You are on page 1of 17

CASE CONTROL STUDY DESIGN

Dr Mohammad Nasim Saba MD Kabul ,PGD Shanghi


Learning objectives
After this session, you will be familiar with:
• The principal types of epidemiological studies
• The basic design features of a case-
control study
• Rationale for applying case-control
designs (advantages)
• Limitations of case-control studies
(disadvantages)
• Calculating measures of association in CC
study and interpreting results
• Example applications case-control designs
Study designs
Strong
est Meta-analysis Experimental
Analytical
Randomized controlled trial
Cohort study Analytical
Case-control study
Observational
Cross-sectional study
Ecological (aggregate) study
Case reports/Case series Descriptive

Weakest
flow-chart of
study designs

Grimes & Schulz


(2002). Lancet,
359: 145-49
In practice
• Studies usually begin simply, as descriptive
studies (case reports, case series, cross-
sectional studies)
• Progress to more complex analytic studies
(cohort, case-control studies)
• And then to experimental studies (RCTs ,
where appropriate)
Design of Case-Control Studies

The investigator selects


cases with the disease,
and appropriate
controls without the disease
 and obtains data regarding past exposure to
possible etiologic factors in both groups.
 The investigator then compares the frequency
of exposure of the two groups.
Case-control studies
Exposed

Cases
(+ disease)
Not exposed
Defined
population

Exposed

Controls
(- disease)
Not exposed

TIME

DIRECTION OF ENQUIRY (about past exposures)


Case-control design
• Advantages Disadvantages
– Relatively efficient (in – Selection of subjects for case
terms of time, money, and control separately
effort) • Prone to selection bias in
control choice
– Suited to diseases with long
latent period between – Measurement of exposure
exposure and disease (e.g retrospectively
smoking and lung cancer) • Exposure is assessed after the
development of the outcome –
– Optimal for rare disease recall bias
(e.g. Circumcision on penial • Measurement errors interviewer
cancer) bias
– Can examine multiple – Can usually only study one
exposures disease or outcome
– Useful for generating – Cannot calculate incidence
hypothesis rates or relative risk
Managing bias
Controls should represent the population
from which cases have come.
• Four strategies for selection of controls:
– Selection of controls from the same hospital/
clinic as the cases
– Matching (by age and sex)
– Using a population based sample
– Using two or more control groups
Continue..
• Recalling past exposure (cases may
remember exposure more than the
controls or vise versa)
• Interviewer bias
Strategies to control bias:
– Use data recorded before the outcome has
occurred
– Use blinding of investigator and subject
Measuring association
• In Case control study we cannot calculate
incidence rate or relative risk
• Can calculate estimate of odds ratio in
rare diseases
Calculating odds ratio
Outcome Total

Yes No

Exposure Yes A b a+b

No C d c+d

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

A two by two table


Odds= a/b
Odds = c/d
Odds ratio= a/b/c/d = a/b*d/c =ad/bc
Example
• The research question is “whether there is
association between intramuscular injection
of Vit K and risk of childhood leukemia”
• The investigator selects:
– 117 cases of leukemia
– 117 children without leukemia as controls
and asked past exposure to Vit K injection
– 69 of 117 cases had received Vit K injection
– 63 of 117 control had received Vit K injection
Example
• Calculate and interpret measures of
association for the data from this study
Leukemia Total

Yes No

Vit K injection Yes 69 63 132

No 48 54 102

117 117 234


Summary
• Experimental and observational designs
• Analytic and descriptive designs
• Case control design
• Advantages and disadvantages of CC
design
• How to control biasing in CC study
• Calculating measures of association in CC
study
References
1. Stephan B. et. al. Designing clinical research: second edition (2001)
by Lippencott Williams & Wilkins

2. Purdie, D., Green, A., Bain, C., Siskind, V., Ward, B., Hacker, N.,
Quinn, M., Wright, G., Russell, P. and Susil, B. (1995) ‘Reproductive
and other factors and risk of epithelial ovarian cancer: an Australian
case-control study. Survey of women’s Health Study Group’.
International journal of cancer 62(6): 678-684

3. Golding J, Greenwood R, Birminghan K, Mott M. Childhood cander


intramuscular Vit K and pethedin given during labour. BMJ 1992; 305:
341-6

4. Mellin, G.W. and Katzenstein, M. (1962) ‘the sega of thalidomide:


neuropathy to embryopathy, with case reports of congenital anomalis.’
new England journal of medicine 267(23): 1184-931

You might also like