Comparing two prevalence rates in a two-phase design study

An epidemiological study often uses a two-phase design to estimate the prev alence rate of a mental disease. In a two-phase design study, the first pha se assesses a large sample with an inexpensive screening test, and then the second phase selects a subsample for a more expensive diagnostic evaluatio n. Furthermore, disease status may not be ascertained for all subjects who are selected for disease verification because some subjects are unable to b e clinically assessed, while others may refuse. Since not all screened subj ects are selected for diagnostic assessments, there is potential for verifi cation bias. In this paper, we propose the maximum likelihood (ML) and boot strap methods to correct for verification bias for estimating and comparing the prevalence rates under the missing-at-random (MAR) assumption for the verification mechanism. We also propose a method to test this MAR assumptio n. Finally, we apply our methods to a large-scale prevalence study of demen tia disorders. Copyright (C) 1999 John Wiley & Sons, Ltd.