Typically, different measures of relative association are calculated for cohort studies versus case-control studies. Name the two measures of relative association typically calculated for cohort studies and they typical measure of relative association used in case-control studies. Explain what is being compared in each of the three measures of association. Include information about the criterion by which statistical significance is determined in each measure of relative association.

Question

What will be an ideal response?

Answer 1

Ans: Students should indicate that for cohort studies, the two types of relative association (or relative risk) are risk ratios (or cumulative incidence proportion) and rate ratios. Case-control studies typically use the odds ratio as the measure of relative association. Both the risk ratio and the rate ratio are measures of the comparison of disease among those with exposure of interest relative to disease among those without exposure. A risk ratio compares disease incidences, is appropriate when follow-up time is similar among those exposed and those unexposed, and usually uses count data. A rate ratio compares disease incidence rates, is preferred when follow-up time varies considerably between the exposed and unexposed, and uses person-time data. An odds ratio is a measure that reflects a comparison of the odds of exposure among cases compared with odds of exposure among controls. Each of the measures of association is deemed statistically significant if its 95% confidence interval does not include the value 1.0.