Designing Methodology

Introduction
• Research methodology is a technique of systematically answering the research
question.
• Researchers need it not only to know the research methods or techniques but also
the methodology too Designing the methodology refers to the development of a
system or method or plan for a solving research question at hand.
• The key purpose of design methodology is to find the best solution for each
experimental condition.
• It involves brainstorming to encourage innovative ideas and collaborative thinking to
work through each proposed idea and achieve the best solution.
• To design a methodology that comply the needs and desires of the end users is the
most significant concern.
• Design methodology can employ basic research methods which involves analysis as
well as testing.
Design Methodology in Technology
• While design methodology is employed in many industries, it is commonly applied in
technology fields, including those using the internet, software and information systems
development Several design methodology approaches have developed in the technology
industry.
• Each was a reaction to a different type of problem.
• Some common technology design methodologies include:
1. Top down Design or Stepwise Refinement:
- This starts from the end solution and works
2. Bottom up Design:
- This methodology starts with a foundation and works up towards backwards, refining each
step along the way.
3. Structured Design:
- This is an industry standard.
- The technique starts by identifying inputs and desired outputs to create a graphical
representation.
4. Structured Analysis and Design Technique:
- This approach utilizes a diagram to describe the hierarchy of a system's
functions.
5. Data Structured Systems Development:
- Data structure determines the system structure in this methodology
6. Object Oriented Design:
- This methodology is based on a system of interacting objects.
Designing an Experiment:
• The major work in designing an experiment involves defining the
problem and related questions to be addressed, defining the
population of interest, determining the need for sampling and
defining the experimental design.
• In order to design an experiment following steps are performed.
(a) Formulating the Research problem
(b) Extensive literature survey:
(c) Development of working hypothesis
(d) Preparing research design
(e) Determining sample design
(f) Collecting the data
(g) Execution of the project
(h) Analysis if data
(i) Hypothesis-testing
(j) Genitalizations and interpretations
(k) Preparation of the report
Sample size Determination
• In research methodology a sample is a set of data collected and/or selected
from a statistical population by a defined method.
• The elements of sample are sample points and sampling units.
• A most important and critical aspect of any research study is determining the
appropriate sample size to answer the research question.
• Sample size determination means selecting the number of observations to
include in a statistical sample.
• The sample size is an important feature in making inferences about a
population from a sample.
• The sample size in a study is determined on the basis of data collection costs
and efforts and the need of sufficient statistical power.
• Studies should be designed to include a sufficient number of samples to
adequately address the research question.
• Studies with either an insufficient or an excessively large number of samples are
wasteful in terms of time, resources, investigational efforts etc.
• These situations can be un ethical as researcher put himself at risk if unable to
answer research question.
• Studies that are much larger than they need to be to answer the research questions
is also wasteful. T
• he formulas resented in following section generate estimates of the necessary
sample size(s) required based on statistical criteria.
• In order to determine desired sample size and to accomplish the experimental
objectives, the level of significance is usually, but not always, set at the traditional
value of 5%.
• The consequences of committing an error should be considered are fully.
• It is preferable to make significance level as small as 1% to avoid costly consequences
resulting out of missing a true difference of practical significance,.
• The difference to be detected should be specified.
• In addition, knowledge of the SD for the significance test is necessary.
 Considerations for Sample Size:
1. Confidence:
➢Confidence is the idea of being certain that the estimate based on the sample
correctly represents the population.
➢This is used in association with an interval, within which the true and unknown
population value is expected to reside.
➢Confidence provides the result's ability to be convincing. Low confidence implies
the results based on the sample data e not convincing.
2. Significance:
➢Significance is the idea that the results are not due to random chance one.
➢It is the notion that there is convincing evidence based on the sample data that
there ally is a difference.
➢This is commonly used when accepting the alternative hypothesis with a specified
level of statistical significance.
3. Variance:
➢The higher the population's variation or spread, the more samples will be
needed to determine the same result over a smaller variation population.
➢The chance of selecting samples further from the mean value will require more
samples to get an accurate estimate than the population that has a very tight
variance.
➢The more precise or smaller the difference that we want to detect, the more
samples will be required.
4. Population size:
➢The larger is the population, larger will be the sample size.
5. Appropriate sample size:
➢When determining the sample size for an experiment, selection of proper sample
size is must to make conclusions about a population.
➢Selection would be made on the basis of sampling risk, population's variance and
amount of change be detected.
➢These factors affect the sample size selection.
➢To minimize risk related to the sample selection that represents the population
better is to consider large samples.
➢For population with homogenous units' small sample size is enough but for
population that has heterogeneous units large sample size is must to obtain reliable
results.
➢In order to eliminate sampling risk, it is advisable to measure every unit in the
population using a method that has very little or no measurement error.
6. Nature of study:
➢If study is intensive and continuous, small sample size is enough but for
studies which are not likely to be repeated may select large sample size.
7. Other considerations:
➢The availability of trained expert, use of appropriate sampling technique,
categorization of population, time constraints, budgetary provisions etc. are
all other considerations for selecting sample size.
Formula for Sample Size:
1. One Sample, Continuous Outcome:
• In designing studies, required sample size is estimated to minimize
the margin of error.
• The margin of error (E) in the one sample confidence interval (𝜇) can
be calculated as:
𝛔
• 𝐄=𝐙
𝐧
• Where, n is sample size,
E is margin of error,
Z is the value from the table of probabilities of the standard
normal distribution for the desired confidence level (for example, Z =
1.96 for 95% confidence) and a is the SD of the outcome of interest.
• The goal is to determine the sample size which ensures that the margin of error
does not exceed a specified value.
• In order to calculate n, the above equation can be simplified as “
𝐸 𝑛 = 𝑍𝜎
𝑍𝜎
𝑛=
𝐸
• Now, square both sides of the equation to get:
𝑍𝜎 2
𝑛=
𝐸

• The formula in equation generates the sample size (n) required to ensure that the
margin of error (E) does not exceed a specified value.
• Sometimes it is difficult to estimate SD of the outcome of interest.
• Using sample size formula in equation one can plan study to estimate the unknown
mean of a particular outcome variable in a population.
• Continuous variable:
➢A continuous variable is one in which the outcome for each subject consists of
a measurement on a numerical scale and the results would usually be
summarized as means (the sum of all the measurements divided by the
number of subjects), and the effect size is summarized as the difference
between the means of the groups compared.
➢ For example, height, weight, temperature etc.
• Dichotomous variable:
➢A dichotomous variable is a nominal variable which have only wo possible
values of categories or levels when observed or measured.
➢The variable value is host often represented as a measured variable, for
example age as under 50 or 50 and over, or an attribute, for example gender
as male or female; or ideas, for example response as yes/no.
Example: An investigator wants to estimate the mean systolic blood pressure in
children with congenital heart disease who are 3 to 5 years of age. How many
children should be enrolled in the study? The investigator plans to use 95%
confidence interval (Z=1.96) and a margin of error of 5 units. Given that, SD is
between 15 and 20.
Solution: To estimate the sample size, first consider the larger SD in order to
obtain the most conservative (largest as possible) sample size.

𝑍𝜎 2
𝑛=
𝐸

1.96 ×20 2
= = 61.5
5

In order to ensure 95% confidence interval, estimate of the mean within 5 units of
the true mean, a sample of size 62 is needed.
 2. One Sample, Dichotomous Outcome:
• To estimate the proportion of successes in a dichotomous outcome variable in
a single population, the formula for determining sample size is:
𝒁 𝟐
n = P(1-P)
𝑬
• Where, Z is the value from the standard normal distribution reflecting the
confidence level that will be used (Z = 1.96 for 95 %),
• E is desired margin of error and
• P is the proportion of successes in the population.
3. Two Independent Samples, Continuous Outcome:
• In order to estimate the difference in means between two independent populations,
the formula to determine the sample sizes required in each comparison group is given
by equation

𝒁𝝈 𝟐
𝒏𝒊 = 𝟐
𝑬

• Where, n, is the sample size required in each group (i= 1, 2),

Z is the value from the standard normal distribution reflecting the confidence
level that will be used and
E is the desired margin of error, and a is the SD of the outcome variable.
 4. Matched Samples, Continuous Outcome:
• In order to estimate the mean difference of a continuous outcome based on
matched data, the formula used for determining sample size is:

𝑍𝜎𝑑 2
𝑛=
𝐸

• Where, Z is the value from the standard normal distribution reflecting the
confidence of t level that will be used (Z = 1.96 for 95%).
• E is the desired margin of error, and
• 𝜎𝑑 is the SD of the difference scores.
5. Two Independent Samples, Dichotomous Outcome:
• In order to estimate the difference in proportions between two independent
populations that is to estimate the risk difference, the formula for determining
the sample sizes requires in each comparison group is:
𝑍 2
𝑛𝑖 = 𝑃1 1 − 𝑃1 + 𝑃2 1 − 𝑃2
𝐸
Where , 𝑛𝑖 , is the sample size required in each group (i = 1, 2).
Z is the value from the standard normal distribution reflecting the
confidence level (Z = 196 for 95%), and
E is the desired margin of error.
P₁ and P₂ are the proportions of successes in each comparison group.
 (c) Steps to Calculate Sample Size:
• Once information is gathered, sample size can be calculated using a formula given in equation
or using any of the specialized software packages, depending on the complexity of the
analysis.
• Regardless of which way it is calculated, following 5 steps need to be performed.
1. Specify a hypothesis test: Most studies have many hypotheses, but for sample size
calculations, choose one to three main hypotheses. Make them explicit in terms of a null and
alternative hypothesis.
2. Specify the significance level of the test: It is usually a 0.05, but it does not have to be.
3. Specify the smallest effect size of interest: This is often the hardest step. The point is to
specify the smallest effect size of scientific interest.
4. Estimate the values of other parameters necessary to compute the power function: Most
statistical tests have the format of effect/standard error. Standard error is generally SD/n. To
solve for n. we need a value for SD. It can be made available from pilot study or from
historical data of another study.
5. Specify the intended power of the test: The power of a test is the probability of finding
significance if the alternative hypothesis is true. A power of 0.8 is the minimum. If it is
difficult to rerun the study or add a few more participants, a power of 0.9 is better.
 Power of Study
• Determining the optimal sample size for a study assures an adequate power to
detect statistical significance.
• The statistical power of a study is its ability to detect a difference really exists.
• The power depends on the sample size (number of subjects), and the effect size
(difference in outcomes between two groups).
• For common studies that involve comparison between two groups, for example
comparison of blood pressure levels between smokers and non-smokers.
• For such small studies, the t-test is usually used and the power of the study is
relatively small and easy to compute if the sample size and the difference in blood
pressure between the two groups are known.
• Many such small studies are under-powered to detect a true difference because they
do not have enough sample size, and research ends up with a large insignificant p-
value.
• The major reason for the lack of significance is a sample size and not an effect size.
• The freely available software package G*Power can be used to compute power of
the study.
• It also determines the desired effect size, or the sample size, for a given power.
• This is very useful software for planning sample size for a study as well as post-hoc
analysis of studies to examine if they had enough power.
• The power and sample size estimations are important components in a study.
• Nearly all studies require studying a subject with a particular characteristic rather
than the whole population.
• These subjects are then used to draw inferences about the whole population.
• These power estimations are used to determine number of subjects required
to answer the research problem (or null hypothesis).
• An example of power of the study is the case of thrombolysis in acute
myocardial infarction (AMI).
• For many years clinicians felt that this treatment would be of benefit given the
proposed etiology of AMI, however successive studies failed to prove the case.
• When adequately powered mega-trials were conducted, a small but
important benefit of thrombolysis was proved.
• Generally, a power of 0.80 (80 %) or higher is considered good for a studies.
• This means, there is an 80% chance of detecting a difference as statistically
significant, if a true difference exists. Power is computed differently for different
kinds of analysis.
• The higher is the power of a study, the more are the subjects and/or the larger is
the effect size (or the smaller is the p-value).
• With a very large number of subjects, a study may have good power to detect
even a very small effect size.
• To detect a small difference between groups, higher number of subjects is
required to contain adequate power.
• To detect a big difference between groups, a smaller sample size will be sufficient
to give adequate power.
• While determining the statistical power of study we need to consider;
1. The null hypothesis that, there is no difference between the
treatments in terms of mortality.
2. The alternative hypothesis that, there is a difference between the
treatments in terms of mortality.
 (a) Reasons to run a power analysis
• We can run a power analysis for many reasons that include:
1. To find the number of trials needed to get an effect of a certain size. This is
probably the most common use for power analysis as it tells us how many
trials we need to do to avoid incorrectly rejecting the null hypothesis.
2. To find the power, given an effect size and the number of trials available.
This is often useful when we have a limited budget, for say, 100 trials, and
we want to know if that number of trials is enough to detect an effect.
3. To validate research. Conducting power analysis is a good science.
• Calculating power is complex and is usually performed with a computer.
There is a list of links to online power calculators available at
https://statpages.info/#Power.
 Factors affecting a power of study:
• There are several factors that can affect the power of a study.
• These factors should be thoroughly considered prior to the development of a
study plan.
• We have control over some of the factors described below, while others we do
not
1. Sample size.
2. The precision and variance of measurements within any sample.
3. The true value of the parameter being tested.
4. Magnitude of a significant difference.
5. How certain we want o avoid type I error.
6. Statistical test applied.
7. Significance level.
 Ways to increase power in a study:
• To increase the power of study following attempts can be made:
1. Increase 𝜶: Set 𝛼 at 0.05 for most scientific studies.
2. Selection of test: Conduct a one-tailed test.
3. Effect size: Increase the effect size as big as possible.
4. Random error: Reduce random error by making it not to be random or use some
sort of repeated measures design. As there will be multiple measurements on a
subject, it would be possible to separate the error variance from the subject
variance.
5. Sample size: Increase sample size.
 Report Writing
• A research report is a formal account of how a scientific research project was
conducted and what its outcome is.
• Writing research reports can be one of the most difficult tasks for the
researchers.
• It involves long and stressed duration with difficult concepts, trying to produce
a cohesive description of how a research project was conducted and what is
meaningful outcome at the completion.
• The research reports should be well-organized, readable and presented in
formats consistent with generally accepted practice.
(a) Significance of Research Report:
• For the researchers, a research report provides a lasting record and reminder of
the work accomplished and its outcomes.
• The feedback other people give on the report can help extend the researchers'
knowledge and understanding of a topic.
• It can also help them to work on research design and management skills for
future projects.
• For readers/audience. a research report makes available the project information
to read about the project's aims, methods and findings, assess the quality of the
project, provide feedback to the project's researchers on what they like or dislike
about the project and incorporate aspects of the project's methods or findings
into their own work or thinking.
Potential readers of research reports:
• The potential readers of any research project are:
1. Other researchers, academicians or knowledge workers.
2. Managers in organizations who use research to inform decisions.
3. Front-line staff delivering programmes or services.
4. Clients or consumers of programmes or services.
5. People living in communities and neighborhoods where a research topic has
special relevance and research participants interested in the outcomes of
studies they have contributed to.
6. Journalists and other media representatives.
7. The general public.
• Using information provided in research reports, people can increase their
knowledge and understanding of a topic.
• This can lead to changes in people's thinking and behavior, or the design and
operation of human services and systems.
 Report Format:
• Research reports are net results of deliberate, painstaking and accurate
induction efforts.
• Report formats are likely to vary with the researcher or research firm conducting
the project, the client for whom the project is being conducted, and the nature
of the project itself.
• The usual steps involved in writing report includes logical analysis of the subject-
matter preparation of the final outline, preparation of the rough draft, rewriting
and polishing preparation of the final bibliography and writing the final draft.
• Though all these steps are self explanatory, in following section, a brief mention
of each one of these is given for better understanding.
• The described general format is intended as a guideline from which the
researchers can develop their own formats for the research projects at hand.
• Most research reports are divided into three parts and include following
elements:
(A) First Part (Formality Part):
1. Cover page
2. Submission letter
3. Title page
4. Certificate or statement
5. Acknowledgement6.
6. Table of contents 7.
7. Preface/forwarding/introduction
8. Abstract
(B) Main Report (Central Part of Report):
1. Introduction
2. Problem definition
3. Methodology and research design
4. Results findings
5. Limitations
6. Summary, conclusions and recommendations
(C) Appendix (Additional Details):
1. Letter of authorization
2. A copy of questionnaire
3. Copies of forms used
4. Sampling techniques
5. Fieldwork
6. Lists including contacts of individuals and organizations
7. Tables not included in findings
8. Statement of expenses
9. Bibliography-list of books, magazines, journals, and other reports
10. Any other relevant information
 Presentation of Data
• Research report writing and its presentation to audience is the final step in the any
research project.
• This process begins with interpretation of data analysis results and leads to
conclusions and recommendations.
• Finally, the formal report is written and an oral presentation made.
• After audience has read the report, the researcher should conduct a follow-up,
assisting audience and undertaking a thorough evaluation of the research project.
• An oral presentation provides a chance for researchers to present their research by
reading a paper and/or showing PowerPoint slides to a group of relevant audience
(viz faculty, students, judges, managers/other researchers/client firm/any related
authority).
• Presentations allow researchers to experience what it is like to present their research
at conferences. analysis s preliminary questions that the audience may have can be
addressed in the presentation.
• Presentation helps the audience to understand and accept the written report.
• Any preliminary questions that the audience may have can be addressed in the
presentations.
• Because many executives may form their first and last impressions about the
project based on the oral presentation, and thus its importance cannot be over
emphasized.
• The key to any effective presentation is its preparation.
• A detailed outline should be prepared following the format of the written research
report.
• The presentation must be focused to the audience.
• For this purpose, the researcher should determine the backgrounds, interests and
involvement of all those in the project, as well as the extent to which they are
likely to be affected by it.
• The presentation should be rehearsed several times before it is made to the audience.
• Visual aids including figures, tables and graphs should be displayed using variety of media.
• Flip charts of large pads of blank paper mounted on an easel ( support stand) enable the
researcher to manipulate numbers.
• They are particularly useful communicating answers to technical questions.
• Visual pages in advance, and the presenter flips through the pages during the
presentation.
• Visual supplements can be prepared on the pages in advance, and the presenter flips
through the pages during the presentation.
• Although not as flexible, over head projector and felt boards allow for instant
presentation of previously prepared information.
• Overhead projectors are suitable to present simple charts as well as complex overlays
produced by the successive additions of new images to the screen.
• Currently, use of computer packages such as Microsoft's PowerPoint is of immense help.
• They can be used for making computer-controlled research project presentations or for
presenting technical information such as analytical models.
 Research Protocol
• Once the research study is properly and completely planned, the plan should be
written down.
• Since protocol is the detailed plan of the study, it can be defined as,
"a document that describes the background, rationale, objectives, design,
methodology, statistical considerations, and organization of a research project’.
• The written protocol compels the investigators to simplify their thoughts and to
think about all aspects of the study.
• Protocols a necessary guide if researchers are working as a team on the project.
• It is essential, if the study involves research on human subjects or is on
experimental animals in order to get the institution's ethical approval.
• In addition, it is an essential component of a research proposal submitted for
funding.
• The first step before beginning and developing a detailed protocol is selection of
research topic.
• The topic should be able to stand alone as an explanation of the study doing so,
researchers need to understand prevailing operational realities and how to wot
within their limits.
• During development of the protocol, investigators can and should obtain advice of
colleagues and experts in refining their plans.
• Once a protocol for the research study has been developed and approved, and the
study has started and progressing, it should adhere strictly to the protocol and
should not change.
• Any violations (changes) of the protocol can discredit the whole study.
• If the changes made are minor, that part of the study should be excluded from the
analysis.
• After writing the protocol, particularly in large studies key is to develop the
operations manual for the study.
• This includes detailed instruction to the investigators in order to assure a uniform
and standardized approach for performing study with good quality control.
• A well-thought out and well-written protocol can be judged on the basis of
following criteria:
1. Adequacy to answer the research question/s and achieve the study
objective.
2. Feasibility in the particular set-up for the study.
3. Provision of enough details that can allow another investigator to do the
study and arrive at comparable conclusions.
• The protocol should outline the rationale for the study, its objective, the
methodology used and how the data will be managed and analyzed.
• In addition, it should indicate how ethical issues have been considered and
handled, how gender issues are being addressed.
 Research Protocol Format:
• The commonly followed research protocol is written according to the following
format.
1. Project title (g) Data management
2. Protocol summary (h) Statistical analysis
3. Introduction 6. Project management
(a) Study questions (a) Personnel
(b) Rationales and previous knowledge on (b) Action-plan
the subject 7. Strengths and limitations
4. Objectives 8. Ethical considerations
5. Design and methods 9. Expected outcome
(a) Study design 10. Budget summary
(b) Study population 11. References
(c) Sample size and sampling 12. Annexure
(d) Study subject: selection, definitions (a) Study formats
(e) Study questionnaires, formats (b) Budget details
(f) Data collection methods (c) Curriculum vitae of investigator/s.
 Cohorts Studies
• The word cohort means a group of people.
• Cohort studies are a type of medical research studies used to determine the causes
of disease and to establish relationships between risk factors and health outcomes.
• The cohort study design identifies a people exposed to a particular factor and a
comparison group that was not exposed to that factor and measures and compares
the incidence of disease in the two groups.
• Cohort studies provide the best information about the causation of disease, because
investigator follows persons from exposure to the occurrence of the disease.
• They may require long periods of follow-up since disease may occur a long time after
exposure. Therefore, it is a very expensive study design When risks are computed in
a study, the risk ratio is the measure that compares the Risk exposed to the Risk non-
exposed.
• The risk ratio is defined as, "the risk in the exposed cohort (the index group) divided
by the risk in the unexposed cohort (the reference group)".
• Cohort's studies can be forward-looking (prospective) or backward-looking
(retrospective).
Types of Cohort Studies:
• The simplest cohort design is prospective which involves following a
group forward in time, Fig. 3.11.
• The cohort study also can be retrospective.
• The descriptor, prospective or retrospective, indicates when the
cohort is identified relative to the initiation of the study.
(i) Prospective cohort studies:
• Prospective studies are planned in advance and carried out over a future
period of time.
• In these studies the investigators visualize and design the study, recruit
subjects, and collect baseline exposure data from all subjects, before any of
the subjects develops any of the outcomes of interest.
• The subjects are then observed into the future to record the development of
any of the outcomes of interest.
• The follow up is usually conducted by mail questionnaires, by phone
interviews, via the internet, or in person with interviews, physical
examinations, and laboratory or imaging tests or combinations of these
methods can also be used.
(ii) Retrospective cohort studies:
• In these studies, subjects are grouped on the basis of their exposure status and
are compared for their incidences of disease.
• Both, exposure status and outcome in these studies are determined
retrospectively.
• These studies examine data that already exist and try to identify risk factors for
particular conditions.
• Interpretations are limited only to current data because the researchers cannot
go back and obtain missing data.
Cohort Study Design:
• A cohort study starts with exposure (risk factor) and follows for the outcome
(disease).
• These studies consist of two groups namely; first group which is exposed to some
risk factor and second group that is free from exposure.
• Cohort studies typically observe large groups of individuals and record their exposure
to certain risk factors to find reasons about the possible causes of disease.
• These long-term studies are sometimes called longitudinal studies.
• The cohort study design is the best available scientific method to measure the effects
of a suspected risk factor.
• In cohort study, researchers raise a question and form a hypothesis about probable
cause of a disease.
• This is followed by observing cohort over a period of time which may take several
years.
• They collect data that may be relevant to the disease.
• Through this procedure, researchers aim to detect any changes in health linked to the
probable risk factors those have been identified.
• For example, investigators may ask participants to record specific lifestyle details over
the course of a study.
• On the basis of responses from cohorts, they analyze any possible correlations
between lifestyle factors and disease.
• Cohort studies are best to find and establish relationships between
health and environmental factors such as chemicals in the air, water
and food.
• These are issues that the World Health Organization ( WHO) helps
researchers to investigate with large-scale cohort studies.
• A cohort study is useful for estimating the risk of disease, the
incidence rate and/or relative risks.
• Advantages of Cohort Studies:
1. Cohort studies precisely indicate the chronological sequence between exposure
and outcome, because in a cohort study, subjects are known to be disease-free at
the initial stage of the observation where their exposure status is established.
2. Cohort studies allow to calculate the incidence of disease in exposure groups, so
absolute risk, relative risk, risk difference and attribution risk proportion can be
calculated.
3. Cohort design can be used to investigate common exposures. These are useful for
evaluating the effects of rare or unusual exposures. The investigators can identify
an adequate number of subjects who have an unusual exposure, for example,
exposure to toxic chemicals, adverse effects of drugs or treatments, unusual
occupational exposures etc.
4. Allow examination of multiple effects of a single exposure.
5. Cohort studies reduce the possibility that the results will be biased by selecting
subjects for the comparison group who may be more or less likely to have the
outcome of interest.
Disadvantages of prospective cohort studies:
1 Investigators have to follow large number of subjects for a long time.
2. They can be very expensive and time consuming.
3. They are not good for rare diseases.
4. They are not good for diseases with a long lag times.
5. Differential loss to follow up can introduce bias
Disadvantages of retrospective cohort studies:
They are not good for very rare diseases.
1. If records that were not part of design are used, the available data
may be of poor quality.
2. There is frequently an absence of data on critical factors if the data
was recorded the past.
3. It is difficult to identify appropriate exposed cohort and an
appropriate comparison group.
4. Differential losses to follow up can bias these studies.
• Limitations Cohort Studies:
➢Cohort studies are recognized as the most robust form of medical research
after experiments such as randomized controlled trials. Unfortunately, they
are not always the best form of observational work. Thus, cohort studies have
certain limitations:
1. As cohort studies follow exposure data and watch for any emerging cases of
disease they are less suited to finding clues about rare diseases.
2. They are typically unsuitable for identifying the causes of a sudden outbreak
of disease.
3. They are expensive to run and usually take many years, often decades, to
produce results.
4. They can only offer clues about the causes of disease, rather than definitive
proof of links between risk factors and health.
5. Participants may leave the cohort, possibly move away, out of contract, or
die from a cause that is not being studied; all these can bias the results.
 Observational Studies
• Research study designs are classified into two types namely;
- observational study
- experimental study
• These are type of studies in which individuals are observed or certain
outcomes are measured.
• No attempt is made to affect the outcome.
• This means, in these studies researchers observe the effect of a risk factor,
diagnostic test, treatment or other intervention without trying to change who
is or is not exposed to it.
• Observational studies do not control any variables; thus the results can only
establish relationships.
• Observational studies only allow us to claim association, and not the
causation.
(a) Cross-sectional Studies:
• This study involves collecting data about individuals at a certain point in time.
• Cross sectional studies are cheap and easy to conduct, but they do not give very
strong results.
• For example, a researcher concerned about the effect of working with asbestos
might compare the cancer rate of those who work with asbestos versus those who
do not.
• We never be sure that those working with asbestos and do not report cancer would
not eventually develop it.
• This type of study only gives a bit of the picture, so it is rarely used by itself.
Researchers tend to use a cross-sectional study to first determine if there might be
any link, and then later conduct another study to further investigate.
(b) Case-control Studies:
• The case control study is one where researchers identify people with an existing
health problem (cases) and a similar group without the problem (controls) and
then compare them with respect to an exposure or exposures.
• Researchers in this study attempt to select homogeneous groups, so that on
average, all other characteristics of the individuals will be similar, with only the
characteristic in question differing.
• For example, research on the link between smoking and lung cancer in the UK. In
the 1950's, almost 80% of adults in the UK were smokers, and the connection
between smoking and lung cancer was not well established. Researchers
interviewed about 700 lung cancer patients and tried to determine a possible
cause.
• This type of study is retrospective, because it asks the individuals to look back
and describe their habits (smoking).
• There are clear weaknesses in a study, because it expects individuals to not only
have an accurate memory, but also to respond honestly.
 Experimental Studies
• A scientific experiment is an organized and detailed series of steps used to
accept or reject a hypothesis.
• An experimental study is the preferred way of hypothesis testing and for that
relevant methods are subjected to continuing improvements.
• Experimental study is one where researchers introduces an intervention and
study the effects. The intervention may be drugs or devices; procedures; or
changes to participants behavior etc.
• Experimental study designs are further classified on the basis of randomization.
- Randomized study
- Non-randomized study
• Randomization is the heart of the experimental study; it means random
selection of eligible individuals from a population.
• Experimental study is commonly used in scientific research designs, and it is the
primary approach used to scrutinize the causal (cause/effect) relationships and
investigate the relationship between one variable and the other.
• This type of research is known as a hypothesis testing or a deductive research method.
This is because it is used to support or disprove the hypotheses which are formulated
by the researcher to address a specific research problem.
• There are key terminologies used in experimental studies that every researcher needs
to know.
Independent variable is a variable that can be manipulated, and thus is called as the
cause or treatment variable.
Dependent variable is the effect or outcome or response of independent variable
manipulation.
The point is that, the outcome must be measurable.
Experimental group is a group that receives the treatment.
Control group is the group that remains fixed, and at the end, it is compared to the
experimental group.
• Clinical experimental studies are of two different types namely;
Therapeutic studies :
- In therapeutic studies (clinical trials), different medicines or medical
procedures for a given disease are compared in a clinical setting.
Prevention studies:
- Preventive studies are trials that are conducted on healthy or apparently
healthy individuals with the aim of preventing future morbidity or
mortality.
- Preventive studies include community study, in which the intervention is
applied to groups, and field study, in which the intervention is applied to
healthy individuals at usual or high risk of developing a disease.
- Clinical and field studies are form of a randomized controlled trial (RCT), in
which the intervention is compared with a control and the allocation to
experimental or control group is randomized.
 (a) Randomized Study:
1) Randomized Controlled Study:
• This is the most reliable design because it has high level evidence to practice health.
profession.
• In this design, control group is present as a comparator to the interventional group.
• The interventional group receives test drug and control group receives either known
(standard) drug or placebo.
• The randomized controlled study is also known as clinical trial study or gold standard
study.
• The comparison between interventional new drug (IND) and known older
(standard) drug is made to identify better and safer drug for the patients to
treat.
• Placebo is used a control if no known (standard) drug is present to treat a
disease.
• Randomized study is further divided in to two types on the basis of treatment
received by two groups.
i) parallel study design.
ii) cross-over study design.
(i) Parallel study design:
• The parallel study design is a simplest and widely used study design.
• In this study design, interventional group receives IND and control
group receive old (known/standard) drug or placebo, throughout the
study.
ii) Cross-over study design:
• In cross-over study design both groups receives both (ie. new and standard) drugs,
Until the washout period interventional group receives IND and control group
receive standard drug or placebo.
• After the washout period, the treatments are exchanged as interventional group
receive standard drug or placebo and control group receive IND.
• This means that, interventional group becomes control group and control group
becomes interventional group.
• In blinding studies subjects are not informed about whether they are receiving a
test product or placebo. If subjects in the control group know that they are
receiving placebo then there might be possibility of reduction or elimination of
placebo effect.
• There are three types of blinding based upon the persons to be blinded in the
study.
- Single blinding: Only patient is blinded.
- Double blinding: Both, patient and clinicians are blinded.
- Triple blinding: In this, all, the patient, the clinician and the statistician is blinded.
• Without the blinding accurate results are not assured. Extra precautions may be
taken for clinicians, treatment administrators and any other related personnel
particularly on patients who receives interventional drug. Blinding process is
exposed after the completion of study. Sometimes unblinding may also works if
benefits or harm is explained in groups.
(2) Randomized Uncontrolled Study:
• In this study, there is only one group present with randomization and has
no control group.
• All the participants are given with a treatment that simply follows for a
period of specific time to examine if they improve, without any
comparison against the control group which is either under another
treatment or no treatment at all.
• This study is thus so weak when compared with the randomized control
study.
(b) Non-randomized Study:
(1) Non-randomized Controlled Study:
• This study is same as that of the randomized control study but has no
randomization.
• The control group receive standard old drug or placebo and interventional group
receive IND and finally outcomes are measured.
• Sometimes hospital records or data obtained from published literature can also be
used as a control group, such non-randomized or non-current studies named as
historical control studies.
(2) Non-randomized Uncontrolled Study:
• This study is also similar to that of randomized uncontrolled study but it has
no randomization and no control group.
• Outcomes are measured finally after the data collection.
(c) Different Randomized and Non-randomized Studies:
1. Cluster Randomized study
2. Latin square Design
3. Factorial randomized study
4. Multicenter trial study
5. Field trials
6. Therapeutic trials and prophylactic trials
7. Quasi experimental deign
Significance of Experimental Studies:
• Experimental research allows researchers to test their thoughts in a controlled
environment before making them commercially viable. It provides the best method
to test the theory behind the design.
• Experimental study has following advantages:
1. Using experimental studies, researchers can have better control over variables to
achieve desired results..
2. The subject or industry has no impact on the effectiveness of experimental study.
3. The experimental study outcomes are more specific.
4. The experimental study outcomes can be applied to obtain findings of similar other
thoughts.
5. It can help to identify the cause and effect of a hypothesis. Thus, researchers can
further analyze this relationship to determine much more in-depth thoughts.
6. Experimental study is an ideal starting point and the data collected can be used as
foundation on which researcher can build more thoughts and conduct more
research.
Designing Clinical Trial - Various Phases
• The development of investigational new drugs (INDs) involves performing clinical
trials to assess the safety and efficacy of these drugs in humans.
• Clinical trials are usually classified into 4 phases of development as Phase-I to Phase-
IV, with each potentially lasting for many months to several years.
• Upon successful completion of each phase of trial, necessary approvals are obtained
from the appropriate regulatory authority or authorities, for example the European
Medicines Agency in the European Union, Food and Drug Administration in the
United States of America, Health Canada in Canada, the Ministry of Health, Labour
and Welfare in Japan, Drugs Controller General of India in India etc. for the
progression to the next phase.
• Satisfactory completion and approval of Phase-I to Phase-III is
required for a IND to be approved for marketing.
• Phase-IV studies are conducted after a drug that has been approved
by regulatory agency, for the primary purpose of post marketing
surveillance.
• Recently, for speeding up the drug development process and to
quickly identify safety issues, Phase O studies (human micro sdosing
studies) have been introduced.
• Phase-0:
- The official name of a Phase-0 study is an 'Exploratory study’.
- Its is to quickly establish whether candidate drug will work as desired in
humans.
- This study is based on in vivo safety, pharmacology and toxicological
preclinical studies.
- At this phase studies, a single sub-therapeutic dose of the IND is administered
to a small number of healthy subjects (10 to 15), over a short duration of 7
days.
- The dose administered at this phase is too low to result in a therapeutic effect
to ensure the absence of any toxic effects, preliminary pharmacokinetic (PK)
and, pharmacodynamic (PD) data are collected evaluation to establish
suitability.
• Phase-I:
- The Phase-l studies are designed to assess the safety of an IND, to
understand its PK and PD properties, and to ideally identify a
potential therapeutic dose.
- These studies are usually conducted in a small number of healthy
volunteers/subjects (15 to 30).
• Phase-II:
- These studies are typically conducted to test the IND in a larger group
of patients who have the disease or illness for which the IND is being
developed, to determine whether it is efficacious, at least in the short
term.
- Phase-Il studies are larger than those conducted earlier in the drug
development, typically comprising up to 300 patients.
• Phase-III:
- The Phase-III studies are designed and performed to assess the
efficacy and effectiveness of an IND in a larger cohort of patients, all
of whom have the disease that the treatment is intended to treat.
- Such studies are typically conducted in several hundred patients, and
are usually conducted at multiple sites in multiple countries.
- Phase-III studies compares the new treatment against the current
'gold standard treatment for the condition for which the new
treatment is being developed.
• Phase-IV:
- Post marketing surveillance involves monitoring (Pharmacovigilance)
for safety concerns once a treatment has been approved by the
regulatory authority/authorities.
- Such surveillance is aimed to detect any uncommon adverse effects
that have not been observed previously or have only been observed
occasionally, and to observe the effects of long term administration in
a wider population.
HUMAN CLINICAL TRIAL PHASES
• Phase I studies :
- It assess the safety of a drug or device.
- This initial phase of testing, which can take several months to
complete, usually includes a small number of healthy volunteers (20
to 100), who are generally paid for participating in the study.
- The study is designed to determine the effects of the drug or device
on humans including how it is absorbed, metabolized, and excreted.
- This phase also investigates the side effects that occur as dosage
levels are increased.
- About 70% of experimental drugs pass this phase of testing.
• Phase II studies:
- It test the efficacy of a drug or device.
- This second phase of testing can last from several months to two years, and
involves up to several hundred patients.
- Most phase II studies are randomized trials where one group of patients
receives the experimental drug, while a second "control" group receives a
standard treatment or placebo.
- Often these studies are "blinded" which means that neither the patients
nor the researchers know who has received the experimental drug.
- This allows investigators to provide the pharmaceutical company and the
FDA with comparative information about the relative safety and
effectiveness of the new drug.
- About one-third of experimental drugs successfully complete both Phase I
and Phase II studies.
• Phase III studies:
- It involve randomized and blind testing in several hundred to several thousand
patients.
- This large-scale testing, which can last several years, provides the pharmaceutical
company and the FDA with a more thorough understanding of the effectiveness
of the drug or device, the benefits and the range of possible adverse reactions.
- 70% to 90% of drugs that enter Phase III studies successfully complete this phase
of testing.
- Once Phase III is complete, a pharmaceutical company can request FDA approval
for marketing the drug.
• Phase IV studies:
- It is often called Post Marketing Surveillance Trials, are conducted after a
drug or device has been approved for consumer sale.
- Pharmaceutical companies have several objectives at this stage:
1. To compare a drug with other drugs already in the market.
2. To monitor a drug's long-term effectiveness and impact on a patient's
quality of life.
3. To determine the cost-effectiveness of a drug therapy relative to other
traditional and new therapies. Phase IV studies can result in a drug or
device being taken off the market or restrictions of use could be placed
on the product depending on the findings in the study.