Estimation of Response Probability in Correlated Binary Data: A New Approach
- PDF / 421,752 Bytes
- 7 Pages / 504 x 719 pts Page_size
- 60 Downloads / 165 Views
Estimation of Response Probability in Correlated Binary Data: A New Approach* Sin-Ho Jung and Chul Ahn Drug Information Journal 2000 34: 599 DOI: 10.1177/009286150003400228 The online version of this article can be found at: http://dij.sagepub.com/content/34/2/599
Published by: http://www.sagepublications.com
On behalf of:
Drug Information Association
Additional services and information for Drug Information Journal can be found at: Email Alerts: http://dij.sagepub.com/cgi/alerts Subscriptions: http://dij.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav Citations: http://dij.sagepub.com/content/34/2/599.refs.html
>> Version of Record - Apr 1, 2000 What is This?
Downloaded from dij.sagepub.com at NATIONAL CHUNG HSING UNIV on April 12, 2014
Drug Information Journal, Vol. 34. pp. 599-604. 2000 Printed in the USA. All rights reserved.
M)92-8615/2000 Copyright 0 2000 Drug Information Association Inc.
ESTIMATION OF RESPONSE PROBABILITY IN CORRELATED BINARY DATA: A NEW APPROACH* SIN-HOJUNG,PHD Associate Professor, Division of Biostatistics, Indiana University School of Medicine, Indianapolis, Indiana
CHULAHN,PHD Associate Professor, Clinical Epidemiology, University of Texas Medical School, Houston, Texas
We consider analysis of binary observations from multiple sites of each subject. In this case, observationsfrom the same subject tend to be correlated. In estimating the common response probability in correlated binary data, two weighting systems have been most popular: equal weights to sites, and equal weights to subjects. When the number of sites varies subject by subject, performance of these two weighting systems depends on the extent of correlation among sites within each subject. In this paper; we describe a new weighting method that minimizes the variance of the estimatol: We apply these methods to data from a study involving an enzymatic diagnostic test to illustrate the estimation of the sensitivity and the specificity of periodontal diagnostic tests. Simulation studies were conducted to compare the performance of the new estimator with that of other estimators. Key Words: Intraclass correlation; Sensitivity; Specificity; Optimal weight
INTRODUCTION CORRELATED BINARY data arise frequently in many fields of applications. Examples include the presence of arthritic pain in one or more joints and the occurrence of regional lymph node metastases in cancer patients. For site-specific observations, the fundamental unit for statistical analysis is “site” rather than “subject.” In periodontal diagnostic tests, “site” refers to a tooth surface within a subject’s mouth. The results for A portion of the work was presented at the DIA Work-
shop “Statistics in Diagnostic Imaging,” March 1-2, 1999, Crystal City, Virginia. Reprint address: Chul Ahn, PhD, Clinical Epidemiology, University of Texas Medical School, 6431 Fannin St., MSB 1.122, Houston, TX 77030. *This work was supported in part by NIH grants M01RR02588 and MH-324
Data Loading...
