Psychiatric and Substance Use Disorder Bed Capacity, Need, and Shortage Estimates in California: Merced, San Joaquin, and Stanislaus Counties
RAND Health Quarterly, 2023; 10(2):6
RAND Health Quarterly, 2023; 10(2):6
RAND Health Quarterly is an online-only journal dedicated to showcasing the breadth of health research and policy analysis conducted RAND-wide.
More in this issuePsychiatric and substance use disorder (SUD) treatment beds are essential infrastructure for meeting the needs of individuals with behavioral health conditions. However, not all psychiatric and SUD beds are alike: They represent infrastructure within different types of facilities. For psychiatric beds, these vary from acute psychiatric hospitals to community residential facilities. For SUD treatment beds, these vary from facilities offering short-term withdrawal management services to others offering longer duration residential detoxification services. Different settings also serve clients with different needs. For example, some clients have high-acuity, short-term needs; others have longer-term needs and may return for care on multiple occasions. California's Merced, San Joaquin, and Stanislaus Counties, like other counties throughout the United States, have sought to assess shortages in psychiatric and SUD treatment beds. In this study, the authors estimated psychiatric bed and residential SUD treatment capacity, need, and shortages for adults and children and adolescents at various levels of care: acute, subacute, and community residential services for psychiatric treatment and SUD treatment service categories defined by American Society of Addiction Medicine clinical guidelines. Drawing from various data sets, literature review findings, and facility survey responses, the authors computed the number of beds required—at each level of care—for adults and children and adolescents and identified hard-to-place populations. The authors draw from these findings to offer Merced, San Joaquin, and Stanislaus Counties recommendations to help ensure all their residents, especially nonambulatory individuals, have access to the behavioral health care that they need.
Psychiatric and substance use disorder (SUD) treatment beds are essential infrastructure for managing the clinical and psychosocial problems associated with behavioral health disorders. The levels of care for psychiatric beds range from acute psychiatric hospitals to community residential facilities (McBain, Cantor, and Eberhart, 2022). SUD treatment beds range from facilities offering short-term withdrawal-management services to others offering longer-duration residential detoxification services (California Department of Health Care Services, 2022d). Patient placement into different settings depends on the facility's alignment with patient needs. For example, some patients may have high-acuity, short-term needs; other patients may require longer-term accommodation and recurrent, episodic care.
General assessment of California and other regions of the United States demarcated need for additional behavioral health beds, but the extent to which beds are available and accessible at each level of care is unknown (Technical Assistance Collaborative and Human Services Research Institute, 2012). The northern San Joaquin Valley counties of Merced, San Joaquin, and Stanislaus have sought to quantify shortages in psychiatric and SUD treatment beds, in an effort to further California's investment to expand the behavioral health care continuum and enhance infrastructure capacity (Sacramento County Mental Health Board, 2019). The current work builds off of two previous reports: first, a report that examined California's statewide and regional adult psychiatric bed needs, and, second, a report that examined the specific adult and adolescent psychiatric and SUD bed need in Sacramento County (McBain, Cantor, Eberhart, Huilgol, and Estrada-Darley, 2022; McBain, Cantor, Eberhart, Crowley, and Estrada-Darley, 2022). This work expands the service area under study to include multiple California counties within a geographic region.
This study estimated inpatient and residential psychiatric and SUD treatment bed capacity, need, and shortages at specific levels of care across Merced, San Joaquin, and Stanislaus Counties. Psychiatric beds were evaluated at three levels of care: acute, subacute, and community residential services. Acute care is directed toward stabilizing patients with the highest-acuity needs, typically provided in the shorter term (days to weeks). Subacute care is directed toward those with moderate- to high-acuity needs for a longer duration (multiple months). Community residential services are intended to address lower-acuity and longer-term care (often multiple years) that is focused on patient recovery. For SUD treatment services, we defined service categories based on the American Society of Addiction Medicine (ASAM) clinical guidelines (American Society of Addiction Medicine, undated). Specifically, we used the definitions for residential detoxification services (ASAM level 3.2); residential long-term treatment, which is typically more than 30 days (ASAM level 3.3); and residential short-term treatment, which is typically 30 days or fewer (ASAM level 3.5).
The geographic area under study included Merced, San Joaquin, and Stanislaus Counties, which we examined at the aggregate level. The populations of interest within these counties consisted of both adults and children. We aimed to conduct survey interviews with administrators at every psychiatric and SUD treatment facility in these counties that provided inpatient or residential services.
To estimate bed capacity, we combined several data sets from state agencies that are responsible for licensure of psychiatric or SUD treatment beds. Additionally, stakeholders from the counties’ Departments of Health Services provided up-to-date feedback to refine the list of closed and in-service facilities. The research team supplemented this information by administering a phone-based survey of all psychiatric and SUD treatment facilities across Merced, San Joaquin, and Stanislaus Counties to collect data on bed occupancy rates, waiting list volume, and requested transfers to higher and lower levels of care.
We triangulated estimates from several approaches to estimate psychiatric and SUD treatment bed need. First, we used the survey data to compute the number of beds required—at each level of care—to reduce bed occupancy rates in the region to 85 percent (a standard ceiling), as well as accommodate waiting list volume and requested transfers. Second, our research team conducted an environmental scan of both the academic and gray literature to identify normative and descriptive benchmarks for psychiatric and SUD treatment bed capacity and need. Using both sources of data, we were able to compare the bottom-up estimates of need, based on observed outcomes at facilities in the three counties, with the top-down estimates of need based on thresholds outlined by experts or otherwise established in various jurisdictions at local, national, and international levels.
Additionally, we conducted online forums with local stakeholders to get their perspectives on bed shortages in the region.
We estimated that Merced, San Joaquin, and Stanislaus Counties have a total of 186 adult beds at the acute level (15.7 per 100,000 adults) and 330 adult beds at the subacute level (27.8 per 100,000). At the community residential level, we estimated that the counties have a total of 467 adult beds (39.4 per 100,000). For children, we found 219 psychiatric beds (49.5 per 100,000) in total: 26 beds at the acute level (5.9 per 100,000 children), 12 beds at the subacute level (2.7 per 100,000 children), and 181 at the community residential level (40.9 per 100,000).
We identified 392 adult beds in Merced, San Joaquin, and Stanislaus Counties. This included 83 available for clinically managed high-intensity residential services (7.0 per 100,000 adults; ASAM level 3.5), 210 available for clinically managed moderate- and low-intensity residential services (17.7 per 100,000 adults; ASAM level 3.3), and 51 available for clinically managed residential detoxification services (4.3 per 100,000 adults; ASAM level 3.2). We also estimated that there are zero child and adolescent SUD treatment beds.
From the facility survey responses, we estimated that Merced, San Joaquin, and Stanislaus Counties require 156 adult acute inpatient psychiatric beds (13.1 per 100,000 adults) and 505 at the subacute level (42.5 per 100,000 adults). At the community residential level, we estimated a need of 354 beds (29.9 per 100,000 adults). From our review of the literature, we estimated a need of 308 adult acute inpatient psychiatric beds (26.0 per 100,000 adults), 291 subacute beds (24.6 per 100,000 adults), and 265 community residential beds (22.3 per 100,000 adults). Although we did not have a sufficient number of responses to estimate child psychiatric bed need, our review of descriptive benchmarks of bed capacity indicated a range from 36 to 41 acute beds (8.1 to 9.2 per 100,000 children and adolescents), 34 to 39 subacute beds (7.7 to 8.7 per 100,000), and 116 to 196 community residential beds per 100,000 (26.3 to 44.3 per 100,000).
From the facility survey responses, we estimated that Merced, San Joaquin, and Stanislaus Counties require between 325 to 362 adult SUD treatment beds (27.4 to 30.5 per 100,000 adults). There was greater demand for higher-intensity services: Beds for ASAM levels 3.5 and 3.3 had higher occupancy rates and more requests for transfers to higher levels of care than requests for transfers to lower levels of care. Our review of the literature showed that reference benchmarks for adult SUD treatment beds were considerably higher: about 507 to 548 beds (42.7 to 46.2 per 100,000 adults). Among children- and adolescent-servicing facilities, we did not receive enough responses to directly estimate SUD treatment bed needs. However, during our review of the literature, we found reference benchmarks of 43 to 69 SUD treatment beds (9.9 to 15.6 per 100,000 children and adolescents).
We estimated that Merced, San Joaquin, and Stanislaus Counties have a shortage of inpatient beds, although the specific level of care experiencing the shortage varied by data source. According to our survey data, there is a shortage of 175 subacute beds and a surplus of 30 acute beds. Likewise, using our expert consensus estimates, we estimated a shortage of 122 acute beds but a surplus of 30 subacute beds. Our results for community residential beds were consistent across sources: We estimated a surplus of between 113 and 202 beds. However, because of beds occupied by clients from outside Merced, San Joaquin, and Stanislaus Counties, these estimates may underestimate the magnitude of the shortage and overestimate the magnitude of the surplus. Using statewide and international reference points for child psychiatric beds, we observed modest shortfalls of acute and subacute beds, while community residential beds ranged from a 15-bed shortfall (using statewide reference points) to a 65-bed surplus (using international reference points).
We estimated that Merced, San Joaquin, and Stanislaus Counties have a shortfall of 156 adult SUD beds. This is 13.2 beds per 100,000 adults shy of California's average rate of 46.2 beds per 100,000 adults. This also represents a shortfall of 115 adult SUD beds when compared with the U.S. national average (9.7 beds per 100,000 adults shy of the national average of 42.7 beds per 100,000 adults). From the facility survey responses, we estimated a modest surplus; however, this direct estimate likely underestimates the magnitude of the shortage because of beds occupied by clients from outside Merced, San Joaquin, and Stanislaus Counties and because some facilities did not accept Medi-Cal patients. Both of these factors could be contributors to our results. Among children and adolescents, we found that the California and national benchmarks indicate a shortfall of 44 to 69 SUD treatment beds (9.9 and 15.6 per 100,000 children and adolescents, respectively).
Overall, we found that the most difficult-to-place populations at psychiatric facilities were individuals with dementia (21 percent of facilities accepted these individuals), nonambulatory individuals (25 percent), and those who required oxygen (25 percent). The most difficult-to-place populations at SUD treatment facilities were nonambulatory individuals (25 percent), individuals with a past sex offense (30 percent), individuals with dementia (30 percent), and individuals with traumatic brain injuries (30 percent).
For adult psychiatric facilities, we found that out-of-region residents consisted of 36 percent of patients in acute facilities, 88 percent of patients in subacute facilities, and 34 percent of patients in community residential facilities. For SUD treatment facilities, the out-of-region estimates were 20 percent of patients in ASAM level 3.5, 18 percent in ASAM level 3.3, and 0 percent in ASAM level 3.2.
Stakeholders reported shortages for all levels of inpatient and residential care, especially for children. Consistent with the quantitative analysis, they reported difficulties placing various populations, especially individuals with co-occurring physical health conditions.
On the basis of these findings, we offer the following recommendations:
In terms of psychiatric beds, focus on addressing the shortage of inpatient beds, especially for hard-to-place populations—including those with dementia and traumatic brain injuries. The majority of psychiatric facilities surveyed in Merced, San Joaquin, and Stanislaus Counties do not accept patients with dementia and traumatic brain injuries. From our survey data, we estimated a shortage of 175 beds at the subacute level and a surplus of 30 beds at the acute level. However, using expert consensus estimates, we estimated a surplus of 39 beds at the subacute level and a shortage of 122 acute beds. The subacute facilities also had a 25 percent waiting list volume and an 88 percent occupancy rate of those who reside outside the region, while acute facilities had a 36 percent occupancy rate from those outside the region. One potential solution to this is to increase the number of beds at these higher levels of care by expanding infrastructure. However, this could lead to an overall surplus—including of unused infrastructure—if even more patients are transferrable from acute or subacute care to community residential care. Merced, San Joaquin, and Stanislaus may therefore want to consider alternatives that allow the county to shift the existing distribution of beds from lower to higher levels of care.
In terms of SUD treatment beds, focus on increasing beds that are available for Merced, San Joaquin, and Stanislaus County residents who are currently hard to place—including nonambulatory individuals. We observed that more than 15 percent of SUD treatment beds are occupied by residents from outside Merced, San Joaquin, and Stanislaus Counties; 20 percent of facilities did not accept patients insured by Medi-Cal; and 80 percent of facilities did not accept patients who were nonambulatory. This indicates that, although these facilities appear to have stable bed occupancy rates and short waiting list volumes, beds at these facilities are not available to many individuals in need of services. This observation is also liable to account for the difference in the estimated need for SUD treatment beds in Merced, San Joaquin, and Stanislaus Counties based on (1) observed outcomes from survey interviews and (2) the reference benchmarks of statewide and national bed capacity. If Merced, San Joaquin, and Stanislaus Counties were to anchor on reference benchmarks (which we believe to be more appropriate), this would indicate a shortage of more than 100 SUD treatment beds for adults and another 40–60 for children and adolescents. Any additional beds would need to be designed to ensure that they reach the high-needs populations that are currently being missed.
Track outcomes of investments in bed capacity over time, including bed occupancy rates, waiting list volume, and bottlenecks that inhibit transfers to higher and lower levels of care. Should the region increase their investments in these facilities, such as by expanding the number of beds, then the counties should monitor how these expansions are associated with changes in outcomes. More-complete data that empirically examine the relationship between changes in bed capacity and outcomes would provide more-valid estimates for future assessments of whether there are shortages in bed capacity and whether expanding the number of beds could alleviate this concern.
The research described in this article was funded by the California Mental Health Services Authority (CalMHSA) and conducted by RAND Health Care.
More in this issueRAND Health Quarterly is produced by the RAND Corporation. ISSN 2162-8254.
Explore RAND Health Quarterly articles on PubMed