Risks for re-hospitalization of persons with severe mental illness living in rehabilitation care settings.

Background: The high rates of psychiatric re-hospitalizations (also termed "revolving door") presents a "wicked problem" which requires a systematic and holistic approach to its resolution. Israel's mental-health rehabilitation law provides a comprehensive set of services intended to support the ability of persons with severe mental illness to rely on community rather than in-patient facilities for their ongoing care needs. Guided by the Health Behavior Model, we examined the relationship between psychiatric re-hospitalizations and the three Health Behavior Model factors (predisposing factor: socio-demographic characteristics and health beliefs; enabling factor: personal and social/vocational relationships facilitated by rehabilitation interventions and services; and need factor: outcomes including symptoms, and mental health and functional status) among persons with severe mental illness receiving rehabilitation services.

Methods: Logistic regression models were used to measure the association between re-hospitalization within a year and variables comprising the three Health Behavior Model factors on the sample of consumers utilizing psychiatric services (n = 7,165). The area under the curve for the model was calculated for each factor separately and for all three factors combined.

Results: A total of 846 (11.8%) consumers were hospitalized within a year after the study began. Although multivariable analyses showed significant associations between re-hospitalization and all three Health Behavior Model factors, the magnitude of the model's area under the curve differed: 0.61 (CI = 0.59-0.64), 0.56 (CI = 0.54-0.58), 0.78 (CI = 0.77-0.80) and 0.78 (CI = 0.76-0.80) for predisposing, enabling, need and the full three-factor Health Behavior Model, respectively.

Conclusion: Findings revealed that among the three Health Behavior Model factors, the need factor best predicted re-hospitalization. The enabling factor, comprised of personal relationships and social/vocational activities facilitated by interventions and services representing many of psychiatric rehabilitation's key goals, had the weakest association with reduced rates of re-hospitalization. Possible explanations may be inaccurate assessments of consumers' personal relationships and social/vocational activities by the mental healthcare professionals, problematic provider-consumer communication on the consumers' involvement in social/vocational activities, or ineffective methods of facilitating consumer participation in these activities. Clearly to reduce the wicked "revolving-door" phenomenon, there is a need for targeted interventions and a review of current psychiatric rehabilitation policies to promote the comprehensive integration of community rehabilitation services by decreasing the fragmentation of care, facilitating continuity of care with other healthcare services, and utilizing effective personal reported outcomes and experiences of consumers with severe mental illness.