Suppose we wish to perform another study of the association between OC use and MI and will control for age. Set up a logistic regression based on the data in Table 13.7, and use the results to estimate the sample size needed for the new study. Assume that the underlying age adjusted OR = 2.0 and we will conduct a two-sided test with α = 0.05 and power = 0.80?

Cardiovascular Disease

Refer to Table 13.7.

Table 13.7 Association between MI and OC use by age

 

Transcribed Image Text:

Cases (MI) Recent OC use Proportion MI Age Controls OR Proportion OC user Yes 62 7.2 25-29 23 No 2 224 Yes 30-34 33 8.9 No 12 390 Yes 35-39 26 1.5 No 33 330 Yes 40-44 6. 3.7 3 16 No 65 362 Yes 45-49 6 3.9 3 24 No 93 301 Total Yes 29 135 1.7 No 205 1607