摘要:Objectives. In November 2004, California voters passed the Mental Health Services Act, which allocated more than $3 billion for comprehensive community mental health programs. We examined whether these county-level programs, known as “full service partnerships,” promoted independent living arrangements (i.e., recovery) among their clients. Methods. We used Markov chain models to identify probabilities of residential transitions among 8 living arrangements (n = 9208 adults followed up to 4 years). We modeled these transitions on the basis of patterns of program participation and clinical and sociodemographic characteristics. Results. Interrupted program participation and substance abuse were significantly associated with a reduced likelihood of independent living and a greater probability of homelessness and incarceration. Persons with schizophrenia were the least likely to live independently, followed by persons with bipolar disorder. Compared with Whites, non-Whites were more frequently found to be homeless or in jail. Conclusions. Clients with sustained enrollment in California’s comprehensive community mental health programs appear most likely to transition to independent living. The likelihood of this transition, however, shows a disparity in that ethnic minority clients appear least likely to transition to independent living. The US mental health system in the 21st century continues to transform. Institutional mental health care, especially public inpatient care, has significantly declined since the mid-1950s. 1–4 As a substitute for public inpatient care, community-based mental health treatment has expanded to serve the increasing fraction of persons with severe mental illness residing in the community. 1–3 In 2007, approximately 14 million persons received care in 1067 community-based facilities nationwide. 5 Between 1981 and 2005, state mental health agencies’ expenditures for community mental health programs as a percentage of total expenditures increased more than 2-fold, from 33% to 70%. 6 Despite this dramatic growth of community mental health programs, critics assert that inadequate funding hinders the successful integration of persons with severe mental disorders into the community. Recently, Yoon and Bruckner 7 reported that at current funding levels in the United States, community mental health programs do not buffer the adverse health effects of the decline in public inpatient care. Markowitz 8 reached a similar conclusion: community mental health programs do not offset the increase in crime and homelessness that coincides with public inpatient reductions. Other researchers, 9,10 moreover, argue that community mental health care may not adequately treat severely mentally ill patients with a history of dangerousness, co-occurring disorders, or arrests. In response to such criticisms, several states have experimented with shifting their public mental health system toward comprehensive and recovery-oriented programs in community settings. 11–15 California’s Mental Health Service Act (MHSA) is the most comprehensive and well-funded effort to date in the United States. 16 As of early 2010, the MHSA distributed approximately $3.7 billion to counties on the basis of county requests. Intensive community treatment programs have served approximately 25 000 clients, and all MHSA programs have served more than 400 000 clients. 17 Funding for the MHSA comes from a 1% tax levied on adjusted gross incomes over $1 million. 16 This tax resulted in an 18% increase in funding in fiscal year 2008–2009 over the level of funding in fiscal year 2003–2004. 18 Full service partnerships (FSPs), which the MHSA created, are community treatment programs that provide intensive case management and services, including housing, employment, education, peer support, and outreach. According to the MHSA, persons may gain admission into FSPs if they meet specific criteria. 19 These criteria include the following: qualifies for public assistance, was diagnosed with a severe mental illness, and is viewed by county officials as unserved or underserved. In addition, depending on age, persons must have at least 1 of the following characteristics: homeless or at risk for homelessness, involved or at risk for involvement with the criminal justice system, frequently hospitalized for mental health problems, or frequent user of emergency department services (Brown et al. 20 provide a detailed description). Two recent reports 20,21 find several benefits of enrollment in FSP programs in the short term, including a reduction in incarceration and emergency hospitalization rates. However, despite California’s efforts to create a more effective community mental health system, little is known about factors affecting the transition to independent living—known as recovery—among clients these comprehensive programs serve. 22 We analyzed, from the perspective of the FSP program, whether participation in FSPs varies positively with clients’ transition to recovery. We followed residential status of FSP clients up to 4 years after the initiation of the FSP program. We used changes in living arrangements as indicators of recovery. We considered a transition to independent living the highest level of recovery. Movements out of jail or prison and out of homelessness also are initial steps toward recovery. We also identified individual characteristics—such as psychiatric and substance abuse diagnoses and race/ethnicity—that may facilitate (or impede) transitions toward recovery.