In a given year, more than 47 million U.S. adults have a diagnosable mental illness (
1). An estimated 11.4 million have a serious mental illness that substantially impairs participation in life activities (
1). For some adults, mental illness can be so disabling that they cannot attend to basic needs. One way to ensure that these adults receive needed care is by temporarily transferring decision-making responsibility to a court-appointed guardian who can monitor engagement in mental health treatment, secure placement in a structured residential setting, and manage income and benefits. Although mental health guardianship (referred to as “guardianship” in the following) has been criticized for promoting restrictive practices (
2), in some cases, it may be the only way to ensure that individuals with disabling mental illness receive needed care. However, because of shortages in mental health treatment at all levels of care (
3), individuals in the guardianship process often face long delays before receiving appropriate care.
Los Angeles County, the most populous county in the United States, has less than half of the number of acute psychiatric inpatient beds deemed minimally adequate for its population (
4,
5). Challenges within the guardianship process contribute to these shortages. In California, guardianship for individuals gravely disabled because of a mental illness is called Lanterman-Petris-Short (LPS) conservatorship (hereafter, “conservatorship”). Conservatorship is initiated when a clinician petitions the court for a temporary conservatorship, usually while the individual is in an acute psychiatric inpatient unit. A court process determines whether the individual meets criteria for grave disability, defined as “a condition in which a person, as a result of a mental disorder, is unable to provide for his or her basic personal needs for food, clothing or shelter” (
6). Individuals may first be held on an involuntary status for up to 20 days before the temporary conservatorship petition; they remain in an acute unit for 1–6 months, awaiting the court process (7). Once placed into conservatorship, individuals may remain in acute inpatient units for several months while waiting for a bed in a locked subacute facility (i.e., a locked institution for mental disease [IMD]). IMDs are state-licensed long-term care psychiatric facilities for individuals no longer needing acute psychiatric care but not clinically ready for community living. Acute inpatient psychiatric care is the most personally restrictive, expensive, and scarce level of care (
4), yet accessing conservatorship requires an individual to remain in such care for an extended period. Given the intensive resources required, many stakeholders suspect that conservatorships are underutilized or inequitably utilized in California.
Leadership in Los Angeles County and across California have considered diverse strategies to address inefficiencies and delays in the conservatorship process complicated by the limited capacity of acute care psychiatric hospitals. Local stakeholders recognize the challenges of serving the population of persons with serious mental illness who may become gravely disabled but have different views on how to address these challenges (unpublished communication: Report to the Los Angeles County Board of Supervisors of the DMH Subcommittee No. 2 on LPS Conservatorship, cochairs Erick Cheung and Jaime Garcia, January 14, 2018).
To critically assess and prioritize these strategies, we conducted a three-round online modified-Delphi panel with a diverse group of experts. Our aim was to provide a consensus-based ranking of potential strategies to improve services throughout the conservatorship process, determine whether experts agree on them, and delineate why certain strategies appear promising in the Los Angeles County context. Our findings have implications for setting priorities and communicating a vision of improved services for persons with serious mental illness who may be gravely disabled. The recommended strategies and associated implementation considerations can apply to other counties aiming to improve their conservatorship process.
Methods
Sample
The study protocol was approved by the Human Subjects Protection Committee at RAND and granted a waiver of review by the Office of the Human Research Protection Program at the University of California, Los Angeles, as a quality improvement program. We assembled experts with extensive knowledge of, or professional expertise in, conservatorship in Los Angeles County, including advocates, mental health court personnel, policy makers, researchers, and clinicians (e.g., social workers and physicians from inpatient and outpatient settings). To ensure that the panelists have the required expertise, we included only clinicians who routinely and over many years have petitioned the court for conservatorship, researchers with expertise in legal guardianship statutes, and policy makers overseeing conservatorship programs. We emailed 63 experts a link to a 2-minute video explaining the importance of the panel and the data collection process. Seven experts opted out, citing time constraints, leaving a sample of 56 experts.
Proposed Strategies
We conducted a rapid review of gray and academic literature to identify already proposed approaches for improving conservatorship of people with serious mental illness in Los Angeles County. The peer-reviewed literature was limited and primarily published in legal journals (
2,
8–
10). Despite the important implications for community mental health care and housing, the topic of improving conservatorship has gone relatively unexplored in peer-reviewed health and social science publications, except for some recent studies (
11,
12). Some research addressed legal or clinical alternatives to conservatorship (
13–
15), but neither were the focus of our panel. However, two sources were particularly informative: the Los Angeles County Department of Mental Health report “Expanding Conservatorship Capacity” (
16) and recommendations from the California LPS Reform Task Force II (
17).
To expand this list, we interviewed two clinicians and one researcher and obtained additional information about 17 candidate strategies through consultation with five individuals from policy, legal, and clinical settings. We grouped similar strategies and eliminated those already implemented in Los Angeles County or not addressing quality or capacity concerns related to conservatorship. We categorized the remaining nine strategies according to whether the improvement would be made before, during, or after the conservatorship process (see the
online supplement to this article).
Data Collection
We used ExpertLens, an evaluated online modified-Delphi platform (
18–
21), and key principles of the RAND/UCLA Appropriateness Method (RAM) (
22) to conduct the expert panel in May–June 2019. Experts provided input on each strategy by using the three-round panel process depicted in the
online supplement. They rated, commented on, and discussed the ethical appropriateness, impact on quality of care, feasibility, and efficiency of each strategy (see
online supplement).
Analysis
Determining agreement.
To determine existence of agreement, we used the RAM approach, which first looks at how answers are distributed across the tertiles of 9-point scales (with a score of 1 indicating “not efficient,” “not ethically appropriate,” “not feasible,” or “very unlikely to have positive impact on the overall quality of services” and a score of 9 indicating “very efficient,” “very ethically appropriate,” “very feasible,” or “very likely to have positive impact on the overall quality of services”) and then uses the median value to determine the panel’s response. If more than a third of responses to a question are in the upper and the lower tertiles, the panelists are deemed to disagree. If there is no disagreement, a median <3.5 indicates a negative final panel determination, meaning that a strategy is not ethically appropriate, unlikely to improve care quality, infeasible, or not efficient. Conversely, a median of ≥6.5 indicates that a strategy is ethically appropriate, likely to improve care quality, feasible, or efficient. A median between 3.5 and 6 denotes uncertain final panel determination.
Rank ordering.
We first identified which strategies experts agreed on and rated highly on all four criteria in round 3. We then identified which strategies lacked agreement or received lower ratings on some criteria. Finally, we rank-ordered all strategies according to their median ratings (starting with the highest median value) for each criterion in this order: ethical appropriateness, positive impact, feasibility, and efficiency.
Qualitative analysis.
We coded experts’ comments, identified the most salient themes, and classified each comment’s sentiment as either positive or negative, or as an implementation suggestion or consideration. We analyzed data with an iterative approach with a low level of inference (
23,
24). Two authors (A.L.S., D.K.) jointly coded comments from two strategies and developed the initial codes. One author (A.L.S.) coded the remaining strategies, refined the codebook, and applied it to previously coded strategies. To increase trustworthiness of the coding results, the same author engaged in peer debriefing with the study team to resolve questions and reach consensus (
25).
Results
Of the 56 experts, 46 (82%) participated in at least one round, of whom 43 (93%) provided input in round 1; 32 (70%) reviewed round-2 charts and discussion boards, with 21 (66%) posting their own discussion comments; and 33 (72%) responded to round-3 questions. Half of the 46 participating experts were women (N=23); 41% (N=19) were hospital-based clinicians and clinical administrators, 28% (N=13) worked at outpatient facilities, and the remaining 30% (N=14) represented legal, advocacy, policy, and forensic organizations.
Although all nine strategies were deemed ethically appropriate, panelists agreed with each other and provided high ratings across all four criteria for only one strategy (
Table 1). Experts either disagreed with each other or saw feasibility of the remaining strategies as uncertain, which could be attributed to either the need for more information or that the strategy required more resources. Seven strategies were considered efficient, and six were seen as likely to improve care quality.
Table 1 shows the rating results, and
Table 2 presents qualitative findings for the three highest-ranked strategies. (Appendix 3 in the
online supplement describes the remaining six strategies.)
The highest-ranked strategy—“improve the administrative functioning of the Los Angeles County Office of the Public Guardian (OPG) and the judicial processes in establishing and managing conservatorship”—was the only strategy rated highly on all criteria. The most salient qualitative theme centered on tele-testimony, which could greatly streamline the process by allowing clinicians—and possibly patients—to testify remotely. Clinicians can spend many valuable hours waiting at court and may not be notified of schedule changes in a timely manner. As one expert put it, “Tele-testimony would be very helpful. Our physicians spend an enormous amount of time at D95 [mental health court].” In the process of establishing and managing conservatorship, clinicians have a disincentive to appear in court and occasionally do not appear, which may result in repeated rescheduling, premature expiration of the conservatorship, and case dismissals. Tele-testimony could result in less wasted time, greater clinician availability, fewer trials missed by clinicians, and shorter hospital stays for patients. Allowing patients to testify offsite could also resolve transportation issues that hospitals face in transferring clients to and from court. Experts doubted that current technological capacities were sufficient (i.e., new systems would likely need to be purchased and installed) and noted potential resistance to change and fear of new technology, which was reflected in a lower score on feasibility. Nonetheless, they did not consider these barriers insurmountable, which may explain why this strategy was rated as more feasible than others. Overall, experts thought that the nominal costs for improving technology would considerably improve the process. However, some noted that streamlining would ultimately enable more people to be placed into conservatorship, so the strategy would be a good use of resources only if guardianship capacity increased correspondingly.
Another common theme was the need for improved coordination among public guardians, outpatient clinics, and hospital staff and the difficulty and cost associated with connecting and integrating them. Other less frequent comments focused on the need to reduce OPG caseloads to speed up the process. Some experts stated that in some cases, the California Supreme Court caused bottlenecks, rather than OPG. Generally positive comments expressing overall approval (e.g., “[This strategy] should be implemented”) greatly outweighed negative comments.
The second-highest-ranked strategy was to enhance the ability of full service partnership (FSP) programs or other intensive outpatient teams to identify and serve individuals before they become gravely disabled. FSP teams are modeled on the assertive community treatment model that provides field-based, multidisciplinary, and intensive clinical care to individuals with severe mental illness in California. Experts considered this strategy to be ethically appropriate, likely to have positive impact, and efficient but had concerns about its feasibility. Of all the strategies, however, the FSP approach raised the most concern among experts, who noted that explicit standards of care would need to be established and monitored and that FSP teams would need ongoing high-quality training in best practices. The resources needed to make this strategy successful may have been the reason for the lower feasibility rating. As one expert explained, “Feasibility will depend on the amount of resources available and consistent standards of services and outreach by the FSP service providers.” In addition, with greater outreach comes increased need for services and housing, which are already limited. Although experts liked that this strategy had the potential to provide care in the least restrictive environment possible, they noted the challenges of serving persons who lack insight into their illness or refuse services. This strategy was considered a good use of resources because it could prevent the need for more costly acute services.
The strategy “develop new unlocked outpatient settings for individuals after they have been conserved by expanding and enriching the continuum of unlocked residential environments” was ranked third. Comments indicated that this strategy may increase patient well-being. “Having a supportive place to live is essential to building on the gains of an inpatient stay,” one expert noted. This strategy was also rated positively on all criteria except feasibility, perhaps because of concerns about funding to implement it and the number of services and beds available. As with all outpatient strategies, experts mentioned the complexities of providing mandated care in an outpatient setting and the potential for elopements. They emphasized the need for screening to determine which patients would be successful in an outpatient setting, for outpatient providers to agree to provide mandated medications, and for establishing and monitoring standards of outpatient care.
Discussion
Our study is the first to critically examine the viability of strategies to improve the guardianship process in Los Angeles County. It incorporated diverse expert perspectives to produce recommendations applicable in other settings. Improving administrative functioning and judicial processes was the only strategy that panelists agreed on and rated highly on all four criteria. Experts perceived some opportunities that could greatly streamline the conservatorship process and make better use of scarce resources through technology. The Los Angeles Superior Court has already made strides toward paperless systems and records for mental health court proceedings and routinely uses tele-testimony in criminal proceedings (
26), as the California Court Technology Advisory Committee (formerly called Task Force) recommended 25 years ago (
27). However, at the time this panel was conducted, Los Angeles County had not yet to used tele-testimony for LPS conservatorship proceedings (
26). A recent survey of 333 California judicial officers found that 85 had used video technology in court proceedings, but only five had used it in conservatorship, guardianship, and mental health proceedings (
27). Video testimony was deemed a particularly beneficial potential improvement, and most judicial officers were satisfied with its use during proceedings (
27). Courts that do not use this technology should consider its potential for speeding up conservatorship trials, preserving clinician availability, and providing more timely care to individuals in need.
Strategies that expanded the use of community-based services were generally rated highly. These strategies include increasing the quality and assertiveness of outpatient treatment and fostering access to high-quality, unlocked residential settings. Although panelists endorsed tenets of deinstitutionalization and community integration in this way (
28,
29), some uncertainty about feasibility of community-based services may be explained by the commonly voiced concern that, until an individual is in a conservatorship, providers lack the authority to provide involuntary treatment. Panelists who raised concerns about community-based services pointed out that the very purpose of conservatorship is to care for those who will not voluntarily engage in needed services. However, proponents of community-based services suggested that with screening to identify appropriate patients, many patients could be successful in a stepped-down environment. Experts were more confident about the ability of outpatient settings to be successful once individuals had been placed in a conservatorship, arguing that it is necessary for outpatient settings to have the added authority that court orders confer or that a period of treatment in a hospital is generally appropriate for someone awaiting a conservatorship determination. Although our panelists indicated that increased capacity was needed at all levels of care, they felt that the high cost of locked IMD beds might add burden to an already underresourced system and draw resources away from less restrictive community-based settings. Nonetheless, increasing the number of locked IMD beds was seen as one of the two most efficient strategies.
Underlying the challenges of providing court-mandated care in unlocked community settings is the ongoing ethical and practical problem of providing involuntary treatment while preserving the patient’s right to the least restrictive environment possible. As our results show, the dominant qualitative theme pertained to appropriate levels of care and supervision, and experts carefully weighed each strategy against its perceived impact on individual autonomy. Pursuing legislative changes was the least desirable strategy in part because experts were concerned that it could impinge on individual rights by increasing the use of restrictive settings. Experts also cautioned that strategies to increase the use or capacity of locked IMDs must carefully consider patients’ right to receive care in the least restrictive environment.
For almost all strategies, experts noted the need to ensure the existence of high-quality services and processes for oversight and accountability. In particular, outpatient settings need to have established standards of care, evidence-based approaches to treatment, and external monitoring. Experts also noted that FSP teams can vary widely in quality and suggested that access to updated training and consistent adherence to best practices would improve the overall quality of the services. Their perception is confirmed by the literature, with a recent study of 93 California FSP teams finding large variation in implementation practices and philosophical approaches to housing placement and consumer choice (
30).
We used an innovative online modified-Delphi expert panel approach that allowed us to solicit input from diverse experts on ways to improve the conservatorship process in Los Angeles County. Although our expert sample was large, only 33 of the 46 experts provided their round-3 ratings used to rank-order the strategies; however, attrition is typical for Delphi panels (
31). Although diverse, our panel was dominated by medical professionals, with less than a third of experts representing legal, advocacy, policy, and forensic organizations. Finally, although our study focused specifically on Los Angeles County and on the strategies that have been discussed in this county, we believe the findings could be useful for other localities that serve individuals with serious mental illness through similar guardianship mechanisms. For example, establishing telecommunication court services and expanding the continuum of psychiatric services are likely to benefit mental health consumers regardless of locale.
Conclusions
Our results suggest a role for holistic reform that encompasses all stages of the guardianship process for individuals with mental illness who are gravely disabled. The three highest-ranked strategies were for interventions before, during, and after the conservatorship process, indicating readiness for improvements across all stages. Since our study, the Los Angeles County Department of Mental Health has already undertaken initiatives that address experts’ priority areas (
32). For instance, it has expanded the availability of unlocked outpatient residential settings able to serve persons in conservatorship and implemented a “conservatee FSP” program to increase the number of individuals in conservatorship living in the community served by an FSP team that coordinates with the OPG. A comprehensive quality improvement process is underway within the FSP program to ensure enrollment of individuals at risk for grave disability and to increase accountability to outcomes. As counties across California and the nation seek strategies to address high rates of chronic homelessness among individuals with mental illness, guardianship as a mechanism to engage hard-to-reach individuals will continue to be a key piece of a comprehensive mental health strategy (
33).
Acknowledgments
The authors thank Dr. Jonathan Sherin, Ms. Kei Nagao, and Ms. Anna Bruce for their commitment to advancing this work. They are also grateful to the following members of the DMH+UCLA Public Partnership for Wellbeing team for their important contributions: Dr. Patricia Lester, Ms. Gita Cugley, Mr. Joseph Mango, Dr. Nichole Goodsmith, Ms. Jaclyn Resnick, Dr. Lisa Davis, Dr. Caryn Bernstein, Ms. Elizabeth Mackey, Ms. Nancy Alfaro, and Ms. Sacha Fernandez. The authors express their gratitude to all participants in the expert panel.