This chapter presents a two-part study. The first part examines the regression to the mean incurred in measuring and remeasuring systolic blood pressure on the same individual. The second part evaluates an experimental diet thought to decrease the severity of benign breast disease when the results are biased by misclassification of individuals assigned to treatment and control groups. The analysis of the repeated blood pressure measurements shows substantial regression to the mean. Necessarily, an intraclass correlation coefficient indicates the same magnitude of regression to the mean. Specifically, the estimated slope is 0.769 from the regression analysis of the first and second blood pressure measurements, and the estimated intraclass correlation coefficient is 0.816. The clinical trial to evaluate the influence of a methylxanthine-free diet on the signs of benign breast disease produces no strong evidence that the diet is effective. The analysis comparing the intend-to-treat individuals yields an odd ratio of 1.420. This odds ratio, however, is reduced by considerable misclassification bias.
Oxford Scholarship Online requires a subscription or purchase to access the full text of books within the service. Public users can however freely search the site and view the abstracts and keywords for each book and chapter.
If you think you should have access to this title, please contact your librarian.