When the Steering Committee of the Physicians’ Health Study Research Group (1988) reported the results of the effect of aspirin on heart attacks, committee members also reported the results of the same aspirin consumption, for the same sample, on strokes. There were 80 strokes in the aspirin group and only 70 in the placebo group. The relative risk was 1.15, with a 95% confidence interval ranging from 0.84 to 1.58.
a. What value for relative risk would indicate that there is no relationship between taking aspirin and having a stroke? Is that value contained in the confidence interval?
b. Set up the appropriate null and alternative hypotheses for this part of the study. The original report gave a p-value of 0.41 for this test. What conclusion would be made on the basis of that value?
c. Compare your results from parts (a) and (b). Explain how they are related.
d. There was actually a higher percentage of strokes in the group that took aspirin than in the group that took a placebo. Why do you think this result did not get much press coverage, whereas the result indicating that aspirin reduces the risk of heart attacks did get substantial coverage?