Multiple informants: Mortality associated with psychiatric disorders in the stirling county study

This paper applies new statistical procedures for analyzing multiple-source information about the relation of psychiatric diagnoses to mortality. The data come from the Stirling County Study, a longitudinal community investig ation of adults, that collected multiple-source reports (self-report and ph ysician-report) about psychiatric disorders. These reports are used as pred ictors of mortality risk over a 16-year follow-up period (1952-1968). Despi te extensive efforts, one or both of these reports were sometimes missing. Missingness of self-report was related to demographic characteristics as we ll as to physician-reports of psychiatric diagnosis. The statistical proced ures used here draw together into a single frame of reference both informan t reports for the initial Stirling survey and relate these to mortality ris k using weighted generalized estimating equation regression models for time to event data. This unified method has two advantages over traditional app roaches: 1) the relative predictiveness of each informant can be assessed a nd 2) all subjects contribute to the analysis. The methods are applicable t o other areas of epidemiology where multiple informant reports are used. Th e results for self-reports and physician-reports of disorders were comparab le: Psychiatric diagnosis was associated with higher mortality, particularl y among younger subjects.