Effect of geographic migration on SMI prevalence estimates.

Abstract

The prevalence of serious mental illness (SMI) varies by the socioeconomic characteristics of communities. This variation is presumed to be due to the differential incidence of disorders caused by adverse social factors (social causation) and differential geographic migration (social selection and drift). The objective of this study was to measure the geographic migration patterns of adults treated for SMI. A sample of 11,725 adults with three or more psychiatric hospital admissions between July 1978 and November 1992 was drawn from inpatient records. At least one third migrated to different counties between first and last admission. Migration rates were higher for Whites than for African Americans and higher for unmarried than for married patients. There were no significant differences in migration rates by gender or mental disorder. Patient migration did not parallel shifts in the general population. Patient migration was generally toward medium-size, low-income urban counties with relatively declining general populations, and movement was away from both the most rural and the largest urban counties. Public needs assessments and resource allocation policies may understate the need in communities with net SMI out-migration and overstate endemic need where there is net in-migration. In the long term, these same policies may induce migration through resource allocation decisions.