Health status, health care utilization, and drug treatment outcomes

Abstract

The relationship between health status and emergency room utilization was prospectively examined in a residential, inpatient drug-free therapeutic community program. Several variables were investigated as possible predictors of emergency room utilization, including physical and mental health, positive and negative affect, and Brickman's Models of Helping and Coping. Treatment retention and resident satisfaction with treatment were also considered. Study participants included 154 admissions (61% male, 39% female, 59% Black, 32% Hispanic, 9% White) to a residential drug-free program over an 18-week period. Approximately two weeks after admission to drug treatment study participants completed a number of study measures and were then prospectively followed for up to four months after admission in regards to health status and emergency room use. Nursing Department records and medical charts were used to check for health problems and to track emergency room use and number of physician visits. Study measures included one of five domains: demographic variables, symptom variables, disease variables, psychosocial variables, and health care utilization variables. Three months after study enrollment participants were re-administered the Tennessee Self-Concept Scale:2 (TSCS:2) as a measure of change in self-concept, with significant improvements noted over time. Overall results indicated a significant relationship between lower physical and mental component scale scores on the Medical Outcome Studies' Short Form-36 Health Survey (SF-36) and the prospective use of an emergency room. Likewise, study participants who had higher levels of negative affect were also more likely to use an emergency room, whereas positive affect was unrelated to emergency room utilization. Important gender differences included women were more likely than men to use an emergency room. Although women had significantly more physician visits than men did overall, when gynecological visits were removed, men and women had a comparable number of physician visits. In addition, women reported significantly more physical symptoms than men did on the Primary Care Evaluation of Mental Disorders (PRIME-MD) and women were also more likely to report experiencing past trauma. Brickman's Models were not related to emergency room use in this sample of drug users. The implications of these results are discussed and suggestions for future research are considered.