Clinical and economic influences on mental health triage decisions

There are few multivariate models to explore emergency mental health triage decisions. To date, no models have been found that assess the effect of clinician concern for liability and the patient's reimbursement resources in concert with client and clinician characteristics on emergency psychiatric admission decisions.; This study explores institution-specific emergency mental health triage decisions for inpatient hospitalization or outpatient follow-up. Hypothesized influences on the disposition decision included: (a) the presence and type of the patient's health insurance; (b) patient demographics; (c) clinician characteristics, including sense of liability if the patient was not hospitalized; and (d) the clinical assessment of the patient.; The sample included 400 patients seen in a midwestern, urban not-for-profit hospital for emergency psychiatric and substance abuse and the clinicians performing the evaluation. Research tools were the patient's medical record, a survey of clinicians, the Global Assessment Scale, the Brief Psychiatric Rating Scale, and two tools developed for this research: the Problem Rating Scale and the Legal Liability Scale. All tools demonstrated adequate reliability.; The clinician's sense of liability if the patient was discharged, the patient's clinical acuity and insurance status were used to correctly predict admission or discharge for 95% of the sample. Other significant predictors were danger to self or others, diagnosis and history of previous admissions; however these were subsumed by the previous three variables. The clinician's sense of liability was the single most important predictor of disposition and was only partially explained by clinical acuity. No relationship was found between disposition and age, ethnicity, marital or employment status, education, the evaluating clinician or time of evaluation.; The model was calculated using factor analysis, stepwise logistic regression and multiple nominal scale analysis. The derived model can be used to assess the relative impact of included variables as contingencies in the disposition decision and as a quality control tool to evaluate clinicians' decision-making and consistency.