Impact of length of stay for first psychiatric admissions on the ratio of readmissions in subsequent years in a large Brazilian catchment area.

Abstract

This study aims to verify the impact that the length of stay has on the rates of readmission for patients who were first admitted to various inpatient psychiatric units in a large catchment area in a middle-income country. The study included all patients who were first admitted to the 108 acute psychiatric beds available in the catchment area of Ribeirão Preto, Brazil, for a period of 8 years. Demographic features, inpatient unit of discharge, diagnosis and length of stay were assessed by bivariate analysis. An analysis of the time span between first admission and readmission was also conducted using survival curves estimated by the Kaplan-Meier formula. For the analyses of the risk of readmissions, a logistic regression analysis was conducted. From a total of 6261 patients admitted in the period of the survey, approximately one-third (2006) had at least one other readmission during the follow-up period. The rates per year of early readmission (within 90 days after discharge) varied from 16.1 to 20.9 %. The risk of readmission was higher immediately after discharge. The survival analysis showed that ultrashort length of stay (1-2 days) was associated with reduced odds of readmission, but multivariate logistic analysis showed no association between length of stay and the odds of readmissions. The predictors of early readmission included the diagnosis of depressive, bipolar, psychotic, and non-alcohol-related disorders, younger ages and unemployment. Duration of the first psychiatric admission was not associated with a higher risk of readmissions. Predictors for early readmissions of first-time-admitted psychiatric patients seem to be more related to the severity of the psychiatric diagnosis and demographic characteristics.