Statistical inference for noninferiority of difference in proportions of clustered matched-pair data from multiple raters. Issue 1 (2nd January 2017)

• Record Type: Journal Article
• Authors: Saeki, Hiroyuki; Tango, Toshiro; Wang, Jinfang
• Is Part Of: Journal of biopharmaceutical statistics. Volume 27:Issue 1(2017)
• Page Start: 70
• Page End: 83
• Publication Date: 2017-01-02
• Subjects: Clustered matched-pair data -- correlated proportions -- multiple raters -- noninferiority; Pharmacy -- Statistical methods -- Periodicals; Drugs -- Testing -- Statistical methods -- Periodicals; Biometry -- Periodicals; Biopharmaceutics -- Periodicals; Pharmacokinetics -- Periodicals; 615.19
• DOI: 10.1080/10543406.2016.1148709
• Languages: English
• ISSNs: 1054-3406

Abstract

ABSTRACT: In clinical investigations of diagnostic procedures to indicate noninferiority, efficacy is generally evaluated on the basis of results from independent multiple raters. For each subject, if two diagnostic procedures are performed and some units are evaluated, the difference in proportions for matched-pair data is correlated between the two diagnostic procedures and within the subject, i.e. the data are clustered. In this article, we propose a noninferiority test to infer the difference in the correlated proportions of clustered data between the two diagnostic procedures. The proposed noninferiority test was validated in a Monte Carlo simulation study. Empirical sizes of the noninferiority test were close to the nominal level. The proposed test is illustrated on data of aneurysm diagnostic procedures for patients with acute subarachnoid hemorrhage.