study designs

By Samrawit.F (BSc, MSc)

 Classifications of study designs

1. Descriptive
• Defines the amount and distribution of health problems in
relation to person, place and time. It answers the questions
who, where and when.
2. Analytic
• involves explicit comparison of groups of individuals to
identify cause and determinants of health and diseases. It
answers the questions why and how
 Descriptive study designs
Features
• Mainly concerned with the distribution of disease by place
person and time
• Provide information for health managers
• Used to generate hypothesis
• Quick and easy
• Less expensive
 Cont…

A. case report and case series

Case Report is a careful, detailed description ,(clinical
report ) or medical profile of a single patient by
clinician/s
 Most basic type of descriptive studies
 documents unusual medical occurrence
 Cont…

Case series is a series of patients with a specific

disease or characteristic of interest by one or more
clinicians.
• Used as an early means to identify the beginning or
presence of an epidemic, generate hypothesis and
inform natural history
• Can suggest the emergence of a new disease
 cont...

Strength:

• very useful for hypothesis generation

• useful for studying signs and symptoms (new

syndromes) and creating case definitions for
epidemiological studies.
• Use available clinical data
 cont...

Limitations:
• Lack of an appropriate comparison group.
• Detailed and complete risk factor information is
difficult to obtain for all cases from records.
• Can’t be used to test for presence of a valid statistical
association
 Cont…

• Are prone to atomistic fallacy (the opposite of

ecological fallacy). Example: at an individual level a
high income may be associated with lower rate of
suicide but this does not mean that societies which
are rich have a lower rate of suicide.
 b. Cross Sectional Studies

• Cross sectional studies or prevalence survey: measure

disease and exposure status simultaneously among
individuals in a well-defined population in a narrowly
defined time period.
 Disease is measured at one point in time
 Exposure may be measured at the same time and/or
historical exposure information may be available
 Characteristics of cross-sectional studies
• Assess both exposure and outcome simultaneously
• Are based on point prevalence rates
• Compare prevalence of disease in persons with and
without the exposure of interest
• Are cheaper, easier & useful for studies of exposures that
are unalterable over time or E.g.-blood group, ethnicity,
sex, etc.
 Cont…
• Can have descriptive or analytic purposes
• The descriptive type is carried out to study prevalence of
health related events at a point in time/snapshot
• The analytic type is carried out to assess association
between exposure and outcome
– Exposure and disease status are assessed simultaneously
among individuals at the same point in time
– Measures of association is made using odds ratio
 cont…
Advantages
• Less expensive: a one-step, one-time collection of data

• It starts with reference population, thus making possible

generalization
• Provide much information useful for planning health services &
medical programs
• No loss to follow up

• multiple factors and outcomes at same point time

• generate hypotheses
 cont…

Disadvantages

• It is difficult to know which occurred first, the exposure or the

outcome, this is known as 'chicken or egg dilemma'.

• Example: study conducted to explore relationship b/n

physical activity & CHD

• Result: prevalence rates among those who were not physically

active in their work was nearly 5 times higher than that among

those who engaged in farm labor.

 c. Correlational/ Ecological study design
 An ecological study is a study in which one or more
exposures (or confounders) is measured at the
population level rather than the individual level
 Uses data from entire population to compare disease

• frequencies
• - between different groups during the same
period of time, or
• - in the same population at different points in time
 cont…

• Does not provide individual data, rather presents

average exposure level in the community
Example:
• Comparing incidence of dental cares in relation to
fluoride content of the water among towns in the rift
valley.
• Mortality from CHD in relation to per capital cigarette
sales among the regions of Ethiopia.
 cont…

Strength
 can be done quickly
 Inexpensively
 can be done using available data (routine records and
reports-- for example, death rates, per capita income,
national food consumption…)
 cont…
Limitation
• Data on exposure and outcome are not linked at
the individual level; association found with
aggregate data (average values) may not apply to
individuals (Prone to ecological fallacy)
• Lack of ability to control for effects of potential
confounding factors.
 Analytical study design

• Types of analytical study design

 Observational studies : Information are obtained by
observation of events.
Case control
Cohort
 Interventional / Experimental : The researcher does
something about the disease or exposure
Clinical trial
Field trial & Community trial
 a. Case control study design

• Starts with cases and comparative groups(controls)

• We determine what proportion of cases were exposed
and what proportion were not
• We also determine what proportion of controls were
exposed and what proportion were not
 cont…

 Subjects are selected with respect to presence or

absence of disease, or outcome of interest
 Unlike cross-sectional studies, may involve incident
cases of disease.
 The presence of the factor or the characteristic of
interest in the two groups is usually determined by
interview and/or a review of records.
 Cont….

Case control study

exposed
cases
Not exposed Source pop

exposed
controls
Not exposed
Study begins here
 Cont….

• At baseline:

– Selection of cases (diseased) and controls (not

diseased)

– Exposure status is assessed in both groups

• Retrospective design – lacks temporality

 cont…

• Generates many insights in to the aetiology of chronic

diseases such as lung, cervical, cancers, congenital
defects, etc.
– Important for study of rare disease
– For diseases with latent period or duration
– Use convenient sampling
– Short study period
– Less expensive than cohort studies
 Cont…
Selections of controls
• Controls should have the same exposure distribution as the
source population from which cases are drawn
• The controls should be similar to the cases in all respects
other than having the disease in question or
• Should be representative of all persons without the disease
in the population from which the cases are selected
 Sources of cases and controls
CASES CONTROLS
All cases diagnosed in the Sample of general population
population
All cases diagnosed in a Non-cases in a sample of the
sample of the population population

All cases diagnosed in all Sample of patients in all

hospitals hospitals who do not have the
disease
All cases diagnosed in a Sample of patients in the same
single hospital hospital who do not have the
disease
Any of the above methods Spouses, siblings or associates
of cases
 Analysis of case control results

• The analysis of case control study is basically a

comparison b/n cases and controls with respect to the
frequency of exposure
• Odds ratio: the ratio of the odds that the cases were
exposed to the odds that the controls were exposed
 case control

exposed a b a+b

Non exposed c d c+d

a+c b+d a+b+c+d

odds of exposure(cases)= a/c

odds of exposure(controls)=b/d
odds ratio= exposure among cases/exposure among controls=
ad/bc
 Cont…
Example
A case-control study was conducted to evaluate the
relationship between artificial sweeteners and
bladder cancer. 3,000 cases and 3,000 controls
were enrolled in the study. Amongst the cases,
1,293 had used artificial sweeteners in the past,
while 1,707 had never used artificial sweeteners.
Among the controls, 855 had used sweeteners and
2,145 had not. Compute the appropriate measure
of association
 Cont…
• A study was conducted in men aged 40-70 in order
to determine whether exercising for 2 or more hours
per week decreases the likelihood of heart attack.
The cases were 1,000 men who had recently had a
heart attack; of these, 236 reported that they had
regularly exercised for two or more hours per week
prior to their heart attack. 1,000 controls were also
selected for the study; of these, 379 reported that
they exercised regularly. Calculate the magnitude of
association between regular exercise and heart
attack.
 Strengths and Weaknesses
Strength Weakness
• Can use prevalent cases
• Is inexpensive and quick to Uncertainty of exposure-
complete disease time relationship
• Well suited for the  Inability to provide a direct
evaluation of diseases with estimate of risk
long-latency diseases  Very prone to biases (recall
• Can study rare diseases & selection bias)
• Can examine multiple  Misclassification of
exposures exposures and outcomes
• Usually requires a smaller  Selection of controls is
study population than a difficult
cohort study  Inefficient for rare
exposures
 b. Cohort study design
• Cohort
– a group of people who share a common experience
or condition
– E.g. Birth cohorts, cohort of smokers, occupational
exposures
• Cohort studies
– The observation of a cohort over time to measure
outcome(s)
– Because the data on exposure and disease refer to
different points in time, cohort studies are
longitudinal
 Cohort study design …

• Cohorts are selected based on their exposure status

• Eligible participants are then followed over a period
of time to assess the occurrence of that outcome
• At the time the exposure status is defined, all
potential subjects must be free from the disease under
investigation
 Cont…

Types of cohort studies

• Prospective Vs. Retrospective

• Closed Vs. Open

• Incidence cohort Vs. Prognostic (clinical)

 Cont…

Prospective Vs Retrospective
 Classification depends on the temporal relationship
between the initiation of the study and the occurrence
of the disease
 Both designs classify subjects in the study on the
basis of presence or absence of exposure
 Cont…

Prospective cohort
• The study participants must be followed into the
future to assess incidence rates of the disease
• More reliable and informative compared to the
retrospective cohort
• Unless specified, cohort study refers to prospective
cohort
 Prospective cohort study
Disease
Exposure Study starts occurrence

time

Disease
Study starts Exposure occurrence

time
 Cont…

Retrospective cohort study

• In this study design, exposure is ascertained from past
records and outcome is ascertained at the time the
study is begun

Disease
Exposure occurrence Study starts

time
 Cont…

Open Vs. closed cohort

1. Closed- members followed without adding others
(100 children=> 2, 3 years
2. Open or dynamic cohort study:
- Others could be added or lost
- People at risk are measured using person- time
 Cont…
Incidence cohort Vs. Prognostic (clinical)
– Incidence Cohort Study
• To assess incidence of disease
• To identify risk factors for disease onset
• Incidence greater in exposed than non-exposed?
– Prognostic Cohort Study
• Follow diseased cohort to assess factors
associated with outcome (recovery or death)
• Goal is to identify explanatory/prognostic
factors of those factors helped to the dev’t of the
disease.
 Analysis of cohort study

• The basic analysis in cohort studies are:

• calculation and comparison of rates of the incidence
of the outcome for exposed and non-exposed.
• comparison of the two groups with baseline
characteristic to ensure similarity.
 Cont...
• Measure of association in cohort studies

Relative risk (RR) or Risk Ratio

RR = Risk in exposed
Risk in unexposed

• RR=1 – no association; RR<1 – negative

association; RR>1 – positive association (Weak
– RR close to zero; Strong – RR far from zero)
 diseased Non diseased

exposed a b a+b

unexposed c d c+d

a+c b+d a+b+c+d

risk (exposed)= a/a+b

risk (unexposed)= c/c+d
risk ratio= [a/a+b] / [c/c+d]
 Cont…
Example
Calculation of the rate ratio from a hypothetical cohort
study of smoking and cancer of the pancreas followed
for 1 year.

cancer No cancer

smoker 42 27,000

Non smoker 7 63,000

 Summary of Measures of association: cohort
studies
Attributable Risk Percent (AR%) among exposed:
AR% = Risk in the exposed - Risk in unexposed
Risk in exposed

= AR -1 X 100
Ie
Population attributable risk ( PAR) = Risk total population - Risk in
unexposed
= AR X proportion of exposed individuals in population.
Population Attribuable Risk Percent (PAR%) :
PAR% = Risk in population - Risk in unexposed X 100
Risk in population
= PAR X 100
Incidence rate of disease in population
 Cont…

Strength of cohort study

• valuable when the exposure is rare
• can examine multiple effects of a single exposure
• can elucidate temporal relationship
• allows direct measurement of risk
• minimize bias in ascertainment of exposure
 Cont…

Limitation
• inefficient in evaluation of rare diseases
• expensive
• time consuming
• loss to follow-up create problem
 c. Experimental/ Interventional study
design

• Always prospective
• Usually begins with disease but sometimes with
cause
• Comparison groups (sometimes)
• The researcher intervenes with natural course of
action
 Cont…

Classification of interventional studies

 Based on population:
- Clinical trial
- Field trial
- Community trial
 cont…

• Clinical trial - usually performed in clinical setting

and the subjects are patients.
• Field trial- used in testing medicine for preventive
purpose and the subjects are healthy people.
• Community trial - a field trial in which the unit of
the study is group of people/community.
 Cont…
Based on the objective:
1. Therapeutic (secondary prevention) trials
– Are designed to test the effect of therapies
– conducted among patients with a particular disease to determine
the ability of an agent or procedure to diminish symptoms,
prevent recurrence, or decrease risk of death from that disease
2. Preventive (primary prevention) trial
– Designed to evaluate prevention strategies
– involves the evaluation of whether an agent or procedure
reduces the risk of developing disease among those free from
that condition at enrollment
 Cont….
Based on design:
• Uncontrolled trial: no control group and control will
be past experience (history)
• Non-randomized controlled trial: There is control
group but allocation into either group is not random
• Randomized controlled trial: There is control group
and allocation into either group is randomized
– Individual randomized controlled trial:
Randomization is at individual level
– Cluster randomized controlled trial: The
randomization is at cluster level
 Cont…
Based on level (phase I-IV) for new drug trials:
Phase I (clinical pharmacologic studies):
• Trial on few subjects (20-80)
• 1st experiment in human for new drug, schedule, or combination
• Primary concern: Safety
• Goal: define the maximum tolerated dose (MTD) in a dose-
escalation study
Phase II (efficacy studies):
• Small randomized, controlled, blinded (100-300)
• Tests tolerability, different doses, efficacy and to further evaluate
its safety
• E.g., optimal dosage without side effects
• Goal - Identify suitable formulation of drug
 Cont..
Phase III (effectiveness studies):
• Referred to as clinical trial
• Trial on large number of subjects (1000+) with therapeutic
dose to evaluate effectiveness of drug and side effects
• Usually randomized, blinded, controlled trial
• If successful, licensed and marketed
Phase IV
• Post marketing surveillance for long term effects.
• Large studies after approval of drug
• Often observational, study long-term effects
• Long term efficacy, rate of serious side effects
• Evaluate drug in “real life”, additional uses
 Cont…
Uncontrolled trials
• The new intervention is studied without any direct

comparison with a similar group of patients on more

standard therapy
• Problems of uncontrolled studies are:

-biased selection of participants

-biased assessment of outcomes
 Cont…
Non randomized concurrent controls
• Non randomized methods are used to allocate participants to
intervention or control group
• Systematic allocation - e.g. date of birth, date of
presentation
• Judgement allocation – investigator or participant is allowed
to exercise judgement in allocation
• Major problem is predictability of allocation group,
especially by investigator
 Cont…
Randomized concurrent controls
• An experimental design used to evaluate drugs and
health interventions
• Subjects should be allocated to treatment group
according to some chance mechanism
• RCT is a research activity that involves the
administration of a test regimen to humans to
evaluate its safety, efficacy, effectiveness
 Main steps in RCT
1. Identify new drug/intervention/prevention
2. Identify comparison –
Example, standard treatment versus placebo (Control v
intervention group)
3. Define eligible patient population/ exclusions (i.e the
sampling frame)
4. Define the outcomes and how to assess them
5. Write the protocol
 cont…

6. Obtain research ethics committee approval

7. Recruit and consent required sample of patients
8. Randomize to treatment, then treat
9. Follow-up and compare/analyze outcome data
10. Publish/disseminate findings
 Problems related to intervention trials
A. Ethical considerations
• Practices or substances known to be harmful should
not be used in this study.
• Therapies known to be beneficial should not be
withheld from any affected individuals.
• Investigators have to have a complete knowledge of
the subject.
 Cont…

• Must have informed consent from each

• Subjects should be left free to withdraw
• A written research protocol is a must.
 cont…

B. Feasibility/practical issues- volunteers, control

C. Cost of experimental studies
The Quality of “Gold Standard“ study design

 The qualities that make experimental studies gold

standard are:
 Randomization
 Blinding
 Use of Placebo
 Randomization

• Maximize the chance of comparability of study

groups at baseline
• Ideally, both groups are at the same stage of the
natural history of disease
• Ideally, only one factor affecting the outcome of
interest would vary between the study groups
 Blinding…
• Single- blind is where only the study subjects do not
know who is getting the active intervention
• Double -blind is where study subjects and health care
giver do not know who is getting the active
intervention
• Triple-blind is where the study subject, the field
investigator and the health care provider do not know
who is receiving the active treatment.
 Cont…
• Blinding procedures:
 Minimize bias related to assignment, assessment, or
compliance
 Minimize the potential for observation bias in
reporting of side effects and in assessing outcome
 Minimize differential compliance
 Minimize differential loss to follow-up
 can be single, double or triple blind
 placebo

• Placebos are inert treatments intended to have no

effect other than the psychological benefit of offering
treatment.
• Can only be used if there is no accepted treatment for
the condition under study.
• Use of placebo minimizes the bias in the
ascertainment of both subjective disease outcomes
and side effects.
 Cont…

 It facilitates that both groups in the study gain equal

attention.
 Placebo effect: tendency for individuals to report
favorable response to any therapy regardless of the
physiologic efficacy of what they received.
 Analysis in Interventional studies
• Similar to cohort studies
• Compare the rates of the outcome of interest in the
treated group(s) and the corresponding rates in the
comparison group(s)
• The roles of chance, bias, and confounding must be
evaluated
• compare the relevant characteristics of the
randomized treatment and comparison groups to
assess balance is achieved
• This comparison should always be presented as one
of the first tables in the report of the study findings
 cont…
• The exclusion of any randomized patients from the
analysis can lead to biased results
• Noncompliance may be related to factors that also
affect the risk of the outcome under study
• Thus, analyze by intention to treat-in other words,
“once randomized, always analyzed”
• Those who are no longer complying with the study
regimen should continue to provide all follow-up
information whenever possible, or at the very least,
their vital status should be ascertained
 Bias
Definition
 Any systematic error in an epidemiologic study that
results in an incorrect estimate of the association
between exposure and risk of disease.
 May result from systematic error (or difference
between exposed and unexposed populations or
between cases and controls) in the collection,
recording, analysis, or interpretation of data.
 Is an error that affects one group more than another
 Types of bias

1. Selection bias - is any error that arises in the

process of identifying the study populations.
A. Non-response bias
B. Loss to follow-up
C. Volunteer/Compliance bias
D. Cohort bias
 cont…
2. Observation or information bias – includes any
systematic error in the measurement of information
on exposure or outcome.
A. Interviewer bias
B. Recall bias
C. Social desirability bias
D. Placebo effect
E. Healthy worker bias
F. Lead-time bias
 Recommendations to minimize bias

 Choose study design carefully

 Choose objective rather than subjective outcomes
 Blinding
 Use well-defined criteria for identifying a case
 Confounding

 Confounding is the mixing of the effect of an

extraneous variable with the effects of the exposure
and disease of interest.
 Confounding is distortion of the estimated effect of
an exposure on an outcome, caused by the presence
of an extraneous factor associated both with the
exposure and the outcome.
 Control for Confounding Variables
 Randomization: assures equal distribution of
confounders between study and control groups
 Restriction: subjects are restricted by the levels of a
known confounder
 Matching: potential confounding factors are kept
equal between the study groups
 Stratification: for various levels of potential
confounders
 Multivariable analysis
Thank you