S5) Cohort Study Designs

Question 1

How does one conduct a Cohort Study?

Answer 1

• Recruit outcome-free individuals
• Classify into exposed and unexposed groups
• Follow-up each group over time:

I. Count the person-years ‘at risk’ (p-y)

II. Count how many develop outcome (d)

III. Calculate incidence rate (IR = d/p-y)

Question 2

How do we analyse Cohort studies?

Answer 2

• Calculate incident rates separately:

I. IR_Exposed = d_Exposed / p-y_Exposed

II. IR_Unexposed = d_Unexposed / p-y_Unexposed

• Compare exposure groups by calculating Incidence Rate Ratio (IRR) to determine relative risk:

IRR = IR_Exposed / IR_Unexposed

Question 3

How do we calculate 95% confidence intervals for Cohort studies?

Answer 3

95% Confidence interval = (IRR/error factor) to (IRR x error factor)

Question 4

Identify 3 advantages of cohort studies

Answer 4

• Study exposures and personal characteristics that are not routinely collected
• Obtain more detailed information on outcomes and exposures
• Collect additional data on potential confounding factors

Question 5

Identify 3 disavantages of cohort studies

Answer 5

• Time consuming
• Large and resource intensive (expensive)
• Not good for rare outcomes

Question 6

Describe how one conducts prospective cohort studies

Answer 6

• Recruit outcome-free individuals
• Classify them according to their exposure status
• Follow-up → count p-y and d
• Data collection may start immediately or later

Question 7

Describe how one conducts retrospective cohort studies

Answer 7

• Recruit outcome-free individuals
• Classify their initial exposure status and subsequent outcome status using historical records
• Follow-up → count p-y and d using historical records

Question 8

Compare and contrast cohort and case-control studies in terms of:

• Comparison of groups
• Time
• Expense
• Rare outcomes and exposures
• Range
• Bias

Answer 8

Question 9

Which 3 ratios are used to describe relative risk?

Answer 9

• Risk Ratio
• Rate ratio
• Odds ratio

Question 10

Define the following terms:

• Risk
• Rate

Answer 10

• Risk = proportion dead at end of time period
• Rate = incidence of death during time period