Predictors of substance abuse treatment outcome in hospitalized veterans.

Background And Objectives: Historically patients consulted for the substance abuse treatment from the medical surgical floors have a very low show rate for the substance abuse treatment. The authors performed retrospective chart review to find predictors of substance abuse treatment outcome in hospitalized veterans at Atlanta VA Medical Center.

Methods: The medical records from all the patients who were admitted to the medical/surgical floor with substance abuse consults from January-December 2009 were reviewed. A total of 235 consults were received. Those records were examined to find the predictors for substance abuse treatment.

Results: Multiple variables were tested for significance - patient demographics, housing status, employment, reason for hospitalization, toxicology screens, co-morbid psychiatric and medical conditions, physician visits, and patients on waiting list. All variables were given cut-off point for the p-value of .10. These variables were then included in the logistic regression model. It was found that homelessness (χ2 = 16.14 and p < .0001) was the only individual variable that showed a statistically significant correlation with starting the program. It was found that homelessness (χ2 = 19.21 and p < .0001) was the only individual variable that showed statistically significant correlation with completing the program.

Conclusions And Scientific Significance: Our study supports that for veterans with substance abuse, housing was the only consistent predictor to enter intensive outpatient program (IOP), complete IOP, and start aftercare. Our study demonstrates the need for and potential benefit of providing stable housing for the homeless veterans.