Multiple Comparisons of Disease Prevalence in the Presence of Misclassification
Other conference paper


Full Text

Other information
AbstractThe comparison of disease prevalence for different populations is an important medical research problem in the area of public health. It is frequent that both high-cost infallible gold-standard classifiers and low-cost fallible classifiers are available to classify individuals as having or not having a disease. In order to obtain a reasonable estimate of the disease prevalence rate under the constraint of limited resources, it is not unusual to proceed with a diagnostic scheme with which a small group of individuals in the sample receives both the gold-standard test and the fallible classifier, while the rest receive only the fallible classifier. Statistical methods to compare disease prevalence of two populations based on the above framework are available. The objective of this study is to extend previous research to situations where multiple comparisons of disease prevalence for several populations are required. We also provide the methods to determine the required sample size when test power is specified. Furthermore, our proposed procedure is illustrated by data extracted from a disease prevalence study.
All Author(s) ListYANG Ping, POON Wai Yin, CHEUNG Siu Hung
Name of Conference2014 Joint Statistical Meetings
Start Date of Conference02/08/2014
End Date of Conference07/08/2014
Place of ConferenceBoston
Country/Region of ConferenceUnited States of America
Proceedings Title2014 Joint Statistical Meetings
Year2014
Month8
Place of PublicationUnited States of America, Boston
LanguagesEnglish-United Kingdom
KeywordsDisease Prevalence; Misclassification; Multiple comparisons

Last updated on 2019-30-10 at 15:11