Van Hylckama Vlieg et al. (2009) investigated the relationship between oral contraceptive use and thrombosis in women. In a sample of 1524 adult female patients who had thrombosis, 103 had taken oral contraceptives regularly. In a second sample of 1760 women from the same population who did not have thrombosis, 658 had taken oral contraceptives regularly.

a. What type of study design was used?

b. Graph the data. Which treatment condition (oral contraceptive use) had the higher proportion of women with thrombosis?

c. Test for an association between oral contraceptive use and thrombosis.

d. What is the odds ratio of thrombosis in women taking oral contraceptives compared to women not taking oral contraceptive? Include a confidence interval for the population odds ratio.

e. Under what circumstances can we say that the odds ratio estimated in part (d) is a reasonable estimate of the relative risk of thrombosis?