American Journal of Epidemiology Vol. 129, No. 2: 430-438
Copyright © 1989 by The Johns Hopkins University School of Hygiene and Public Health
research-article |
ESTIMATING THE MORBIDITY RISK OF ILLNESS FROM SURVEY DATA1
Reprint requests to Dr. Stephen C. Newman, Department of Psychiatry, Mackenzie Centre, University of Alberta, Edmonton, Alberta, Canada, T6G 2B7
A method is proposed for using survey data to estimate lower and upper bounds for the lifetime risk of an illness (morbidity risk). The mathematical model used, which is based on a three-state Markov process, assumes that the illness is irreversible, but allows differential mortality. The data required include information on age at onset collected from prevalent cases, and an estimate of the magnitude of differential mortality, which ordinarily must come from published research. The method is illustrated using data from a community survey of psychiatric illness conducted in Edmonton, Alberta, Canada
biometry; data collection; epidemiologic methods; morbidity; probability
1Department of Psychiatry, University of Alberta, Edmonton, Alberta, Canada, and Mental Health Services, Alberta Department of Community and Occupational Health, Edmonton, Alberta, Canada