Good for master program

1. STUDY DESIGNS
Richard Malumba

2. Lecture objectives
• How do we design and conduct studies to explain the etiology of and
risk factors for human disease?

3. Intro..
• Epidemiological study designs are broadly divided into observational
and Experimental studies
• Observational studies are further divided into;
• Descriptive Studies
• Case reports/studies/series
• Ecological studies
• Cross-sectional studies
• Analytical Studies
• Case-Control Studies (retrospective)
• Cohort Studies (prospective/longitudinal)

4. Cont..
• Experimental Studies
• Clinical Trials (RCTs)
• Community Trials

5. Case Reports and Case Series
• One of the most common and early origins of medical research
conducted through careful observations by physicians and other
health care providers of what they see during their clinical practice
• Such individual-level observations can be documented in a case
report, describing a particular clinical phenomenon in a single patient,
• Or in a case series that describes more than one patient with similar
problems

6. Cont..
• The main objective of case reports and case series is to provide a
comprehensive and detailed description of the case(s) under observation
• This allows other physicians to identify and potentially report similar cases
from their practice, especially when they share geographic or specific
clinical characteristics
• case reports and case series are merely descriptive in nature with no
reference group to make a strict comparison

7. Advantages and Disadvantages
• Case reports and case series are key hypothesis-generating tools,
especially when they are simple, inexpensive, and easy to conduct in
the course of busy clinical settings
• However, the lack of a comparison group is a major disadvantage
• Furthermore, external validity (generalizability) is limited, given the
biased selection of cases (all identified in clinical practice)

8. Cross-sectional study design
• Cross-sectional studies are mainly used for examining the
characteristics of populations so as to detect associations with events
of interest
• The relationship between exposure and disease is assessed
simultaneously e.g., we may want to associate high cholesterol levels
to increased risk of Congenital Heart Disease. In this case, we assess
for congenital heart disease and cholesterol

9. Cont..
• The investigator measures the outcome and the exposure(s) in the
population, and may study their association
• Participants in a cross-sectional study are just selected based on the
inclusion and exclusion criteria set for the study not basing on
exposure or outcome status
• These types of designs will give us information about the prevalence
of outcomes or exposures

10. Cont..
• The general design of a cross-sectional (or prevalence) study follows the
sequence that;
• We define a population and determine the presence or absence of exposure and the
presence or absence of disease for each individual at the same time
• both exposure and disease outcome are determined simultaneously for each study
participant
• views a snapshot of the population at a certain point in time in relation to exposure
and disease

11. Challenges in using cross-sectional studies
• Cross-sectional studies identify prevalent other than incident disease.
If there is an association therefore, it is more related to survival with
the disease
• In addition, it is not possible to establish a temporal relationship – we
cannot tell what preceded the other: the heart disease or the
cholesterol

12. Case-Control Studies
• Suppose you are a clinician and you have seen a few patients with a
certain disease
• You observe that many of them have been exposed to a particular
agent— biological or chemical
• You hypothesize that their exposure is related to their risk of
developing this disease
• How would you go about confirming or refuting your hypothesis?

13. Cont..
• Case Control studies are “observational” studies
• The investigator does not actively intervene in the exposure-outcome
process, but simply observes the process and then goes on to analyze
those observations so as to make rational inferences

14. Cont..
• In case control studies the relationship between cases of a particular
disease with a possible exposure is investigated
• In this study-design the investigator goes through the following steps:
Identifying persons with the disease of interest i.e. the cases: the starting
point is the people with the disease
Identify a comparable group of people without the disease who are the
“control” group
Establishing history of previous exposure to the suspected agent in both the
cases and the controls
Comparing exposure in the 2 groups

15. Cont..
• The investigation process in case-control studies is retrospective. This
is so because the process begins with identifying cases of a disease
and then proceeds with inquiries about past exposure to the
suspected causative agent
• In the analysis of case-control studies we compare past exposure to
the suspected agent between cases and controls
• We compare the proportion of cases that were exposed to the
proportion of controls that were exposed

16. Cont..
• Inquiries about past exposure are through interviews, questionnaires,
review of past medical and other records (e.g. employee records) and
results of chemical/biological assays e.g. of blood and urine
• Odds ratios are a measure of association used

17. Advantages
• They can be conducted quickly. This means that they can provide
quick answers to explain current problems
• They are usually cheap compared to other types of studies
• Since there is no follow up period, study subjects cannot be lost
• They are suited to investigating rare conditions

18. Disadvantages
• Does not establish sequence of events
• Potential for recall bias; either limitations in recall (forgetting), recall
bias (cases may be more keen to recall historical exposures more than
controls) or reporting bias (deception)

19. Cohort study design
• It is an observational and analytical study design that is very often used to
test certain hypotheses
• It is used to test the relationship between certain risk factors or exposures
and disease
• In Epidemiology, a cohort is a group of people who are defined by certain
attributes like age, residence, occupation, etc. whom are followed up in
time and in whom a record of events e.g. disease or death is noted

20. Why cohort study design?
• Usually done because epidemiologists want to test certain
hypotheses of “causal associations” between a certain exposure/risk
factor, and disease
• Such hypotheses are usually derived from other less rigorous studies
in analytical sense like case reports or case series, or cross sectional
studies that generate such hypotheses

21. Cont..
• The cohort study design starts with Selection of groups of individuals;
exposed and non-exposed individuals
• Follow up the groups in time to compare incidence of events in
exposed vs non-exposed (the design may include two or more groups
in each category

22. Features
• At the beginning of the study, the persons to be studied have not yet
developed the disease of interest
• The exposure is known and can be measured in one way or another
• Two groups of people from the same population are followed up; one
group has had or is under the exposure factor while the other group
(the control group) is free of exposure factor. Both groups are
followed up in time; both groups are initially disease free

23. Cont..
• During follow-up, incident cases of disease are noted and
comparisons made through calculations of risk of developing the
disease in both groups
• By the very nature of this design, any suspected causal factor
precedes the onset of disease and therefore this type of design is
quite useful in proving causal relationship between exposure
factor and disease

24. Advantages
• They are useful to investigate multiple effects of an exposure or
exposures, over time e.g. One can test for example effect of smoking
on the coronary vessels, as well as on the lungs over time, and can
test the effects of smoking, H.T, Obesity on coronary heart diseases
(CHD)
• Cohorts are very useful in measurements of time-relationships,
relating exposure and time of onset of disease
• With cohorts, it is easy to calculate the risk of developing disease
through the use of incidence rates

25. Disadvantages
• The nature of follow-ups makes the design quite expensive in a
number of things: money, time and personnel
• Loss to follow-up of subjects is a very serious drawback in this design.
This leads to bias if loss is not uniform in the two study group
• There is potential bias in assessing the outcome of exposure
especially if such person doing the assessment knows the hypothesis
under test and which subject was exposed and not exposed

26. Randomized Trials/RCTS
• Our objective, both in clinical practice and in public health, is to modify the
natural history of a disease so as to prevent or delay death or disability and
to improve the health of the patient or the population
• The challenge is to select the best available preventive or therapeutic
measures to achieve this goal
• Randomized trials is considered the ideal design for evaluating both the
efficacy and the side effects of new forms of intervention

27. Cont..
• Randomized trials are experimental groups, which we follow up
forward in time to detect incident events of interest and associate
them to the exposure (intervention/ treatment)
• incidence of events of interest(outcome) in the exposed group is
compared with that in the non-exposed
• We cannot intentionally expose a group of people to a putatively
harmful substance (e.g. a suspected carcinogen) simply because we
want to study an association
• The “exposure” in most randomized trials therefore is a “treatment”
or “preventive measure”

28. Why RCTs?
• Evaluate new approaches to treatment and prevention, thus
the concept of therapeutic and preventive trials.
• Evaluate tests of new health and medical care technology
• Assess new programs for screening and early detection of
disease
• Assess new ways of organizing and delivering health care
services

29. Forms of Randomized trials
• Clinical trial; a research activity or an experiment that involves the
administration of a test regimen to humans to evaluate its efficacy
and safety
• In this experiment the unit of randomization is the individual and
the site of the trial is usually a clinic or a hospital ward

30. Cont..
• Community trial; an experiment in which the unit of allocation is a
cluster of individuals or an entire community
• Clinical or community trials can be either therapeutic or preventive
depending on the intervention being evaluated

31. Randomization
• The process of assigning trial participants to treatment group using an
element of chance to determine the assignments in order to reduce
bias
• Increases likelihood that the groups will be comparable with regard to
baseline characteristics except the intervention under study

32. Blinding
• Concealing the allocation to the treatment groups is important so as
to reduce bias in assessing treatment outcomes
• Masking or blinding can achieve concealing the allocation and use a
code that is only broken after the analysis

33. Forms of blinding
• Open Label; Everyone i.e. the subjects, the investigators and the
assessors know who is in which treatment group
• Single Blind; Either the subject or the investigators are unaware of the
treatment group.
• Double Blind; Either the subjects and the investigators or the subjects
and the assessors are unaware of the treatment group
• Triple Blind; Everyone i.e. the subjects, the investigators and the
assessors are unaware of treatment group