C. Stratify by the potential confounder, and calculate stratum-specific OR's Stratified by age Ages 20-39 Ages 40-49 Ages 50-54 _II-I-I-I-I-I mi:- 6 64 9 1 1 5 11 Total 59 336 1 7 5 542 _-_-_-_ Odds ratio (age 20-39) = Odds ratio (age 40-49) = Odds ratio (age 50-54) = Summary (age-adjusted) odds ratio* = 1.57 * The summary OR was calculated using a statistical procedure known as the Mantel-Haenszel weighted odds ratio. You will learn about this measure in future epidemiology and biostatistics classes. d. Compare the crude odds ratio with the adjusted odds ratio. Do you think that age was a confounder in the association between OC use and stroke? Why or why not? e. As the investigator, which measure of effect (crude or adjusted) would you choose to report to your audience? 6. Stratification: Example 2 The Boston Area Health Case-Control Study looked at the association between inactivity and myocardial infarction (MI). Information on smoking was collected. It has been shown in previous studies that smoking increases the risk for MI and that people who smoke are more likely to be inactive. Furthermore, it is hypothesized that smoking is not on the causal pathway between inactivity and MI. Based on this information, the investigators considered smoking status to be a potential confounder of the inactivity-MI relationship. Therefore, they stratified on smoking status and obtained the following fahlp