Psychiatric rehospitalization among elderly persons in the United States

Psychiatr Serv. 2008 Sep;59(9):1038-45. doi: 10.1176/ps.2008.59.9.1038.

Abstract

Objective: This study examined predictors of psychiatric rehospitalization among elderly persons.

Methods: Readmission within six months of an index hospitalization was modeled by using Medicare data on all hospitalizations with a primary psychiatric diagnosis in the first half of 2002 (N=41,839). Data were linked with state and community-level information from the U.S. census.

Results: Twenty-two percent of beneficiaries were rehospitalized for psychiatric reasons within six months of discharge. After the analyses adjusted for sociodemographic factors, readmission was most likely among persons with a primary diagnosis of schizophrenia (hazard ratio [HR]=2.63), followed by bipolar disorder (HR=2.51), depression (HR=1.75), and substance abuse (HR=1.38) (reference group was "other" psychiatric conditions). A baseline hospital stay of five or more days for an affective disorder was associated with a reduced readmission hazard (HR=.68, relative to shorter stays), yet the opposite was true for a nonaffective disorder (HR=1.26). For persons with nonaffective disorders, an elevated hazard of readmission was associated with comorbid alcohol dependence (HR=1.32), panic disorder (HR=1.76), borderline personality disorder (HR=2.33), and drug dependence (HR=1.17). However, for persons with affective disorders, having a personality disorder other than borderline personality disorder or dependent personality disorder (HR=1.27) and having an "other" anxiety disorder (HR=1.15) were significantly associated with an increased risk of rehospitalization. Obsessive-compulsive disorder increased the readmission hazard in both groups.

Conclusions: Readmission risk factors may differ for affective disorders and nonaffective disorders. Very short hospitalizations were associated with increased risk of rehospitalization among persons with an affective disorder, which underscores the need for adequate stabilization of this group of patients during hospitalization. Results also highlight the specific types of comorbidities associated most strongly with rehospitalization risk.

Publication types

  • Research Support, N.I.H., Extramural
  • Research Support, Non-U.S. Gov't
  • Research Support, U.S. Gov't, P.H.S.

MeSH terms

  • Aged
  • Aged, 80 and over
  • Alcoholism / epidemiology
  • Alcoholism / therapy
  • Anxiety Disorders / epidemiology
  • Anxiety Disorders / therapy
  • Bipolar Disorder / epidemiology
  • Bipolar Disorder / therapy
  • Comorbidity
  • Depressive Disorder / epidemiology
  • Depressive Disorder / therapy
  • Female
  • Humans
  • Length of Stay / statistics & numerical data
  • Male
  • Medicare / statistics & numerical data
  • Mental Disorders / epidemiology*
  • Mental Disorders / therapy
  • Patient Readmission / statistics & numerical data*
  • Personality Disorders / epidemiology
  • Personality Disorders / therapy
  • Proportional Hazards Models
  • Risk
  • Schizophrenia / epidemiology
  • Schizophrenia / therapy
  • Substance-Related Disorders / epidemiology
  • Substance-Related Disorders / therapy
  • United States