What does relative risk (RR) measure in the context of a cohort study on diet and cardiac disease?

What Relative Risk Measures in a Cohort Study

Relative risk (RR) measures the ratio of the probability of an outcome occurring in the exposed group (bad diet) compared to the probability of the same outcome occurring in the unexposed group (good diet). 1

Definition and Calculation

• RR is calculated as the risk of disease in the exposed group divided by the risk of disease in the unexposed group 1
• In your cohort study example: RR = (Risk of cardiac disease in bad diet group) / (Risk of cardiac disease in good diet group) 2
• An RR of 1.0 indicates no difference in risk between groups 1
• An RR > 1.0 indicates increased risk in the exposed group (bad diet increases cardiac disease risk) 1
• An RR < 1.0 indicates decreased risk in the exposed group (bad diet decreases cardiac disease risk, which would be unexpected) 1

Interpretation Framework

The RR provides a measure of the strength of association between the exposure (diet) and outcome (cardiac disease): 2

• RR = 2.0 means the bad diet group has twice the risk of developing cardiac disease compared to the good diet group 1
• RR = 1.5 means 50% increased risk in the exposed group 1
• RR = 0.5 means 50% decreased risk in the exposed group 1

Critical Limitations to Understand

The RR alone does not tell you the absolute magnitude of risk—it only tells you the relative comparison between groups: 3

• A RR of 2.0 could mean going from 1% to 2% risk (1% absolute increase) or from 20% to 40% risk (20% absolute increase) 3
• Both relative and absolute risks must be reported together to understand clinical significance 3
• The RR can appear impressive while the absolute risk difference remains clinically trivial 3

Relationship to Baseline Risk

The RR shifts toward the null value (1.0) as the outcome becomes more common in the population: 4

• When cardiac disease prevalence is high (≥10%), the RR underestimates the true strength of association 4
• The RR is most reliable when the outcome is rare (<10% in the unexposed group) 2, 5
• This mathematical property exists because RR is a ratio of two probabilities, both influenced by baseline prevalence 4

Practical Application in Your Study

In your cohort comparing bad diet versus good diet for cardiac disease: 2

• Follow both groups forward in time and count cardiac disease events in each group 2
• Calculate: RR = (Events in bad diet group / Total in bad diet group) / (Events in good diet group / Total in good diet group) 2
• Always report the absolute risks alongside the RR (e.g., "15% vs 10%, RR = 1.5") 3
• Include 95% confidence intervals to assess precision of the estimate 2

Common Pitfall

Do not confuse RR with odds ratio (OR): 2, 5

• OR is appropriate for case-control studies 5
• When outcome prevalence is <10%, OR approximates RR 2, 5
• When outcome prevalence is ≥10%, OR exaggerates the true RR and should not be used as a substitute 2, 5
• In cohort studies, always calculate and report RR directly rather than OR 2