5272.1: Wednesday, November 15, 2000 - Board 8

Abstract #1601

Effect of geographic migration on prevalence estimation

Bruce Dembling, PhD, Southeastern Rural Mental Health Research Center, University of Virginia, Box 393 Health Sciences, Charlottesville, VA 22908, 804-982-3276, dembling@virginia.edu and Michael B. Blank, PhD, Center for Mental Health Policy and Services Research, University of Pennsylvania, 3600 Market St, Room 707, Philadelphia, PA 19104-2648.

Introduction: The prevalence of serious mental illness (SMI) varies by socioeconomic characteristics of communities. Variation is assumed due to a combination of differential incidence (social causation) and differential migration (social selection and drift). This study determines the predictors of differential migration, and assesses the impact of migration on epidemiological estimates of SMI. Methods: A population of 11,725 adults with 3 or more admissions who used inpatient state psychiatric care in Virginia between July 1978 and November 1992 was sampled. Models were fit to predict individual migration, and to simulate the effect on county level prevelance estimates over time. Results: One third of cases migrated to different counties between first and last admission. Migration was higher for whites than African Americans, and was lower for women and married patients. Patient migration was generally from small rural and large urban counties toward medium sized urban counties. This flow was not correlated with general population trends. In a logistic regression model the odds of migrating decreased with age and for married persons, and increased with the number of admissions, total length of inpatient stays, and overall period of time under study. Conclusions: A substantial percent of psychiatric patients change county residence over time. The concentration of SMI prevalence in certain communities is due in part to this migration. Public needs assessments and resource allocation policies may under-state the need in communities with a net SMI out-migration, and in the long term these same policies may induce migration through resource allocation decisions.

Learning Objectives: 1. Recognize factors associated with differential migration of SMI populations 2. Identify service system policies which induce migration in an SMI population 3. Identify potential bias in synthetic prevalence estimations

Keywords: Rural Populations, Social Class

Presenting author's disclosure statement:
Organization/institution whose products or services will be discussed: None
I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.

The 128th Annual Meeting of APHA