2. Objectives
To understand basic description of the study
design
To show real life example of a Cohort study
To highlight key design issues, biases and
how to control
6. • More than 5,000 people 30 and 62
years of age, free of coronary
heart disease were recruited.
• Examined every 2 years – 70 years
and still going on
• Now three generations are
enrolled-children, parents and
grandparents
http://www.framinghamheartstudy.org/
Started in Framingham,
Massachusetts, USA in 1948
9. Advantages and
disadvantages of
Cohort studies
Advantages
• High level of evidence about
establishment of disease
• Causal inferences can be established
• High external validity
Drawbacks
• Misclassification
• Expensive
• Time consuming
• Lost to follow up
11. Retrospective or historical cohorts
• Requires good record (such as birth record) of past exposure for a group of people who
can then be traced to determine their health status.
• Record linkages
• Birth records
• Hospital admissions
• Emergency presentations
• Midwives notifications
• Cancer registries
12. Comparing the strengths and weakness of different
study designs
Cross sectional Randomized
controlled Trials
Cohort
Investigation of rare disease or outcome
Investigation of rare exposure
Testing multiple effects of an exposure
Study of multiple exposures
Establishing temporality (exposure come before outcome)
Give a direct measure of incidence
Explore exposures which changes over time
Time required
cost
Ethical problems
13. Analysis of RCT &
Cohort and cross
sectional studies
• What type of analysis are
appropriate for these study
designs?
• Give formulas
Statistical test Cross-
sectional
RCT Cohort
Incidence
Prevalence
Relative risk
Attributable
risk
Prevalence
ratio
Odds ratio