Predictors of Psychiatric Hospitalization After Discharge From Inpatient Neurorehabilitation for Traumatic Brain Injury.

Abstract

Objective: To examine, among persons discharged from inpatient rehabilitation for traumatic brain injury (TBI), the degree to which pre-TBI factors were associated with post-TBI hospitalization for psychiatric reasons. The authors hypothesized that pre-TBI psychiatric hospitalization and other pre-TBI mental health treatment would predict post-TBI psychiatric hospitalization following rehabilitation discharge, up to 5 years post-TBI.

Setting: Five Veterans Affairs Polytrauma Rehabilitation Centers.

Participants: Participants with nonmissing rehospitalization status and reason, who were followed at 1 year (N = 1006), 2 years (N = 985), and 5 years (N = 772) post-TBI.

Design: A secondary analysis of the Veterans Affairs TBI Model Systems, a multicenter, longitudinal study of veterans and active-duty service members with a history of mild, moderate, or severe TBI previously admitted to comprehensive inpatient medical rehabilitation. This study examined participants cross-sectionally at 3 follow-up timepoints.

Main measures: Psychiatric Rehospitalization was classified according to Healthcare Cost and Utilization Project multilevel Clinical Classifications diagnosis terminology (Category 5).

Results: Rates of post-TBI psychiatric hospitalization at years 1, 2, and 5 were 4.3%, 4.7%, and 4.1%, respectively. While bivariate comparisons identified pre-TBI psychiatric hospitalization and pre-TBI mental health treatment as factors associated with psychiatric rehospitalization after TBI across all postinjury timepoints, these factors were statistically nonsignificant when examined in a multivariate model across all timepoints. In the multivariable analysis, pre-TBI psychiatric hospitalization was significantly associated with increased odds of post-TBI psychiatric hospitalization only at 1-year post-TBI (adjusted odds ratio = 2.65; 95% confidence interval, 1.07-6.55, P = .04). Posttraumatic amnesia duration was unrelated to psychiatric rehospitalization.

Conclusions: Study findings suggest the limited utility of age, education, and pre-TBI substance use and mental health utilization in predicting post-TBI psychiatric hospitalization. Temporally closer social and behavior factors, particularly those that are potentially modifiable, should be considered in future research.