Baldness and myocardial infarction, lurking variables. This exercise is a continuation of Exercise 18.17. The investigators found differences in cases and controls besides those have to due to baldness. For example, the median age of cases was 47 years, while the median age of controls was 43 years.

(a) Explain why careful consideration of age differences is important when studying the relationship between baldness and heart disease.

(b) The investigators used a logistic regression model to adjust odds ratios for the following factors: age, race, religion, years of education, body mass index, use of alcohol and cigarettes, family history of myocardial infarction, history of angina, hypertension, diabetes, hypercholesterolemia, gout, exercise, personality, and number of doctor visits in the prior year. The unadjusted and multivariate adjusted odds ratios reported by the article are listed in Table 18.22. Do the multivariate adjustments materially affect your interpretation?
TABLE