A comparison of decisions to discharge committed psychiatric patients between treating physicians and district psychiatric committees: an outcome study

BACKGROUND: The Israel Mental Health Act of 1991 stipulates a process for involuntary psychiatric hospitalization (IPH). A patient thus hospitalized may be discharged by either the treating psychiatrist (TP) or the district psychiatric committee (DPC). The decision rendered by the DPC is often at odds with the recommendation of the TP. Although much has been written about the ethical issues of restricting patients’ rights and limiting their freedom, far less attention has been devoted to the psychiatric, medical, and social outcome of legal patient discharge against the doctor’s recommendation. METHODS: In our study we examined the outcomes of the decisions made by the DPC using readmission data, an internationally recognized indicator of the quality of hospital care, and compared them to the outcomes of patients discharged by the TP. All IPH discharges resulting from the DPC’s determination for the year 2013 (N = 972) were extracted from the Israel national register. We also collected all IPH discharges owing to the TP’s decision for 2013 (N = 5788). We defined “failure” as readmission in less than 30 days, involuntary civil readmission in less than 180 days, and involuntary readmission under court order in less than 1 year. RESULTS: The rehospitalization pattern was compared in the two groups of patients discharged from their psychiatric hospitalization during 2013 (index discharges) and followed up individually for a year. We found a statistically significant difference between the DPC and the TP group for each of the time frames, with the DPC group returning to IPH much more frequently than the TP group. Using cross-sectional comparison with logistic regression adjusted for age, gender, diagnosis and length of hospitalization, we found the probability of a decision failure in the TP group was significantly less with an OR of 0.7 (95% CI .586–.863), representing a 30% adjusted decrease in the probability for failure in the TP group. CONCLUSIONS: The results we present show that the probability of decision “failure” (readmission) was found to be significantly higher in the DPC group than in the TP group. It is often assumed that IPH patients will fare better at home in their communities than in a protracted hospitalization. This is frequently the rationale for early discharge by the DPC (30.1 days vs. 75.9 DPC and TP groups, respectively). Our results demonstrate that this rationale may well be a faulty generalization.