly cited. The Creative Commons Public Domain Dedication waiver
(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
Hamilton-Fairley et al. BMC Medical Education 2014, 14(Suppl 1):S17 Page 2 of 10
http://www.biomedcentral.com/1472-6920/14/S1/S17

Background therefore, night work should be reduced as much as pos-


Impact of working long hours and working at night sible. The rapid rotation of shifts (a change every few
The impact of physician fatigue on patient safety should days) is preferable to a slow rotation, as the former inter-
be the principal focus of this discussion. The number of feres less with the circadian rhythms. As well, clockwise
hours worked per week or month is less important than rotation (morning, afternoon, night) is preferable to
the impact of working long hours and, in particular, of counter-clockwise rotation in an eight-hour shift model.
working at night. At night, complex tasks become more Twelve-hour shift models with quick changeovers (e.g.,
difficult to execute, and the acuity of decision making the morning and the night shift in the same 24-hour per-
can be blunted. Medical errors, adverse events, and iod) should be avoided to allow for longer rest periods
attention failure have been noted in interns working between shifts. Finally, a later start for the morning shift
extended shifts (those greater than 24 hours) [6]. reduces the truncation of the previous sleep period, parti-
A systematic review of the impact of working long cularly for REM sleep. Factors that can ameliorate shift
hours (more than 48 per week) and shift work revealed work include workplace improvements in catering,
adverse effects on the worker’s health as well as on his supervision, health care, transportation, and recreational
or her family and social life [7]. Other reported effects facilities. Some limited evidence also suggests that bright
include an association between road traffic accidents light on the night shift might offset some of the circadian
and shifts longer than 24 hours; this risk increases with effects of day–night changes [5].
the frequency of these extended shifts [8].With this evi-
dence in mind, efforts should be concentrated on mini- The impact of daytime system inefficiencies on nighttime
mizing the number of people involved in providing care work
at night, reducing the number of night shifts per indivi- Understanding the nature of clinical work done at night
dual, and reducing the total number of hours worked. is key to developing organizational structures that sup-
port safe patient care. Many of the tasks that doctors
Doctors in training and patient safety undertake “out of hours” have been shown to be rou-
Evidence suggests that patients fare less well at night tine, “inappropriate,” or unrelated to acute need [5].
when they are cared for by doctors in training programs. Although a certain proportion of patients will require
The UK National Confidential Enquiries into Periopera- expert and sometimes life-saving medical attention at
tive Deaths (CEPOD and NCEPOD), which began in night, an analysis of the competencies required for out-
1982, found an association between perioperative death of-hours care of inpatients found that the need for spe-
and being operated on by a junior surgeon out of “nor- cialist intervention was rare; most calls required doctors
mal working hours” [9]. In light of these national ana- with generic competencies in all specialties (from ortho-
lyses, UK hospitals set up daytime “CEPOD” theatre lists paedics to oncology). On average, only five skills (e.g.,
to ensure that the majority of emergency operations were management of acute coronary syndrome, compartment
done during normal working hours. More recently, 8% of syndrome) were identified as being specialty-specific
trainees who responded to the General Medical Council (unpublished data). Nursing and allied health profes-
annual trainee survey reported that they had made a ser- sionals already perform many generic tasks, and other
ious or potentially serious error in the last month; a sub- tasks can readily be performed by practitioners in
stantial proportion of trainees attributed these errors to extended-practice or alternative health care roles (e.g.,
sleep deprivation and working longer hours [10]. physician assistants). System inefficiencies prevent the
completion of routine daytime tasks, which are rolled
Mitigating the effects of working at night over to the night, and so most nighttime calls prove not
Working patterns should reflect an approach that is to be of a “life-and-limb” nature (Figure 1) [5].
safest for patients and healthiest for doctors and other
staff. Although shift work can harm health, this harm Night teams for modern care
can be mitigated. Fatigue, but not necessarily perfor- Although changes in medical technology over the past
mance, is worse on longer shifts. Shift schedules two decades have brought about significant improve-
designed by workers encourage good performance, ments in the effectiveness of care, hospitals still require
although many workers favour longer shifts to gain working patterns that enable the safe delivery of care
longer rest periods. Frequent 24-hour shifts (more than while minimizing negative effects on the health of care
1:3 days) have been shown to increase the risk of serious providers. Two fundamental points about delivering safe
medical errors [7,11]. patient care at night have arisen from the research
Most research into minimizing the deleterious effects regarding working at night: (1) tasks done at night
of shift work has concentrated on rotating three eight- should be minimized, such that as much work as possi-
hour shifts. Night shifts are the most harmful to health; ble is done during the “normal” day and (2) only
Hamilton-Fairley et al. BMC Medical Education 2014, 14(Suppl 1):S17 Page 3 of 10
http://www.biomedcentral.com/1472-6920/14/S1/S17

care for patients at night. The project has moved on from


H@N to Hospital 24/7, as it is clear that good team func-
tioning is critical for delivering high-quality and safe care
both during the day and at night [5].
Changing to H@N from a standard “firm” and “on-
call-rota” approach to night care requires significant
organizational and cultural adaptation. We report here
the experience of two London hospitals – Guy’s and St
Thomas’ NHS Foundation Trust, a teaching hospital,
and Homerton University Hospital NHS Foundation
Trust, a district general hospital – that have successfully
implemented H@N and Hospital 24/7. These case stu-
dies illustrate not only the degree of organizational
change required, but also the benefits of this new
Figure 1 Average number of calls received per overnight shift. Data approach. A comparison of the changes made at these
source: UK National Audit 2003. two sites is outlined in Table 1.

Case 1: Guy’s and St Thomas’ NHS Foundation


competent practitioners should undertake complex tasks Trust
during night hours. Working patterns must also allow Context
doctors and other health care professionals to be trained Guy’s and St Thomas’ NHS Foundation Trust is a large
in a manner that does not put patients at risk. From the secondary and tertiary teaching hospital with an annual
evidence available, it is possible to derive principles for budget of £1 billion. It serves a deprived local popula-
designing team approaches for delivering safe nighttime tion of 1.3 million, providing regional and national refer-
care. These principles include the following: ral services in most major specialties and subspecialties.
For example, this hospital is the specialist cancer centre
• Those expected to contribute to out-of-hours work for South East London (with a population of 2.6 million)
should be involved in the rota design while ensuring and the children’s referral centre for South East England
that the duration of shifts and the hours worked (with a population of 3.5 million). This hospital operates
each week meet the agreed standards. at two sites:
• Those on-call at night should be asked to do night
work for three or four nights at a time, as this has • Guy’s – predominantly an elective surgery and
less impact on health and is safer for patients than cancer centre with a total of 350 beds
either working either weeks of nights or single • St Thomas’ – an acute-care site with 850 beds as
nights. well as the 120-bed Evelina Children’s Hospital
• Those on-call at night should be competent to
undertake the predicted work. The Guy’s and St Thomas’ NHS Foundation Trust has
• Those on-call at night should work as members of approximately one million patient contacts per year.
a team with clearly defined roles and a designated Each year there are approximately 120,000 visits to the
leader. adult emergency department and 30,000 to the chil-
• Those working at night should be encouraged to dren’s emergency department. In addition, the hospital
eat well-balanced meals and avoid caffeine. delivers approximately 6,800 babies each year.
• Provision should be made for those working at The hospital employs over 11,000 staff, including
night to take short naps in a quiet environment. 1,970 doctors, of which 870 are consultants and 720 are
• Night teams must be multi-professional so that doctors in postgraduate training (all grades of residents)
they have the necessary skill mix. and 380 in non-training grades.

Outline of problem
The Hospital at Night project In 2003 there were a total of 144 on-call rotas. These
In the United Kingdom, the pressure to reduce junior medical teams – consisting of two levels of doctors on-
doctors’ hours prompted a national project sponsored by call at any one time, with a consultant on-call from
the Department of Health: Hospital at Night (H@N). The home – were on-call only for their own patients. The
H@N project gathered evidence and provided guidance work intensity experienced by these teams varied
for the development of safe and functioning teams to according to specialty. Emergency and admitting,
Hamilton-Fairley et al. BMC Medical Education 2014, 14(Suppl 1):S17 Page 4 of 10
http://www.biomedcentral.com/1472-6920/14/S1/S17

Table 1 Comparison of changes at Guy’s and St Thomas’ NHS Foundation Trust versus Homerton University Hospital
NHS Foundation Trust
Guy’s and St Thomas’ NHS Foundation Trust Homerton University Hospital NHS Foundation Trust
Key drivers Patient safety Patient safety
Regulatory requirement to reduce doctors’ hours Regulatory requirement to reduce doctors’ hours
Controlling costs Maintaining/improving medical training
Maintaining/improving medical training
Key outcomes Reduction in HSMR Reduction in HSMR
Reduction in serious incidents No increase in serious incidents
Reduction in health care–associated infections Reduction in health care–associated infections
100% compliance with EWTR 100% compliance with EWTR
Maximum cost of £2.4 million
Data collection Analysis of on-call duties Presentations and discussions on how to improve patient care
Analysis of rotas
Creation of competency matrix
Hospital at SNPs with both clinical and site management responsibilities Clinical Site Manager Team with both clinical and site
Night Structured handover at the same time for all specialties management responsibilities
Baton bleeper for face-to-face handovers First point of contact for wards and other areas
First point of contact for wards and other areas Single team for emergency admissions via emergency
Twilight shifts for specialties (majority removed for overnight) department
On-call teams covering patients from all specialties Single team to cover inpatients
Consultant ward rounds by Surgery and General Internal
Medicine every 12 hours for all admissions
SNPs see, assess, treat, and /or refer acutely ill patients
18 pathways (with associated protocols) for common
emergencies
Taking Care 24/ Extension of H@N into the day Separation of elective and emergency work
7 Physician of the week for surgical inpatients working with Single admissions area for elective work
surgical teams Acute Care Unit
Single escalation system for both sites Doctors work only in one or the other pathway for set periods
Single admissions area for elective surgical patients of time, thereby maximizing training opportunities
Handovers for planned discharge and weekend care Consultant in General Internal Medicine present in emergency
Regular contact with wards and doctors by SNP every 6 hours area 12 hours per day
24/7 Factual handover at 8 a.m.
Elective to emergency team handover at 4 p.m.
Impact on Initial reduction in HSMR Initial reduction in HSMR
patient care Sustained reduction in serious incidents Reduction in health care–associated infections
Reduction in health care–associated infections
Reduction in in-hospital cardiac arrests
Reduction in lengths of stay
Financial impact H@N: £4.1 million saving; £2.4 million in recurrent costs H@N: £100,000 saving
24/7: closure of 250 beds 24/7: £600,000 saving; £250,000 in recurrent costs
Educational H@N: no change in feedback from junior doctors H@N: no change in feedback from junior doctors
impact 24/7: improved teaching time participation for most junior staff Sustained hours of direct supervision/elective work
and physicians (daily seminar from physician of the week) Reduction in hours spent in “acute care team” for each doctor
Educational handover at 10 a.m.
Lessons learned Need involvement of all staff, not just medical Need involvement of all staff, not just medical
Need good, relevant data Need good, relevant data
Training for staff who are extending/changing their role Training for staff who are extending/changing their role
The change is part of a whole system change that continues to The change is part of a whole system change that continues to
evolve; 24/7 is only one part that contributes to the evolve; 24/7 is only one part that contributes to the
improvement as a whole improvement as a whole
Sustainability Yes – no appetite to return to the on-call system Yes – no appetite to return to the on-call system
EWTR = European Working Time Regulation; HSMR = Hospital Standardised Mortality Ratio; SNP = senior nurse practitioner

General Internal Medicine, Obstetrics, Anaesthetics, The junior member of the medical team was typically
Intensive/High Dependency Care, and Pediatrics teams the first to be called to attend to a patient. Senior doc-
tended to work hard throughout the day and night. In tors were usually called later and only if the junior doc-
contrast, general ward patients tended to require medi- tor thought it was necessary. It is well documented that
cal attention only until 8:30 p.m., and most other speci- this was often the root cause of a poor outcome [9].
alty physicians slept for at least five hours per night The hospital’s risk record in 2002 included 16 serious
without being disturbed. adverse incidents that were considered to be avoidable.
Hamilton-Fairley et al. BMC Medical Education 2014, 14(Suppl 1):S17 Page 5 of 10
http://www.biomedcentral.com/1472-6920/14/S1/S17

There were also a significant number of cases of methi- priority, starts treatment according to agreed-upon
cillin-resistant Staphylococcus aureus (MRSA) (81 per protocols, and refers the patient to the relevant doc-
year). In addition, the Hospital Standardised Mortality tor, physiotherapist, nurse, etc.
Ratio (HSMR) was at the United Kingdom average. • There is an agreed-upon escalation policy that
Over a six-month period ending in August 2004, the involves, as necessary, referral to the Medical Emer-
hospital started to reduce junior doctors’ working hours gency Team (critical care outreach). (See Figure 2
to an average of 48 per week to comply with the for additional details.)
EWTR. As the hospital’s first priority was patient safety, • Handover was synchronized across the Trusts, and
any proposed solution had to improve patient care. a 30-minute period was built into shift times for
handover. (See Figure 3 for more details.)
Key measures of improvement • Each team holds a “baton” bleeper that, at the end
The vision was to reorganize care so that all patients of shift, has to be handed over. This ensures that
had access to the right person with the right skills for there is a face-to-face handover.
their needs at the right time. Patient care and patient • Specialist teams (referred to as twilight teams) stay
safety were the prime goals, along with the achievement on-site until 9 p.m. (with handover at 8.30 p.m.) to
of the EWTR. Identified outcome measures included a complete all routine tasks and ensure that all their
reduction in serious adverse incidents, a reduction in patients are stable with no ongoing needs.
the number of cases of MRSA, and a reduction in the • A representative from each twilight team attends
HSMR. handover to report on all patients requiring input/
additional observation through the night.
Process of gathering information • The overnight non-acute rotas were reduced to a
Colleagues in Human Resources collected information minimum. As well, junior surgical rotas were merged
about all medical rotas. A financial costing was done of and staff were shared.
maintaining on-call rotas. In addition, an analysis of on-
call work from 5 p.m. to 8 a.m. was undertaken. Meet-
ings to understand patients requirements out of hours
were held with all staff, including porters, transport,
medical records, and IT. This allowed us to identify all
required competencies in addition to clinical competen-
cies, such as the ability to access databases and medical
records, equipment such as wheelchairs, and services
such as radiology, the Intensive Therapy Unit, High-
Dependency Care, and the morturary.

Strategy for change


A review of all data showed that better care could be
provided on-call by fewer people with the introduction
of three multi-professional teams across the hospital
sites, with a fourth for the Evelina Children’s Hospital.
In addition, only minimal changes were made to the
emergency admission teams. The strategy was operatio-
nalized as follows:

• Each team is led by a senior nurse practitioner


(SNP) who takes all calls from all areas. This strategy
of using SNPs as leaders provides consistency
because these individuals are permanent staff who
have a thorough knowledge of their hospital and
ward staff, and they are able to provide effective site
management, including bed management.
• All patients are given a score (Patient at Risk
[PAR]) at each point of observations, and any Figure 2 Escalation of care for sick patients at Guy’s and St.
changes in their scores are reported to the SNP who Thomas’ NHS Foundation Trust. SNP = senior nurse practitioner; PAR
= patient at risk.
then assesses the patient, determines his or her
Hamilton-Fairley et al. BMC Medical Education 2014, 14(Suppl 1):S17 Page 6 of 10
http://www.biomedcentral.com/1472-6920/14/S1/S17

• Training of all ward staff to calculate accurate PAR


scores, with a weekly audit to ensure this was being
done correctly.
• Training on induction of the surgical trainees so
that they could cross cover urological and ear, nose,
and throat emergencies.

Effects of change on patients


The analysis of benefits of the night teams – which had
been first introduced in 2004 – took place in 2009.
Overall, analysis showed that patients are seen in a
Figure 3 Average total calls received per day at Guy’s Hospital, timely fashion and that more than 95% of the time rou-
London, UK, after the implementation of Taking Care 24/7, 2008– tine tasks are completed before the night shift starts.
2009. Data source: Senior Nurse Practitioners, Guy’s and St Thomas’
Foundation Trust.
More importantly, we found only one serious adverse
incident in 2007; this was maintained to less than three
per annum up to 2011. There was also a reduction in
• Some specialties (e.g., Obstetrics, Anaesthesia, HSMR to 0.87 in first year of implementation; the hos-
Neonatology, Intensive Care) could not be included pital has remained in the top three in London since that
because of the requirement for specialty-specific time. In addition, there was a reduction in MRSA cases
competencies throughout the 24 hours. from 81 per annum to 7 per annum.
• Team Standard Operational Procedures were
developed. These included roles, competency check- Effects of change on education
lists, handover, escalation policies for acutely sick On the night team, most junior doctors work alongside,
patients, and agreed-upon protocols and guidelines and are supported and supervised by, SNPs who are
for the most common conditions occurring over- trained to do procedural assessments. Under this sys-
night. Nurses were trained to prescribe against tem, the junior doctors’ education continues overnight.
agreed-upon protocols for commonly occurring con- More senior medical staff are always available for com-
ditions (e.g., oliguria, asthma, cardiac failure, pain plex cases, and junior staff are never asked to perform
relief). Competencies were checked at induction on beyond their competence. Reports from junior staff
starting work in the Trust and confirmed with edu- include high praise for the SNPs, the handover is con-
cational supervisors from the doctor’s portfolio of sidered effective and educational, and those in Founda-
training/log book. tion years (the first two years post-qualification) have
requested to do more night shifts. More senior surgeons
£2.4 million was available for this project. The (more than five years post-qualification) are on-site until
resources were used to pay for the following: 11 p.m., as no surgical procedures are done at night
unless they are considered to be life-saving. This has
• A clinical champion for one day per week who minimized the effect on training during daytime hours.
worked with a project manager (full-time). Consultants are called in a little more frequently; how-
• Human Resources time to analyze and help redesign ever, this is always considered to be appropriate (e.g.,
rotas with input from clinicians and junior doctors. for life-saving surgery).
• Training for the SNPs in advanced clinical skills,
including full patient assessment (particularly cardi- Effects of change on finances
orespiratory and neurological systems, interpreting The estimated cost to maintain the original 144 on-call
electrocardiograms, chest X-rays, and performing rotas in 2003 was £7 million. As described, it cost £2.4
and interpreting arterial blood gases). million to introduce the changes. The number of rotas
• Increasing the number of doctors on the on-call was reduced to 102, of which only 25 are physically pre-
roster for emergency surgery from six to nine to sent on-site. The remainder go home at 9 p.m. Only 15
achieve compliance. remain on-duty from home, responding by phone more
• Increasing the number of SNPs. than 80% of the time; however, they are expected to
• Funding for advanced nurse practitioners in some attend within 30 minutes if required. Consultants are
areas (e.g., Emergency, Gynecology, Ophthalmology) also on-call from home. This system reduced the recur-
to reduce the amount of work requiring doctor rent costs from £7 million per annum to £2.6 million
input. (in 2009 dollars).
Hamilton-Fairley et al. BMC Medical Education 2014, 14(Suppl 1):S17 Page 7 of 10
http://www.biomedcentral.com/1472-6920/14/S1/S17

Other changes introduced at the same time included SNP for the first 12 hours; this is in addition to
the following: increased observation times by the nurses on the wards.
Surgeons can now spend longer in theatre, as physicians
• Extending this system into the day. are on duty for post-operative medical care. Residents
• Developing a medical perioperative support team and other doctors-in-training report that they are being
for the elderly. supported and that they receive good training and
• Naming a physician for each week to oversee the teaching.
elective surgical and oncology patients; this has
reduced the in-patient length of stay and reduced Case 2: Homerton University Hospital NHS
the number of in-hospital cardiac arrests from an Foundation Trust
average of 13 per month to 3 per month for the Context
nine months between March and December 2011 Homerton University Hospital is a 550-bed district gen-
(with no in-hospital cardiac arrests for 102 consecu- eral hospital located in Hackney, an inner city area with
tive days). a multi-ethnic, deprived population of approximately
250,000 to 300,000 people. The hospital provides acute
Since 2004, shortening the length of stay has led to a hospital and community services for the local popula-
reduction in the bed base by 250 beds, despite an overall tion. It also offers some tertiary services, including bar-
increase in in-patient episodes of 13% (average of 3% iatric surgery, neurological rehabilitation, neonatal
per annum). This reduction in the bed base has saved intensive care, and high-risk maternity care.
£1 million per annum per 25 beds closed and has On average there are 120,000 emergency department
increased the income generated as a result of the visits per year. In addition, the hospital delivers 5,000
increased number of Finished Consultant Episodes (the babies each year. The annual turnover is in excess of
basis on which NHS acute trusts are paid). £230 million.

Lessons learned Outline of problem (patient-centred)


This major cultural change took time both to start and to There was a recognition in 2003 that significant and
embed. The main hurdle was for specialists to fully trust robust service redesign was the only way to guarantee
that the overnight urgent care needs of their patients were good quality care and patient safety day and night while
mostly generic and that they very rarely required specialist also providing adequate training opportunities, control-
intervention. The impact on the hospital has been more ling costs, and meeting the EWTR (i.e., 48 hours per
far-reaching than we had imagined. What began as H@N week with a maximum shift length of 13 hours) without
has enabled the introduction of many other organizational increasing the number of doctors required.
changes that have improved care 24/7.
Process of gathering information
Message for others In 2003 we monitored all junior doctors’ activity during
It is key to have a senior clinical champion who believes their night shift, in some cases by using diaries and in
that this change will improve patient care. It is also others by shadowing them overnight.
important to do the following: collect initial data; engage This exercise revealed the following:
in detailed discussions and ensure that there is sufficient
time to engage with all groups of staff; train those indi- • Activity was variable across grades and specialties
viduals whose roles are changing; and focus on the qual- from minimal to extremely busy.
ity and safety of care as well as on the number of hours • Some work should have been completed during
worked. The cultural change is perhaps greatest for doc- the day or could have waited until the next day.
tors, but nurses had to be convinced that they are the • Some work did not need to be done by doctors.
most appropriate professional to lead the teams and that • “Life-and-limb-threatening” work was relatively
they have the capability to enhance their roles. rare, and sometimes it should have been done by
more senior medical staff.
Next steps • There was evidence of poor systematic handover of
The day and night systems are now the same. There is patients.
now a single point of contact, available at one number
24/7, for urgent care. SNPs work proactively by “round-
ing” all wards every six hours. All patients who are Strategy for change
stepped down from level 3 (Intensive Care) and 2 (High We introduced H@N in 2003 with a Clinical Site Man-
Dependency Care) are reviewed every 4 hours by the ager Team. In August 2007, two years ahead of the
Hamilton-Fairley et al. BMC Medical Education 2014, 14(Suppl 1):S17 Page 8 of 10
http://www.biomedcentral.com/1472-6920/14/S1/S17

2009 deadline, we took this to the next stage: “Taking As Taking Care 24/7 evolved, it became obvious that
Care 24/7,” with a maximum 48-hour week for all we had made a number of errors that had to be cor-
doctors. rected. For instance, handovers were often unfocused
Stage 1 – Hospital at Night (2003) and lengthy, and they were occurring too frequently.
Clinical Site Managers – highly trained nurses usually To remedy this situation, we changed handovers to a
with a background in acute/emergency/critical care brief factual handover at 8 a.m., a more contemplative
medicine – were introduced to coordinate the night and educational handover after the post take ward
work and to be the first on-call. These managers round at 10:30 a.m., and a cold to hot team handover
screened out tasks that did not need to be done by a at 4 p.m. The night handover continued as it had as
doctor and ensured that the most-appropriate doctor, part of H@N.
one who was often not the most junior, was called to
perform any required tasks. Having this one point of Resources needed
initial contact also helped to prevent multiple calls For both service redesign projects we employed a pro-
being made to different doctors for the same task. ject manager for a little over 12 months. We used no
These individuals also supervised – and ensured maxi- other additional resources, although most staff input
mum attendance at – the hospital-wide nighttime hand- took place during normal working hours and, therefore,
over. In addition, an analysis of work performed by theoretical costs were clearly involved.
resident on-call junior doctors showed that many of
them had little to do overnight. Accordingly, these doc- Effects of change on patient care
tors were removed from the on-call rotas. We looked at a number of indicators to assess the
Some senior doctors were concerned that these impact of these changes on patient care. The HSMR fell
changes were driven by agendas other than patient care shortly after we introduced Taking Care 24/7, before
(e.g., to save costs or to make consultants work more gradually returning to baseline (Figure 4). It is important
hours), and this was partly true. To help alleviate these to emphasize that improvements to patient care cannot
concerns, a series of presentations and discussions were be attributed solely to the changes in the junior doctors’
held for all grades of medical staff, other clinicians, and hours of work. The way services were delivered changed,
managers to describe how patient care could be in particular the introduction of an extended working
improved with the new approach. In a short amount of day for senior doctors. In addition, the elective and
time the value of the Clinical Site Manager Team as emergency admissions were each placed in a specific
first point of contact for out-of-hours care was recog- location with dedicated medical and nursing teams spe-
nized and welcomed. cific to each group with no possibility for split or con-
Stage 2 – Taking Care 24/7 (2007) fused duties for each group of staff. Other, less
Staff were divided into elective (“cold”) and emergency quantifiable, things inevitably changed at the same time.
(“hot”) teams. Trainee doctors spent time, usually peri- Nevertheless, we implemented what had been regarded,
ods of weeks, on each team, and were thus able to focus by some, as dangerous reforms to junior doctors’ hours
exclusively on either “hot” or “cold” work for both ser-
vice delivery and training purposes at different times.
Following a review of patient flows prior to imple-
menting Taking Care 24/7, all elective surgical admis-
sions were directed to a single ward area. From there,
they went to theatres and, finally, to their in-patient
recovery ward. This reduced congestion on receiving
wards, as patients scheduled for discharge could stay
until mid-morning, and also prevent the unnecessary
displacement of newly admitted patients. All emergency
admissions (except maternity and children) were
admitted to the Acute Care Unit, which was staffed by
the Acute Care Team.
Importantly, between 2003 and 2007 we had already
changed the working patterns of consultants so that
they were required to provide senior cover for an Figure 4 Quarterly in-hospital relative risk of mortality, with 95%
extended 12-hour working day for emergency care on confidence intervals, for all non-elective admissions, Homerton
weekdays, with a slightly shorter time commitment on University Hospital, London, UK. Values shown in green are
statistically significant deviations from expected values.
weekends.
Hamilton-Fairley et al. BMC Medical Education 2014, 14(Suppl 1):S17 Page 9 of 10
http://www.biomedcentral.com/1472-6920/14/S1/S17

without detriment to patient safety. The following It is essential to focus on the demands of the service
points, in particular, were observed: and the training needs, not on the hours per se. Dealing
with the hours of work in isolation – or even as the
• There were no deleterious effects on length of stay prime mover – will not provide solutions, but will
or the number of admissions and readmissions. instead exacerbate problems and fault lines in the cur-
• No significant untoward incidents were reported at rent system.
the time of implementation or since that are attribu-
table to the changes. Current approach
• The incidence of health care–associated infections We plan to continue with the service redesign. It is
fell – and continues to fall – after the implementa- becoming clear that a senior consultant presence is
tion of Taking Care 24/7. required seven days a week and even, possibly, around
the clock for acute care. This currently presents major
financial challenges.
Effects of change on education
We were able to maintain hours of direct supervision of Discussion
trainees by consultants in their elective time because Working excessive hours puts patients at unacceptable
during that time there was no night on-call or weekend risk, is bad for the health of practitioners, and increases
work. Taken over a whole year, the time devoted to the risk of both accidents in the workplace and car acci-
cold shifts was maintained, while that devoted to hot dents after leaving work. Health care in general – and
shifts was reduced. medicine in particular – has been slow to develop ways
of working that enable safe delivery of modern medical
Effects of change on finances care within a safe environment for the staff who are
The changes associated with H@N saved approximately delivering that care. Changes to working hours have
£100,000 per year (in 2003 dollars). The savings asso- perceived, but not proven, negative effects on continuity
ciated with Taking Care 24/7 were approximately of care and the training of junior doctors. The case stu-
£600,000. For the latter, additional costs incurred were dies we describe outline how it is possible to organize
£250,000 (in 2007 dollars) to pay for the increased nur- health care staff so that the twin aims of safety for
sing presence and some junior staff numbers. Both of patients and staff and a good environment for training
these sums were recurrent. Had we not radically chan- can be realized. While the principles might be the same,
ged the service model, the additional costs would have it is clear that different hospitals will take somewhat dif-
been very much greater and no savings would have been ferent approaches to the task of maintaining sufficient
realized. and appropriate medical cover for patients being
The savings were achieved by reducing all doctors’ admitted as emergency or elective patients. An overview
hours of work to 48 per week and removing some from of the systematic approach described in these two hospi-
resident on-call work completely. tals is summarized in Table 1.
The extended role of nurses in both of these models is
Lessons learned striking, in particular their role as coordinator for the
The key to successful implementation of these changes hospital site and for the clinical assessment of deterior-
was to involve those responsible for delivering the ating patients. The availability of, and appropriate train-
changes in the planning process and to manage the ing for, these staff members is crucial to the success of
changes in clear phases. We used evidence from the the system. Initial fears that this would erode the role of
first phase to build up confidence; this allowed us to the doctor have not been borne out as this system tends
progress to the further phases of implementation. It was to allow doctors to be called appropriately to optimize
important for both system changes to encourage direct the management of patient care.
and indirect feedback so that problems could be dealt The coordination of all admissions and discharges by
with and successes celebrated as soon as possible. a single person who is aware of the organization of the
whole hospital is key to the success of separating the
Message for others elective and acute pathways for patients. This also
Effective communication is vital to implementing change allows teams to concentrate on one pathway without
of this magnitude. Some of the better suggestions for worrying about the other. This appears to improve
improvement to the system were made by doctors in patient care through improved teamwork and increased
training, those who were living through these changes efficiency in moving through the elements of the path-
to the way they worked. Feedback, even hostile, must be way facilitated by a single team. Concentrating on the
encouraged and actively sought. way of working and minimizing the opportunity for
Hamilton-Fairley et al. BMC Medical Education 2014, 14(Suppl 1):S17 Page 10 of 10
http://www.biomedcentral.com/1472-6920/14/S1/S17

medical errors (reducing fatigue through proper rest and Authors’ details
1
Guy’s and St Thomas’ NHS Foundation Trust, London, UK. 2Homerton
minimizing disruption to the circadian rhythm) also
University Hospital, NHS Foundation Trust, London, UK. 3London Deanery,
appears to minimize costs and can lead to sustained London, UK.
financial savings.
Published: 11 December 2014
Work on competencies required for overnight care has
allowed doctors with generic skills to be available for
References
overnight emergencies for a group of specialties (e.g., for 1. Chantler C: The role of education in the delivery of health care. The
surgical patients, junior residents can cover inpatients Lancet 1999, 353:1178-1181.
2. McKee M, Black N: Hours of Work of Junior Hospital Doctors: Is there a
from all surgical specialties with distant support from
Solution? J Manage Med 1991, 3:40-54.
senior residents for the rare specialty-specific 3. Dyer C: Junior doctors’ hours; fighting fund set up after writ. BMJ 1989,
emergency). 298:910-911.
4. Chikwe J, de Souza AC, Pepper JR: No time to train the surgeons. BMJ
Senior residents can be on-call from home, taking
2004, 328:418.
referrals from the H@N team or other centres. A single 5. Department of Health: The implementation and impact of Hospital at
team was needed to cover admissions for the emergency Night pilot projects: An evaluation report. 4117 [http://www.dh.gov.uk/
en/Publicationsandstatistics/Publications/PublicationsPolicyAndGuidance/
department. Thus fewer doctors are present in the hos-
DH_968].
pital, although they work harder in short bursts. The 6. Barger LK, Ayas NT, Cade BE, Cronin JW, Rosner B, Speizer FE, Czeisler EA:
key to the success of such a system is that all clinical Impact of extended-duration shifts on medical errors, adverse events,
and attentional failures. PLoS Med 2006, 3:e487.
professionals work together with an organized coordi-
7. Harrington JM: Health effects of shift work and extended hours of work.
nated approach that includes assessment, prioritization, Occup Environ Med 2001, 58:68-72.
treatment, and, if appropriate, further referral. The team 8. Barger LK, Cade BE, Ayas NT, Cronin JW, Rosner B, Speizer FE, Czeisler CA:
Extended work shifts and the risk of motor vehicle crashes among
leader, usually a senior advanced nurse practitioner,
interns. N Eng J Med 2005, 352:125-134.
receives all calls and coordinates all activity within the 9. National Confidential Enquiry into Patient Outcome and Death: Deaths in
hospital. Acute Hospitals: Caring to the End? 2009 [http://www.ncepod.org.uk/dah.
htm].
The important messages are that changing from a
10. Postgraduate Medical Education and Training Board: National Training
standard approach to doctors’ duty hours to a coordi- Surveys 2008-2009 Key findings. 2008 [http://www.gmc-uk.org/
nated H@N or Hospital 24/7 approach requires signifi- National_Training_Surveys__09_20090929.pdf_30512348.pdf].
11. Czeisler CA, Moore-Ede MC, Coleman RH: Rotating shift work schedules
cant organizational change that will have an impact on
that disrupt sleep are improved by applying circadian principles. Science
the working practices of many clinicians. This change is 1982, 4558:460-463.
not easy, and it requires considerable determination and 12. Moss F: Junior doctors’ pay: a wake-up call to the NHS. BMJ 2000,
320:1224.
continuing input from senior staff to embed and main-
tain it. The difficulty lies in persuading many dedicated doi:10.1186/1472-6920-14-S1-S17
individuals that working differently will indeed provide Cite this article as: Hamilton-Fairley et al.: Hospital at night: an
organizational design that provides safer care at night. BMC Medical
better and safer care. The hospitals in the United King- Education 2014 14(Suppl 1):S17.
dom that have implemented this change have all shown
improvement in patient safety, with some evidence of
reduced costs overall.

Competing interests
The authors declare that they have no competing interests.

Declarations
Resources and secretariat support for this project was provided by the Royal
College.
This article has been published as part of BMC Medical Education Volume 14
Supplement 1, 2014: Resident duty hours across borders: an international
perspective. The full contents of the supplement are available online at Submit your next manuscript to BioMed Central
http://www.biomedcentral.com/bmcmededuc/supplements/14/S1. This and take full advantage of:
article was submitted and peer reviewed in 2012. Final acceptance for
publication as part of this supplement was in 2014.
• Convenient online submission
Publication of this supplement was supported and funded by the Royal
College of Physicians and Surgeons of Canada. The funding agency played • Thorough peer review
no role in the design, in the collection, analysis and interpretation of data; in • No space constraints or color figure charges
the writing of the manuscripts; and in the decision to submit the
• Immediate publication on acceptance
manuscript for publication. The articles have been through the journal’s
standard peer review process for supplements. The Supplement Editors • Inclusion in PubMed, CAS, Scopus and Google Scholar
declare that they have no competing interests. • Research which is freely available for redistribution

Submit your manuscript at


www.biomedcentral.com/submit

You might also like