Bed Check: Inpatient Psychiatric Care in Three California Counties APRIL 2020 AUTHORS Amanda Lechner, Matthew Niedzwiecki, Megan Dormond, Jasmine Little, and Melissa Azur, Mathematica Contents About the Authors 3Introduction and Background Amanda Lechner, MPP, is a health researcher Availability Analysis Spurred by Bed Reductions at Mathematica. Also with Mathematica Inpatient Psychiatric Beds as Part of a Larger System of Care are Matthew Niedzwiecki, PhD, health researcher; Megan Dormond, MSW, health Data Sources and Methods analyst; Jasmine Little, MA, health analyst, 5Findings: Inpatient Beds and Use and Melissa Azur, PhD, associate director of health. County Differences in Availability and Use Trends Respondents’ Concerns About Inpatient Bed Availability Working at the intersection of strategy, eval­ uation, and practice, Mathematica helps 7Findings: Factors Affecting Inpatient Beds maximize the impact of foundation invest­ and Use ments, ensuring they are evidence-informed Intermediary Care and move the needle on complex challenges. Routine Outpatient Care We offer learning and evaluation services to strengthen program design and execution Emergency and Crisis Services across the strategy lifecycle. For more infor­ Workforce Availability mation, visit www.mathematica.org. County Funding of Mental Health Services 13Key Takeaways for Stakeholders About the Foundation and Policymakers The California Health Care Foundation is dedicated to advancing meaningful, measur­ 15 Appendices able improvements in the way the health care A. Case Studies delivery system provides care to the people of B. Methods California, particularly those with low incomes and those whose needs are not well served 28 Endnotes by the status quo. We work to ensure that people have access to the care they need, when they need it, at a price they can afford. CHCF informs policymakers and industry leaders, invests in ideas and innovations, and connects with changemakers to create a more responsive, patient-centered health care system. For more information, visit www.chcf.org. California Health Care Foundation www.chcf.org 2 Introduction and living in poverty and a high percentage of residents covered by Medi-Cal. San Diego County was selected Background in part because of the urgency generated by the clo­ sure of the Tri-County Medical Center’s psychiatric units. Availability Analysis Spurred by Bed Reductions Case studies of each of the three counties, set out in In June 2018, San Diego’s Tri-City Medical Center Appendix A, summarize qualitative information gath­ announced it would eliminate 30 inpatient psychiatric ered through interviews with key stakeholders and beds — including an 18-bed locked behavioral health quantitative information based on publicly available unit and a 12-person crisis-stabilization unit.1 This data. closure is part of a 25-year reduction in California’s supply of inpatient psychiatric beds, during a time when demand for mental health care services across Inpatient Psychiatric Beds as Part of the state has been increasing. In response to these a Larger System of Care developments, an investigation was conducted on the Assessing the availability of and need for inpatient factors contributing to the availability of and demand beds requires an understanding of the system as a for inpatient psychiatric care in the state. The findings whole. Inpatient services represent one modality in are discussed in this report. a complex, interconnected system of care for mental health conditions. People with mental health needs The research looked in depth at three California coun­ also engage with providers in emergency settings, as ties — San Diego, Fresno, and Contra Costa Counties well as in routine outpatient environments such as pri­ — which represent geographically diverse regions mary care and mental health clinics. Patients may also of the state and vary in terms of key demographic use intensive outpatient settings, such as partial hospi­ characteristics (Table 1). For example, Contra Costa talization or assertive community treatment programs, County has a relatively low share of residents covered which function as intermediary care between inpatient by Medi-Cal and a low share of residents living in pov­ and routine outpatient care (see “Levels of Psychiatric erty compared with the state average. By comparison, Care” sidebar on page 4). Fresno County has a relatively high share of residents Table 1. Demographics of Study Counties, 2017 CONTRA COSTA FRESNO SAN DIEGO COUNTY COUNTY COUNTY CALIFORNIA Population (in millions) 1.15 0.99 3.34 39.54 Residents Living in Poverty 9.3% 21.1% 11.8% 15.1% Medi-Cal Model* Two Plan Two Plan Geographic NA Residents with Medi-Cal 18.4% 42.6% 23.5% 29.3% Uninsured Residents 5.5% 6.5% 6.9% 7.3% *In California, there are six models of Medi-Cal managed care. Under the Managed Care Two-Plan Model, the Department of Health Care Services contracts with two managed care plans in the county (one private plan and one county-owned public plan) to provide medical services to most Medi-Cal enrollees. Under the Geographic Managed Care Model, the Department of Health Care Services contracts with several commercial plans in the county. Sources: U.S. Census Bureau, QuickFacts (Washington, DC: U.S. Government Printing Office, 2018); American FactFinder, “Share of Residents Living in Poverty” (2017), American Community Survey, U.S. Census Bureau’s American Community Survey Office, accessed March 20, 2019; UCLA Center for Health Policy Research, “Share of Residents with Medi-Cal” and “Share of Uninsured Residents” (2017), accessed March 19, 2019. Factors Affecting Inpatient Psychiatric Bed Availability in California: Findings from San Diego, Fresno, and Contra Costa Counties 3 Importantly, when patient throughput –– or flow across Data Sources and Methods the mental health system –– is obstructed by a lack The case studies presented in Appendix A draw on of access at one or more of these different points of information from publicly available secondary data care, the effects may be felt throughout the system. sources and from qualitative interviews. The case For example, a perceived lack of inpatient beds in a studies are also based on analysis of secondary data community might be rooted in a dearth of interme­ using publicly available information from 2010 – 2017 diary options, such as partial hospitalization services, (depending on the data source) to create county-level because providers may not be able to safely discharge measures of the following: patients who no longer need inpatient care but still $ Utilization of outpatient clinics, emergency require intensive support. A perceived lack of inpa­ departments (EDs), and inpatient facilities tient beds might also be the result of a shortage of mental health providers to operate beds rather than $ Supply of hospital psychiatric beds an actual shortage of beds. At the same time, gaps in $ Staffing of outpatient and inpatient psychiatric outpatient care and crisis services within a community facilities may contribute to greater use of the inpatient system, since people’s mental health needs may escalate in $ Mental Health Services Act (MHSA) funding the absence of accessible community-based care. State psychiatric hospitals are excluded from the analy­ The purpose of this research is to provide information ses. The data sources included the Office of Statewide on the availability and use of inpatient psychiatric ser­ Health Planning and Development (OSHPD) (clinic and vices in the context of related factors. ED utilization and financial summaries), the California Mental Health Services Oversight and Accountability Commission, the US Census Bureau’s US Census and Levels of Psychiatric Care American Community Survey, the Health Resources and Services Administration’s Area Health Resources Inpatient. 24-hour care provided in psychiatric hos­ pitals or general hospital psychiatric units. Files, and the UCLA Center for Health Policy Research’s California Health Interview Survey. Appendix B con­ Intermediary. Care that provides more support than tains a full description of the outcomes, as well as data routine outpatient care but that is less intensive than inpatient hospitalization. Includes residential care, sources and years of data included in the analyses. which provides long-term care in settings that are typically more comfortable than hospitals; partial Interviews were conducted with 23 respondents in San hospitalization and day programs, in which indi­ Diego, Fresno, and Contra Costa Counties between viduals regularly receive partial-day mental health April and July 2019. Respondents included executives services for several hours per day; and assertive from hospitals, health clinics, Medi-Cal and commer­ community treatment programs, through which community-based multidisciplinary teams provide cial health plans, and other local mental health care treatment, rehabilitation, recovery, and support ser­ leaders. Interviews with several associations and gov­ vices to individuals with serious mental illness. ernment agencies at the state level also informed this Routine outpatient. Less than 24-hour care report. Interview topics included trends in use and provided in a wide range of care settings, such as availability of mental health care and county-level community mental health centers, private therapy funding for mental health care over the preceding five offices, and primary care clinics. Care is generally years (2014 – 2019). Because the qualitative analyses provided for less than three hours at a single visit. include reflections on trends that extend three years Emergency. Care provided in emergency depart­ beyond the quantitative analysis, some observations ments and crisis intervention and stabilization from interview respondents may be based on devel­ centers. opments that are not reflected in the quantitative data. California Health Care Foundation www.chcf.org 4 Figure 1. A vailable Psychiatric Inpatient Beds Per One Findings: Inpatient Beds Million Persons, by Study County, 2010 – 2016 and Use Contra Costa Fresno San Diego CA This section provides an overview of the state of 300 inpatient psychiatric beds in San Diego, Fresno, and Contra Costa Counties during the study period. 250 246.2 208.0 County Differences in Availability 200 and Use Trends 164.4 From 2010 to 2016, the number of licensed psychiatric 150 166.7 inpatient beds per capita in California remained fairly stable, declining by 7%. As of 2016, the most recent 100 year for which OSHPD data are publicly available, 94.9 there were 170 beds per 1,000,000 people (or about 83.6 94.8 82.6 50 17 for 100,000 people) in the state. In the absence of definitive information on the “right” number of beds per capita, it is useful to compare California’s inpatient 2010 2011 2012 2013 2014 2015 2016 availability to the rest of the country. Analysis from the California Hospital Association indicate that California Figure 2. P sychiatric Inpatient Discharges Per 1,000 has 1 bed for every 5,834 people (or about 170 per Persons, by Study County, 2010 – 2016 1,000,000 people), compared with 1 bed for every 4,383 people (or about 228 per 1,000,000 people) Contra Costa Fresno San Diego CA nationwide.2 9 In the three counties studied, the number of inpatient 8 psychiatric beds per capita moved in different direc­ 7.6 7.3 tions (Figure 1).3 For example, the number of beds per 7 capita decreased in San Diego but increased in both 6 Fresno and Contra Costa. Despite the decline in beds 5.5 5.6 in San Diego, the county still had 209 beds per one 5 4.9 4.4 million persons in 2016, which was substantially more 4 than the other two counties and more than the aver­ 3 age statewide. Fresno and Contra Costa Counties, 2.9 2.5 each with approximately 95 beds per one million per­ 2 sons, continued to have substantially fewer beds per 1 capita compared with the state average. 2010 2011 2012 2013 2014 2015 2016 Use of beds remained fairly stable statewide over the analysis period. The number of admissions to psychi­ FIGURES 1 and 2: atric hospitals, as measured by discharges, decreased Notes: State hospitals are excluded from the analysis. Count of psychiatric inpatient discharges to hospitals in each county. Outcomes are scaled to only slightly across the state as whole and in each 1,000 persons based on county population from Census data. Data from of the three counties (Figure 2). The average length 2017 are not reported because a significant number of hospital reporting periods for FY 2016–2017 (the most recent data) did not include 2017. of stay, or number of days that admitted patients Source: Data are from the Office of Statewide Health Planning and remained in the hospital, was fairly stable statewide Development “Hospital Annual Financial Disclosure Report – Complete Data Set” and “Hospital Annual Financial Data – Selected Data & Pivot Tables” financial databases for fiscal years 2009–2010 through 2016–2017. Factors Affecting Inpatient Psychiatric Bed Availability in California: Findings from San Diego, Fresno, and Contra Costa Counties 5 and in both San Diego and Contra Costa Counties. Respondents’ Concerns About However, Fresno County saw a 38% increase in aver­ Inpatient Bed Availability age length of stay (Figure 3). While stakeholders across counties expressed similar Figure 3. A verage Length of Stay for Inpatients, concerns about the availability of inpatient beds in by Study County, 2010 – 2016 their counties, county trends differed in ways that may suggest the need for county-specific responses: Contra Costa Fresno San Diego CA $ InSan Diego County, the number of beds declined, 9 while the use of beds remained fairly stable. As 8.3 8.0 8.2 a result, more beds were filled more of the time, 8 7.8 7.3 indicating a decrease in available capacity relative 7 6.7 to need. This raised concerns among respondents 6.6 that additional hospital closures in the future could 6 cause substantial strain on the inpatient system. 5 4.8 If that happens, it might suggest a need for addi­ 4 tional inpatient beds. 3 $ InFresno County, both the number and use of beds 2 increased. Respondents in the county expressed concerns that the increased use of inpatient care is 1 outpacing the growth in capacity. The major health system in the county is reportedly adding beds to 2010 2011 2012 2013 2014 2015 2016 help keep up with the need. Notes: State hospitals are excluded from the analysis. Average length of stay as calculated by total inpatient psychiatric census days divided by total discharges. Data from 2017 are not reported because a significant $ In Contra Costa County, the number of beds number of hospital reporting periods for FY 2016–2017 (the most recent increased, while the use of beds decreased slightly, data) did not include 2017. Source: Data are from the Office of Statewide Health Planning and suggesting an increase in available beds relative to Development “Hospital Annual Financial Disclosure Report – Complete the need for them. Nevertheless, respondents in the Data Set” and “Hospital Annual Financial Data – Selected Data & Pivot Tables” financial databases for fiscal years 2009–2010 through 2016–2017. county expressed concern about a lack of available beds, which may reflect observations more recent than the trends captured in the quantitative data. Analysis of more recent quantitative data, available either from the county or when more recent years of OSHPD data are released, could shed insight into respondents’ concerns. California Health Care Foundation www.chcf.org 6 Findings: Factors We don’t have a lot of intensive outpatient Affecting Inpatient Beds or mental health outpatient services. . . . It’s and Use difficult to find intermediate levels of care. This section delves into the different factors identified So it’s not easy to step somebody down by respondents as affecting inpatient beds and use: from inpatient to intensive outpatient or access to intermediary care, access to routine outpa­ tient care, access to emergency and crisis services, partial hospitalization, because that really workforce availability, and how counties use mental doesn’t exist. health funding. — Psychiatrist at an outpatient clinic Intermediary Care Figure 4. P artial Hospitalization Days Per 1,000 Persons, More partial hospitalization and assertive by Study County, 2010 – 2016 community treatment programs are needed across all three counties. Respondents in the three counties emphasized the Contra Costa Fresno San Diego CA need for additional intermediary care, such as partial 60 hospitalization and assertive community treatment programs (see sidebar on page 4 for definitions). Such 50 services are needed for patients exiting acute inpa­ 43.0 tient psychiatric care as well as for individuals who 45.1 40 may not need to be hospitalized but who require more intensive services than can be provided in routine out­ patient visits. 30 22.9 Over the analysis period, the number of partial hos­ 20 18.2 pitalization days per capita remained fairly stable statewide and within each county but varied widely 10 across the three studied counties (Figure 4). For 6.9 6.2 example, San Diego County had many more partial 0.0 0.0 hospitalization days per capita than the state average, 2010 2011 2012 2013 2014 2015 2016 with 45 days per 1,000 persons in 2016, compared Notes: State hospitals are excluded from the analysis. Count of partial with the statewide average of 18 days. Contra Costa hospitalization days at hospitals in each county. Outcomes are scaled to 1,000 persons based on county population from Census data. Data from County had approximately six partial hospitalization 2017 not reported because a significant number of hospital reporting days per 1,000 persons. Of particular note, Fresno periods for FY 2016–2017 (the most recent data) did not include 2017. County had no partial hospitalization services avail­ Source: Data are from the Office of Statewide Health Planning and Development “Hospital Annual Financial Disclosure Report – Complete able during the analysis period. Data Set” and “Hospital Annual Financial Data – Selected Data & Pivot Tables” financial databases for fiscal years 2009–2010 through 2016–2017. While quantitative data on availability of intermediary care is limited to data on partial hospitalization ser­ vices, across all three counties respondents described a need for other types of intermediary care services, including assertive community treatment. Factors Affecting Inpatient Psychiatric Bed Availability in California: Findings from San Diego, Fresno, and Contra Costa Counties 7 Many respondents emphasized that the shortage of Routine Outpatient Care intermediary care has hindered the ability of inpatient Large increases in use in San Diego and Fresno units to discharge patients since they cannot develop Counties did not appear to offset inpatient use. a clinically appropriate plan for referring patients to The use of routine outpatient care varied substantially community providers. The result has been longer across and within counties over the analysis period stays than would otherwise be medically necessary, (Figure 5) but did not appear to have a major impact more frequent readmissions, and reduced inpatient on use of inpatient beds. Across the state, the num­ bed availability for those who need it. All of these ber of outpatient visits to licensed community and free have significant implications for patients with mental clinics for mental health needs (including both visits illnesses and their families. to primary care and psychology specialty clinics) per capita increased by 43%. The number of visits rose The lack of intermediary care is especially acute for substantially in Fresno County (by 62%) and in San people covered through Medi-Cal. Across the three Diego County (by 157%). counties, intensive outpatient services are limited for people on Medi-Cal, and partial hospitalization Figure 5. M ental Health Visits Per 1,000 Persons to services are reportedly nonexistent. Access to inter­ Outpatient Clinics, by Study County, 2010 – 2017 mediary care appears to be complicated by the state’s divided coverage of mild-to-moderate mental health Contra Costa Fresno San Diego CA conditions under Medi-Cal managed care plans and coverage of severe conditions under the county men­ 80 tal health plans. While many types of intermediary care 76.5 70 fall under coverage of specialty mental health services by county mental health plans,4 across counties, some 58.4 60 providers, payers, and other stakeholders expressed 52.7 uncertainty as to whether partial hospitalization ser­ 50 vices are covered by county mental health plans. This 40 confusion may contribute to limited provision of inter­ 34.7 mediary services for people with Medi-Cal coverage. 30 28.5 20 26.0 24.6 13.6 There aren’t enough crisis houses available 10 in the county. The patients end up staying 2010 2011 2012 2013 2014 2015 2016 2017 [longer] in the hospital in the acute-care Notes: Visits to primary care clinics for mental health care need and site. The patients that are coming in to the specialty psychology clinics. Outcomes are scaled to 1,000 persons based on county population from Census data. hospital that need acute treatment aren’t Source: Data are from the Office of Statewide Health Planning and Development “Primary Care Clinic Annual Utilization Data” databases for getting it because those beds are being held the years 2010–2017. up by patients that are waiting for placement into a crisis home. — Behavioral health director at an inpatient hospital California Health Care Foundation www.chcf.org 8 In contrast to statewide trends, the number of visits per I think we are continuing to see increased capita remained fairly stable in Contra Costa County (decreasing, but only slightly). In both Fresno and San challenges with anxiety, stress, substance, Diego Counties, respondents attributed the increase alcohol, drug disorders. . . . We have had to growth in the number of people with Medi-Cal cov­ erage following the state’s 2014 Medi-Cal expansion. volume increases in general, especially The increased use of routine outpatient care in these in the Medi-Cal population for all of our two counties did not appear to offset the need for inpatient care, since neither county saw a meaningful services. I’d probably say outpatient reduction in use of inpatient care; in fact, inpatient use services had the greatest increase. increased in Fresno County during the analysis period. — Health system behavioral health director Across counties respondents described a continued need for more access to outpatient services. Emergency and Crisis Services ED use rose in Contra Costa and San Diego Counties while inpatient use remained stable. FQHC [Federally Qualified Health Center] The number of people seeking care in EDs for men­ tal health needs increased statewide from 2012 primary care providers are screening more through 2017, growing by 12%. While the use of the and more for mental health needs and ED for mental health needs rose in both San Diego and Contra Costa Counties, it remained fairly stable thus referring more patients to mental in Fresno County, decreasing slightly over the analy­ health providers than ever before, which sis period (Figure 6, page 10). The increased use of is affecting the wait times and frequency EDs in San Diego and Contra Costa Counties was not associated with a larger number of admissions to the of appointments. This isn’t necessarily a hospital, as the numbers of admissions in those coun­ bad thing. We are working on meeting ties remained fairly stable over the analysis period. In San Diego County, a couple of respondents attributed the demand. the rise in ED use to problems accessing outpatient — Director of an FQHC care and to increased use of methamphetamine among county residents. The county has added emer­ gency and crisis services to alleviate the pressures on EDs, but respondents emphasized that the commu­ nity still needs additional services. In Contra Costa County, people may be using the ED as a source of outpatient treatment, possibly because of access chal­ lenges in certain areas of the county or wait times for outpatient care. Factors Affecting Inpatient Psychiatric Bed Availability in California: Findings from San Diego, Fresno, and Contra Costa Counties 9 Figure 6. E mergency Department Visits for Mental Health Diagnoses Per 1,000 Persons, by Study County, Workforce Availability 2012 – 2017 Shortages of mental health care providers creating problems across care settings. Contra Costa Fresno San Diego CA Many California counties have a shortage of mental health providers, and respondents highlighted work­ 20 force shortages as substantial problems impacting 18 access to mental health care across inpatient and 16.5 16.9 16 16.1 outpatient settings. During the analysis period, the 15.4 14.1 15.3 number of psychiatrists per capita remained fairly sta­ 14 13.8 ble within the three counties and the state as a whole. 13.6 12 However, the numbers varied widely across counties, 10 ranging from 99 psychiatrists per one million persons in Fresno County to 156 in San Diego County. 8 6 Respondents in all three counties described a need for 4 more psychiatrists to serve county residents, though respondents in Fresno County emphasized the need 2 as especially severe. In addition to the shortage of psychiatrists, there is a need for other types of men­ 2012 2013 2014 2015 2016 2017 tal health clinicians — such as psychologists, clinical Notes: Outcomes are scaled to 1,000 persons based on county popula- social workers, and marriage and family therapists — tion from Census data. Data report total emergency department visits (discharged outpatient and admitted) that have an associated diagnosis particularly in Fresno County and other counties in the related to mental health (ICD-9-CM codes 290–319, “psychoses and neuro- sis,” and ICD-10-CM codes F01–F99, “mental disorders”). San Joaquin Valley.5 All three study counties are pur­ Source: Data are from the Office of Statewide Health Planning and suing a range of strategies to bolster the workforce, Development “Hospital Emergency Department – Characteristics by including ramping up efforts to recruit more psychia­ Facility (Pivot Profile)” databases for the years 2012–2017. trists and using telehealth where there are not enough local providers to meet the mental health care needs of the population. The biggest challenge really lies within psychiatry. We indeed have had challenges, in terms of recruiting, to meet our population’s needs. — Health system behavioral health director California Health Care Foundation www.chcf.org 10 Medi-Cal enrollees with severe mental health service County Funding of Mental needs and California residents with low income. Health Services Counties used Mental Health Services Act Individual counties have substantial discretion in (MHSA) funds in different ways to alleviate how they allocate their resources to meet the spe- pressure on inpatient beds. cific needs for these mental health services in their California has a complex funding structure for public counties (Figure 7). This report presents quantitative mental health services that includes local realignment analyses of MHSA expenditures and qualitative results revenues, MHSA funds (Proposition 63), state general of perceptions of broader mental health services fund- funds, and federal funds.6 These resources fund County ing. MHSA funds composed approximately 22% of Mental Health Services (CMHS) for care provided to county behavioral health funding during California’s 2019–2020 budget year.7 Figure 7. Public Funding for Mental Health Services Local Mental Health Federal Funds State Realignment Services Fund (primarily through General Fund Revenues (Prop 63) Medi-Cal) County Mental Health Services Medi-Cal Managed Care TARGET POPULATION   Adults and children enrolled and Fee-for-Service in Medi-Cal who meet criteria* for specialty mental TARGET POPULATION   People health services; residents with low incomes enrolled in Medi-Cal whose mental health needs do not meet criteria for Medi-Cal Specialty Services include: specialty mental health services $ Care coordination and case management Services include: $ Therapy $ Psychotropic drugs $ Day treatment and rehabilitation $ Outpatient $ Crisis intervention and stabilization $ Psychiatric inpatient $ Psychiatric inpatient services $ Residential treatment Mental Health Services Act Programs $ Intensive and wraparound services $ Prevention and early intervention activities $ Innovation projects Safety-Net Services $ Mental health services for residents with low incomes, including the uninsured *Criteria for children under 21 are less restrictive than those for adults. Source: Figure is adapted from two sources: (1) Deborah Reidy Kelch, Locally Sourced: The Crucial Role of Counties in the Health of Californians (Sacramento: California Health Care Foundation, October 2015); (2) Legislative Analyst’s Office, Overview of Funding for Medi-Cal Mental Health Services (PDF) presented to Assembly Committee on Health and Senate Committee on Health, February 26, 2019. Factors Affecting Inpatient Psychiatric Bed Availability in California: Findings from San Diego, Fresno, and Contra Costa Counties 11 During the analysis period, MHSA expenditures var­ Figure 8. M HSA Spending Per Capita, by Study County, ied across study counties. For example, in 2017, 2013– 2017 MHSA spending ranged from $33 per capita in Contra Costa County to $45 per capita in San Diego County Contra Costa Fresno San Diego CA (Figure 8). To alleviate pressures on inpatient systems $50 and to provide care for county residents with mental health conditions, study counties are using available $45 $44.73 MHSA funding in a variety of ways: $40 $40.24 $37.52 $39.49 $35 $34.77 $ San Diego County reported using MHSA dollars to $33.31 fund crisis services, assertive community treatment $30 programs, housing programs for people with mental $25 $25.07 illness, and psychiatry residency training programs. $23.29 $20 The county has also been allocating MHSA funds for housing and homelessness programs, including $15 permanent supportive housing for individuals with $10 serious mental illness and temporary rental assis­ $5 tance housing support. 2013 2014 2015 2016 2017 $ Fresno County reported using MHSA dollars to Notes: Spending does not include other funding outside of the Mental fund several full-service partnerships with commu­ Health Services Act (MHSA). nity organizations that provide an array of services Source: Data are from the California Mental Health Services Oversight and for people with serious mental illness or emotional Accountability Commission for fiscal years 2012–2013 through 2016–2017. disorders. Services include individual and group therapy, medication, case management, and hous­ ing support. Full-service partnership programs also reserve some slots specifically for individuals who are being discharged from inpatient settings. $ Contra Costa County reported using MHSA funds to operate and contract with mental health ser­ vice providers to support full-service partnerships in order to provide clients with the full spectrum of mental health and social services. MHSA funds have been used to add a new mobile crisis response team to serve adults and to expand the hours dur­ ing which the child mobile crisis response team is able to operate. California Health Care Foundation www.chcf.org 12 Limited access to intermediary services Key Takeaways for contributes to use of the inpatient system. Stakeholders and While it is difficult to quantify the extent to which expanded intermediary care could offset inpatient Policymakers use, many respondents reported that enhanced inter­ mediary care would reduce inpatient admissions and lengths of stay. They also said that intermediary care Expansion of intermediary, routine could more appropriately serve the needs of many outpatient, and crisis services people with mental health conditions. should be the priority. Across the three studied counties, respondents Respondents emphasized that people with Medi-Cal described high need for mental health services in inpa­ coverage, in particular, lack sufficient access to interme­ tient and other care settings and insufficient resources diary care. As one partial solution, Federally Qualified to serve the need. Although the trends in availability Health Centers (FQHCs) in San Diego County are and use of inpatient care differed, stakeholders in all partnering with hospitals to care for patients recently three counties expressed concerns about bed avail­ discharged from inpatient psychiatric units. The large ability. In Fresno County, use of beds increased and increase in utilization of FQHCs and other community appears to be outpacing recently added capacity. In clinics in San Diego from 2010 to 2016 reflects this San Diego County, use of beds remained relatively growing partnership to meet the needs of people stable as bed capacity declined — raising concerns leaving inpatient care, especially those covered by about the potential for shortages in the future. In Medi-Cal. However, although FQHCs may be able to Contra Costa County, respondents’ perceptions of the provide some intermediary care, it may not be feasible need for more bed availability may reflect changes for them to offer more intensive intermediary services, that have occurred more recently than publicly avail­ such as partial hospitalization programs. able data show. Within counties, payers and providers could consider Across counties, many respondents emphasized that developing collaborative strategies for increasing the need for inpatient beds is contingent on the avail­ availability of partial hospitalization services or other ability of other levels of mental health care; to the intensive intermediary care, especially for people cov­ degree that community-based care is available, the ered through Medi-Cal. need for inpatient care may decrease. For this rea­ son, many respondents prioritized expansion of other Use of outpatient clinics increased substantially levels of care, including intermediary care, routine out­ statewide and in Fresno and San Diego patient care, and crisis services. Counties over the analysis period; however, challenges in accessing outpatient services Several related themes emerged from this study for persist. policymakers, providers, and mental health care stake­ Respondents attributed the large increase in volume holders to consider. of people seeking outpatient mental health care during the analysis period to a rise in the number of people with health care coverage following the 2014 Medi-Cal expansion. Despite the large increase in the volume of outpatient visits to licensed community and free clinics for mental health needs, many respondents emphasized that access to outpatient care remains challenging for certain populations, especially those Factors Affecting Inpatient Psychiatric Bed Availability in California: Findings from San Diego, Fresno, and Contra Costa Counties 13 with Medi-Cal coverage, and in certain geographic Provider shortages, particularly in psychiatry, areas within the counties. While it is unclear whether are a major challenge that limits expansion of expanded access to routine care would alleviate pres­ mental health services. sures on the inpatient system, addressing the need Respondents in all three counties described a need for for outpatient care is nonetheless important. Counties more psychiatrists to serve county residents, as well as are pursuing a range of strategies to increase access a less dire need for other types of clinicians, including to outpatient care, including relying on FQHCs to psychologists, social workers, and marriage and fam­ provide some mental health services. Going forward, ily therapists. The magnitude of provider shortages is FQHCs may be able to expand their roles as providers especially acute in Fresno County and neighboring of outpatient mental health care, but they may need counties, although provider shortages appear to be more support to do so. present across counties. Each of the studied counties is working to expand its mental health workforce, but Crisis services are important for serving the some respondents suggested that state-level initia­ needs of county residents and for alleviating use tives to increase the supply of mental health providers of EDs for mental health needs. may also be needed. State policies intended to bol­ The number of people seeking care in EDs for men­ ster the mental health workforce should consider the tal health needs increased across the state and in San stark disparities in provider supply in different areas Diego and Contra Costa Counties during the analy­ of the state when deciding how to allocate resources. sis period. Some respondents in San Diego County reported increased use of methamphetamine as a contributing factor. Given the volume of people expe­ riencing mental health crises, respondents across counties described the importance of crisis services, such as mobile crisis response teams and crisis stabi­ lization units, in serving their communities. While all three study counties have expanded crisis resources over the last few years, respondents still described high volumes of people seeking care in EDs. Further expansion of crisis services may provide an alternative to better serve people experiencing mental health emergencies. California Health Care Foundation www.chcf.org 14 Appendix A. Case Studies SAN DIEGO COUNTY Hospital, located in the northern portion of the county, Decline in Beds Leads to Concern closed its 18-bed psychiatric unit, and other hospitals About Capacity are reportedly reducing bed availability temporar­ San Diego County saw a decline in the number of ily to make the renovations required for compliance inpatient psychiatric beds from 2010 to 2016 but with the state’s seismic standards and federal ligature relatively stable admissions and lengths of stay. As a requirements.8, 9 Other hospitals may close units if they result, inpatient facilities have been operating at or determine these requirements are too costly to meet. near full capacity most of the time. Some respondents expressed concern that additional bed closures would While the number of beds declined, the volume of result in substantial capacity constraints, as some admissions and the average length of stay remained parts of the county are reportedly on the brink of bed fairly stable from 2010 to 2016, and therefore, more shortages. Respondents pointed to a need for more inpatient beds were occupied more of the time. For intermediary services to facilitate discharges once example, utilization of psychiatric beds increased by patients are stable, to alleviate pressures on the inpa- 17%, with the average bed filled 78.5% of the time tient system. in 2016, up from 66.9% in 2010.10 Also, although the volume of admissions remained fairly stable from 2010 The county has seen an increase in use of emergency to 2016 — decreasing by about 5% — respondents services and a very large increase in use of outpatient from major hospitals in the county said that their inpa­ services for mental health needs. These may be driven tient units have been operating at full capacity. Two by the increase in Medi-Cal enrollment following the respondents observed that volumes have increased state’s 2014 Medi-Cal expansion and by an increase over the past few years, possibly reflecting trends that in mental health needs among county residents. have been occurring more recently than quantitative Respondents emphasized a need for more outpatient data show. services to prevent mental health crises and to serve people already in crisis. To address these needs, health Table A1. Inpatient Psychiatric Bed Capacity and systems and hospitals are partnering with Federally Utilization, San Diego County, 2010 and 2016 Qualified Health Centers (FQHCs) to provide inten- 2010 2016 CHANGE sive outpatient programs. The county has also been adding assertive community treatment programs and Licensed Psychiatric 249.4 209.2 –16% Inpatient Beds expanding crisis services. (per one million persons) Percentage of Licensed Bed 66.9% 78.5% +17% Decrease in Beds and Stable Utilization Days Filled In line with statewide trends from 2010 to 2016, the Psychiatric Discharges 7.6 7.3 –5% number of psychiatric beds in San Diego County (per 1,000 persons) declined 16% (Table A1). As of 2016, San Diego had Average Length of Stay (days) 8.0 8.3 +3% 12 hospitals with a total of 694 licensed psychiat­ ric beds. Yet the county still had substantially more licensed inpatient psychiatric beds per capita than With the decline in beds in recent years, several California as a whole (209.2 beds per one million pop­ respondents expressed concerns about bed short­ ulation in 2016 versus 170.1 statewide). According to ages, noting that patients can wait up to a few days interview respondents, since 2016, there have been in EDs for inpatient beds to become available. Bed both additions and closures of beds. For example, shortages in the northern portion of the county were interviewees reported that some hospitals in the of particular concern because of recent and antici­ county have added bed capacity. However, Tri-City pated closures there. Factors Affecting Inpatient Psychiatric Bed Availability in California: Findings from San Diego, Fresno, and Contra Costa Counties 15 Concerns About Long-Term Care To help fill the gap in intermediary care, hospitals and Bed Availability FQHCs are partnering to provide post-hospitalization Several respondents expressed concern about a need outpatient services. Family Health Centers, the largest for additional long-term care beds, including more FQHC organization in the county, with eight mental skilled nursing facilities to serve individuals with seri­ health clinics, partners with 10 hospitals to provide ous mental illness. They reported that the lack of care for patients who are coming out of the hospital. long-term beds contributes to pressures on the acute The FQHC created an electronic system to coordinate inpatient system, because some individuals remain in discharge planning with the hospitals and to schedule acute care beds for several months or even years. The outpatient appointments with patients following dis­ county is considering adding more long-term beds to charge. Medication management and individual and help address the need for this level of care. group therapy are provided for children, adolescents, and adults. In addition, there may be one-to-two-year wait times for placements in state psychiatric hospitals, which, in Services Needed to Prevent Crises and Care for California, provide mental health services to individu­ Those in Crisis als referred by a prison, parole board, or county court. There has been a dramatic increase in use of outpa­ In 2017, almost 90% of patients admitted to these tient mental health care in San Diego County over hospitals were forensic commitments.11 the past few years, and many respondents mentioned being overwhelmed by the volume and severity of Bottlenecks Created by Gaps in needs. The number of visits to outpatient primary care Intermediary Care and psychiatric clinics was two and half times higher in Insufficient availability of intermediary care impacts 2016 than in 2010 (Table A2). inpatient bed availability in San Diego County. Several respondents explained that there are not enough Table A2. M ental Health Visits at Community and intermediary outpatient programs for patients being Outpatient Mental Health Clinics, discharged from the hospital and therefore patients San Diego County, 2010 and 2016 are held in inpatient beds longer than clinically nec­ 2010 2016 CHANGE essary. Wait times for intermediary care programs Total Clinic Visits for 26.0 67.0 +157% are typically several days. This delay contributes to Mental Health Diagnoses high volumes of “administrative days,” when patients (per 1,000 persons) remain in the hospital after they are fully stabilized. Visits to Outpatient Clinics 24.7 65.9 +167% This problem occurs across the county. for Mental Health Services (per 1,000 persons, primary care) For people with Medi-Cal coverage and for those Visits to Psychiatric 1.3 1.0 –24% Outpatient Clinics who are homeless, the lack of intermediary care is a (per 1,000 persons) particular problem. For example, partial hospitaliza­ tion programs are unavailable for Medi-Cal enrollees, although the county does provide some other interme­ Two respondents attributed the large increase to diary services such as assertive community treatment. growth in the number of people with Medi-Cal cov­ Two respondents explained that there are not enough erage seeking mental health care following coverage supportive housing programs for homeless individu­ expansions under the Affordable Care Act and to an als with serious mental illness, which leads to longer overall increase in the prevalence of mental health stays in inpatient beds as well as frequent readmis­ conditions. FQHCs in the county, some of which have sions among this population. long provided integrated behavioral health care,12 California Health Care Foundation www.chcf.org 16 continue to offer a large share of the county’s outpa­ The county behavioral health agency and hospitals tient mental health care. For example, Family Health have expanded emergency and crisis services to allevi­ Centers provides more than 2,300 mental health visits ate the pressures on EDs, but respondents emphasized per week and continues to expand the volume of ser­ that the community still needs additional services. For vices it provides each year. Also, a major health system example, over the past several years, the county has in the county has reportedly been expanding outpa­ invested in creating additional Psychiatric Emergency tient services as a result of the increased demand. Response Teams (PERTs), which have grown from approximately 50 to 75 teams. These teams consist Respondents emphasized the need for more outpa­ of licensed mental health clinicians and uniformed law tient care to prevent crises, especially for the Medi-Cal enforcement officers who work together to provide population and in certain regions of the county. For emergency assessment and referral for individuals example, wait times for outpatient appointments for experiencing behavioral health crises. San Diego also the Medi-Cal population have been increasing, and has seven Short Term Acute Residential Treatment some attribute use of emergency services to a lack (START) programs for adult crisis stabilization. In 2014, of timely access to outpatient care. While the north­ Rady Children’s Hospital opened a crisis stabilization ern and central regions of the county have dedicated unit for children and adolescents, and more recently, walk-in centers for behavioral health, helping to fill a the county expanded its youth emergency screening gap in care and prevent unnecessary ED use, other unit from 4 to 12 beds. portions of the county lack these services. Despite this added capacity, respondents said there is Use of the ED for mental health needs in San Diego a need for more services and that crisis beds are often County increased from 2012 to 2017 (Table A3). full. A few hospitals are also considering or in the pro­ Consistent with observations reported by respon­ cess of expanding their crisis stabilization units, and dents, the prevalence of ED use for mental health Rady Children’s plans to open the county’s first pedi­ needs grew by 18% in the county, compared with atric psychiatric emergency department later this year. 12% statewide. Two respondents attributed the rise in These developments may help address the reported ED use to problems accessing outpatient care and to gaps. increased use of methamphetamine in the county. Despite Strength of Workforce, Some Concern Table A3. E mergency Services Utilization for Mental About Shortages Health Diagnosis, San Diego County, According to a recent workforce study, the San Diego Selected Years region, which encompasses both San Diego County 2010 2016 CHANGE and adjacent Imperial County, had ratios of psycholo­ gists, marriage and family therapists, and clinical social Visits to Psychiatric EDs 4.5 7.2 +58% (per 1,000 persons) workers comparable to the state average in 2016.13 San Diego County has more psychiatrists per capita 2012 2017 CHANGE than other parts of the state — 156.2 psychiatrists Total ED Visits for 13.6 16.1 +18% per one million persons in 2016, which is substantially Mental Health Diagnosis (per 1,000 persons) higher than the numbers in Contra Costa and Fresno Counties; it is also higher than the state average Admitted ED Visits for 2.7 2.9 +6% Mental Health Diagnosis (Table A4, page 18). Still, respondents expressed con­ (per 1,000 persons) cern about an inadequate supply of clinicians to serve ED Discharges to 1.9 2.6 +36% county residents, especially people with Medi-Cal. Psychiatric Care (per 1,000 persons) Factors Affecting Inpatient Psychiatric Bed Availability in California: Findings from San Diego, Fresno, and Contra Costa Counties 17 Table A4. M ental Health Staff, San Diego County, Conclusion 2010 and 2016 San Diego County has more inpatient psychiatric beds per capita than the California average, but the num­ 2010 2016 CHANGE ber of beds decreased over the analysis period. While Active Psychiatric Hospital 165.0 209.2 +27% use of inpatient care has remained fairly stable, the and Non-hospital Staff (per million persons) decline in beds has raised concerns about capacity, especially in areas where more hospital closures are Psychiatrists 155.9 156.2 0% (per million persons) expected. Lack of intermediary and long-term care in the county creates pressure on the inpatient system, because patients may be held in acute inpatient set­ Despite the county’s relatively strong workforce, tings longer than they would be if intermediary levels respondents universally commented on shortages as a of care were available. While San Diego has a rela­ problem. While the county has more psychiatrists than tively strong mental health workforce compared with the state average, and the number of psychiatrists in other parts of the state, the county has seen a large the county remained stable from 2010 to 2016, most increase in use of outpatient care, which respondents respondents described provider shortages as most described as outpacing their capacity. The county acute in psychiatry. has been working to increase crisis and emergency services, but respondents reported a need for more Also, some respondents noted that across provider capacity in outpatient settings and more resources to types, the number who are willing to serve people serve people in crisis. enrolled in Medi-Cal is very small because of low reimbursement rates. To help expand the available workforce, San Diego County has partnered with the University of California, San Diego, to place three psychiatry fellows and two nurse practitioners in com­ munity settings. However, respondents cited a need for additional funding to create more residency and workforce training programs and particularly for a larger pipeline of clinicians willing to serve people enrolled in Medi-Cal. MHSA Funding MHSA funding is higher per capita in San Diego than in Contra Costa and Fresno Counties, and above the California average. The county has used MHSA dollars to fund crisis services, assertive community treatment programs, housing programs for people with mental illness, and psychiatry residency training programs. The county also allocates MHSA funds for housing and homelessness programs, including permanent supportive housing for individuals with serious mental illness and temporary rental assistance housing sup­ port. The San Diego County Department of Health and Human Services is conducting a study on cur­ rent needs in the county’s behavioral health delivery system. California Health Care Foundation www.chcf.org 18 FRESNO COUNTY Despite the addition of beds, use of inpatient psy­ Increased use of services outpaces growth chiatric services increased at a faster rate, placing in capacity pressure on the inpatient system. Over the analysis Fresno County has ongoing capacity issues with the period, use of beds rose by 8%, with the average inpatient psychiatric system. While the county has bed filled 84.1% of the time in 2016, up from 77.8% expanded the number of psychiatric beds over the in 2010. The greater use of beds was driven by an past few years, an increase in use of beds is outpacing increase in the number of days that individuals stayed the newly added capacity. Admissions and ED visits in the hospital rather than by growth in the number of for mental health needs have remained fairly stable, people being admitted. For example, average length but higher lengths of stay have placed additional pres- of stay increased by 38%, with individuals staying in sure on bed availability. the hospital 6.6 days on average in 2016, up from 4.8 days in 2010. The number of admissions to the hospi­ The relatively stable use of the ED for mental health tal remained stable. needs could be related to the addition of alternative crisis services and other community-based care in the Table A5. Inpatient Bed Capacity and Utilization, Fresno county over the past few years. However, respondents County, 2010 and 2016 emphasized the need for more intermediary services 2010 2016 CHANGE for individuals released from the hospital. The county is expanding access points for outpatient care as well Licensed Psychiatric 82.6 94.9 +15% Inpatient Beds as adding inpatient beds. (per million persons) Percentage of Licensed Bed 77.8% 84.1% +8% Chronic workforce shortages in the county are a chal- Days Filled lenge in addressing mental health needs, and the Psychiatric Discharges 4.9 4.4 –10% county is working to increase the workforce by invest- (per 1,000 persons) ing resources into recruiting more psychiatrists and Average Length of Stay (days) 4.8 6.6 +38% using telehealth. Growth in Number and Use of Inpatient Beds Respondents pointed to a need for additional beds While the number of inpatient psychiatric beds per overall and for certain populations in particular. For capita decreased statewide from 2010 to 2016, the example, one hospital executive explained that it is number of beds per capita in Fresno County increased very common for people to have to wait in the ED for a from 82.6 to 94.9 per one million persons (Table A5). psychiatric bed to become available; these wait times As of 2016, the most recent year for which data are are approximately two or three times those of patients publicly available, the county had three inpatient with medical needs. Some respondents highlighted psychiatric facilities with a total of 93 licensed beds. the need for more beds specifically for adolescents, Despite the addition of new beds, the number of although others perceived that there are enough in beds per capita in 2016 remained substantially lower the county with the recent addition of an adolescent in Fresno County than in the state as a whole. Since inpatient crisis unit. then, Fresno has added several beds to existing facilities and opened additional facilities, including a A need for beds for people with both medical and mental health rehabilitation center, a crisis residential psychiatric conditions was also highlighted as a gap. unit, and an inpatient crisis unit for adolescents.14 To help address the need, the county’s dominant health system is in the process of adding another 12 to 24 beds to its inpatient behavioral health facility Factors Affecting Inpatient Psychiatric Bed Availability in California: Findings from San Diego, Fresno, and Contra Costa Counties 19 over the next year or two. Respondents were uncer­ settings for both adults and adolescents. These efforts tain whether these additional beds will be enough to have helped with care transitions from inpatient to meet the demand; they pointed out that quantifying intermediary settings. the need for beds is challenging because it depends on the availability of other services such as intermedi­ Stable Use of ED Services ary care and crisis services. Use of the ED for mental health needs in Fresno County remained fairly stable from 2012 to 2017, and Need for More Intermediary Care to Reduce the share of people being admitted from the ED to Inpatient Use facilities in the county with inpatient psychiatric beds Almost all respondents expressed a need for more also remained fairly stable — declining, but only intermediary services for patients being discharged slightly (Table A6). The county has expanded crisis ser­ from inpatient settings and to serve people who vices over the past few years, adding beds to its crisis require intensive services but not necessarily hos­ stabilization units for both adults and children and pitalization. No partial hospitalization services were pairing clinicians with police officers who are respond­ available in the county during the study period, a gap ing to crises in the community. These newly added that appears to place pressure on the inpatient psy­ crisis services may be helping to divert use of EDs. An chiatric system. executive from a major hospital system felt that the county’s crisis stabilization services have been some­ The lack of intermediary care strains the inpatient what helpful in decreasing the number of patients in system because patients occupy beds longer than its ED. needed or when other settings could better serve their needs. Individuals are sometimes inappropriately Table A6. E mergency Services Utilization for Mental hospitalized because they require more intensive Health Diagnosis, Fresno County, Selected Years care than can be provided in routine outpatient vis­ 2010 2016 CHANGE its. Several respondents noted that patients can be discharged when they no longer meet the clinical Visits to Psychiatric EDs 0* 0* NA (per 1,000 persons) criteria for hospitalization, but they are not always connected to appropriate step-down services to meet 2012 2017 CHANGE their needs. In some cases patients are provided with Total ED Visits for 16.5 15.3 –7% instructions or connections to step-down services but Mental Health Diagnosis (per 1,000 persons) may be unable to follow through on them. As a result, patients can be readmitted to the hospital soon after Admitted ED Visits for 1.9 0.9 –54% Mental Health Diagnosis discharge. (per 1,000 persons) ED Discharges to 3.7 2.6 –30% A few respondents pointed to a need for more sup­ Psychiatric Care (per 1,000 persons) portive housing for individuals with serious mental illness who are homeless, both to alleviate use of inpa­ *Fresno County does not have a psychiatric emergency department. tient beds and to more appropriately serve the needs of that population. Large Growth in Use of Outpatient Care To address the need for more intermediary care, the The county saw a large increase in the use of outpa­ county behavioral health department has expanded tient services during the analysis period, although services over the last few years, including adding a utilization per capita remained below the state aver­ mental health rehabilitation center and a new contract age. For example, the number of visits to outpatient for residential care. The county also has teams in place clinics for mental health diagnoses grew by 62%, ris­ to help coordinate discharge planning from inpatient ing from 28.5 per 1,000 persons in 2011 to 46.1 in California Health Care Foundation www.chcf.org 20 2016 (Table A7). This increase and the overall rate are million persons, versus 140.5 statewide (Table A8). The in line with statewide trends. Consistent with trends San Joaquin Valley region as a whole has some of the reflected in quantitative data, respondents reported lowest ratios in California of other behavioral health seeing higher volumes for outpatient mental health professionals, including psychologists, marriage services over the past few years. Some respondents and family therapists, and clinical social workers.15 attributed the increased use to 2014 coverage expan­ Interviewees noted workforce gaps that are especially sions under the Affordable Care Act and “pent-up” severe, including the shortage of psychiatrists, mental demand for mental health care. health specialists for children, and Spanish-speaking providers. Table A7. M ental Health Visits at Community and Outpatient Mental Health Clinics, Fresno One respondent said that the number of psychia­ County, 2010 and 2016 trists employed by the county declined over the past 2010 2016 CHANGE 15 to 20 years as the result of several factors, includ­ ing psychiatrists relocating away from the county, Total Clinic Visits for 28.5 46.1 +62% Mental Health Diagnoses and the building of prisons around the Fresno area, (per 1,000 persons) which may compensate psychiatrists at a higher salary. Visits to Outpatient Clinics 14.1 30.3 +114% Respondents explained that the supply of licensed for Mental Health Services marriage and family therapists, licensed clinical social (per 1,000 persons, primary care) workers, case managers, community mental health Visits to Psychiatric 14.3 15.8 +10% specialists, and peer specialists is somewhat better, Outpatient Clinics (per 1,000 persons) although more of these types of providers are also needed in the county. While use of outpatient care has increased in the Table A8. M ental Health Staff, Fresno County, county, some access challenges persist, particularly 2010 and 2016 in the outlying areas where the geographic distance to providers is large. This creates a barrier to care, 2010 2016 CHANGE particularly for people with low incomes who lack trans­ Active Psychiatric Hospital 149.1 89.8 –40% and Non-hospital Staff portation. The county behavioral health department is (per million persons) pursuing a wide range of strategies to expand access Psychiatrists 91.2 99.0 +9% to outpatient services, including placing clinics in (per million persons) remote areas and adding access points for outpatient services in school settings, faith-based organizations, and libraries. Respondents also described an Urgent Fresno County has been pursuing several strategies Care Wellness Center in the county as an important to increase the available workforce, including using resource that provides mental health screenings and social workers and tele-psychiatry to serve EDs and assessments and expedites access to community pro­ some inpatient units and using locums — physicians viders across the continuum of care. serving on temporary assignments that last between three and six months. The county is working to recruit Workforce Shortages Contributing to more psychiatrists to the area, which is challenging Capacity Constraints because of the region’s generally poor payer mix and Like other counties in the San Joaquin Valley, Fresno certain quality-of-life factors. Several respondents County continues to experience a major shortage of described the limited cultural offerings and less desir­ mental health providers. In 2016, the county had a able weather in Fresno County relative to the Bay Area much lower number of psychiatrists per capita than and Southern California as major recruitment barriers the California average, with 99.0 psychiatrists per one without obvious solutions. Factors Affecting Inpatient Psychiatric Bed Availability in California: Findings from San Diego, Fresno, and Contra Costa Counties 21 MHSA Funding of Full-Service Partnerships for CONTRA COSTA COUNTY Community-Based Services Bed Capacity Generally Adequate but Some MHSA spending is higher per capita in Fresno County Concerns About Outpatient Access than in Contra Costa County but lower than in San Contra Costa County saw a 13% increase in the num- Diego County. Fresno uses MHSA dollars to fund ber of inpatient psychiatric beds per capita from 2010 several full-service partnerships with community to 2016. Despite this increase, respondents said more organizations that provide an array of services for inpatient beds are needed. Interviewees also empha- people with serious mental illness or emotional dis­ sized that more intermediary services are needed for orders; these include individual and group therapy, people exiting the inpatient system, especially for peo- medication, case management, and housing sup­ ple enrolled in Medi-Cal. Over the analysis period, the port. Full-service partnership programs also reserve number of overall visits to outpatient clinics decreased some slots specifically for individuals who are being slightly, and more outpatient care was provided in pri- discharged from inpatient settings, though several mary care settings. At the same time, use of EDs for respondents emphasized the need for additional mental health needs rose substantially, though this did capacity for such individuals. not lead to an increase in psychiatric admissions. The findings may indicate that individuals are using the ED Conclusion as a result of challenges accessing care in outpatient Despite additions of beds during the analysis period, settings. Like other parts of California, Contra Costa increased utilization, driven by longer lengths of stay, County is experiencing a shortage of mental health has placed added pressure on the inpatient system providers. The county is implementing telepsychiatry in Fresno County. The major hospital system in the and working to recruit additional providers to expand county is planning to add more inpatient beds, but its workforce. it remains to be seen whether these will be enough to address needs. Use of the ED for mental health Increase in Beds but Concerns About Capacity needs has remained fairly stable, possibly because the for Select Populations county has added more community-based resources The number of inpatient psychiatric beds per cap­ and access points. Many respondents speculated ita in Contra Costa County increased 13% between that the availability of intermediary services to dis­ 2010 and 2016, growing from 83.6 to 94.8 per one charge patients from the hospital could alleviate some million persons (Table A9, page 23). As of 2016, the pressure on the inpatient bed capacity. Workforce most recent year for which data are publicly available, shortages, which are worse in Fresno than in many the county had two hospitals that provided inpatient parts of California, remain a fundamental problem psychiatric services with a total of 108 beds. As the across the continuum of care. number of beds increased, utilization, as measured by the percentage of days throughout the year that beds are filled, decreased from approximately 70% in 2010 to 48% in 2016. The number of discharges for mental health conditions also decreased over this time period, while the length of stay remained fairly stable. Taken together, these findings suggest that, overall, the county may have sufficient inpatient beds to meet the needs of residents. However, despite the appearance of sufficient bed capacity reflected in quantitative data, a few respondents commented that Contra Costa County needs more inpatient beds. Two respondents cited California Health Care Foundation www.chcf.org 22 substantial wait times in EDs for patients in need of also mentioned that there are typically wait lists for inpatient psychiatric beds. It is possible that since crisis residential beds as well as for board-and-care 2016, inpatient bed use has increased and created facilities, where individuals can receive less inten­ new pressures on capacity. Also, while the availability sive services for up to 18 months in smaller, privately of inpatient psychiatric beds grew overall in the county owned houses. from 2010 to 2016, the number of beds (per capita) for children and adolescents decreased by 8%. A Outpatient Care Access Challenges and respondent from a major health system in the county Increased ED Use noted that use of inpatient care among the adolescent Contra Costa County saw a slight decrease in the population has increased in that system over the past overall use of outpatient services from 2010 to 2016 few years. A couple of respondents said that finding and a shift in the settings where individuals received beds for adolescents is an ongoing challenge, and outpatient care. For example, in 2010, 88% of out­ another respondent explained that many adolescents patient visits to licensed community and free clinics are placed in facilities outside county lines, such as in occurred in specialty mental health clinics, and 12% neighboring Alameda or Solano Counties. of outpatient clinic visits occurred in physical health clinics (Table A10). In 2016, the percentage of visits to specialty mental health clinics decreased to 75% Table A9. Inpatient Bed Capacity and Utilization, Contra Costa County, 2010 and 2016 and the percentage of visits to physical health clinics increased to 25%. This shift could reflect initiatives to 2010 2016 CHANGE integrate physical and behavioral health care whereby Licensed Psychiatric 83.6 94.8 +13% primary care providers are taking a more active role in Inpatient Beds (per one million persons) treating mental health problems. Percentage of Licensed Bed 69.6% 48.2% –31% Days Filled Table A10. M ental Health Visits at Community and Outpatient Mental Health Clinics, Contra Costa Psychiatric Discharges 2.9 2.5 –14% (per 1,000 persons) County, 2010 and 2016 Average Length of Stay (days) 7.3 6.7 –9% 2010 2016 CHANGE Total Clinic Visits for 24.6 22.8 –7% Mental Health Diagnoses Need for More Intermediary Care, Especially for (per 1,000 persons) Medi-Cal Enrollees Visits to Outpatient Clinics 2.9 5.6 +93% Intermediary care is reportedly a gap in Contra Costa for Mental Health Services (per 1,000 persons, primary care) County, especially for individuals enrolled in Medi-Cal. According to respondents, partial hospitalization Visits to Psychiatric 21.7 17.3 –20% services are not available for people with Medi-Cal Outpatient Clinics (per 1,000 persons) coverage. Respondents had differing views about whether coverage of partial hospitalization services falls under the purview of county mental health plans. Although use of outpatient mental health services This uncertainty may contribute to the gaps in care. decreased from 2010 to 2016, some respondents According to one respondent, individuals without noted an increase in use in recent years, especially access to partial hospitalization are often discharged among the Medi-Cal population. Respondents to a crisis residential facility, where they may receive from two major health systems in the county noted care for up to one month. While these services partly increased use of outpatient services at their systems. fill the intermediary care gap, respondents said that Some respondents also described long wait times to partial hospitalization services, if available, could bet­ see psychiatrists and other mental health providers. ter serve the needs of some individuals. Respondents Factors Affecting Inpatient Psychiatric Bed Availability in California: Findings from San Diego, Fresno, and Contra Costa Counties 23 The county has been making efforts to expand outpa­ Growing Workforce Shortages tient capacity. For example, to help reduce wait times Like other California counties, Contra Costa has a for psychiatry appointments, the County Behavioral mental health workforce shortage. The number of Health Department hired additional psychiatrists and psychiatrists in the county declined slightly from 122.5 implemented tele-psychiatry within the county-funded psychiatrists per one million persons in 2010 to 120.3 system of outpatient clinics. Also, in 2018, a major in 2016 and remains lower than the statewide num­ health system in the county moved its outpatient men­ ber of 140.5 psychiatrists per one million persons tal health services to a larger building, which allowed (Table A12). Consistent with these findings, all the the system to expand capacity for outpatient services respondents in Contra Costa County commented on and provide new service offerings, such as smoking a need for more psychiatrists, nurse practitioners, psy­ cessation programs and educational programs related chologists, and social workers. One respondent said to co-occurring conditions. The system’s outpatient provider shortages are getting worse over time, par­ mental health services primarily serve people with ticularly in areas of the county where the population commercial insurance. is growing. The county saw a 19% increase in use of EDs for Table A12. M ental Health Staff, Contra Costa County, mental health issues from 2012 to 2017, which may 2010 and 2016 be related to challenges with access to outpatient 2010 2016 CHANGE services (Table A11). In addition, visits to the county’s designated psychiatric ED at Contra Costa Regional Active Psychiatric Hospital 134.9 190.6 +41% and Non-hospital Staff Medical Center increased by 32% over the analysis (per one million persons) period. Despite increased use of the ED, data indicate Psychiatrists 122.5 120.3 –2% that the share of people being admitted from the ED (per one million persons) to inpatient facilities declined slightly. Taken together, these findings suggest that people in the county may be using the ED as a source of outpatient treat­ Contra Costa has been pursuing several strategies to ment, possibly as a result of gaps in access in certain increase its mental health workforce, including using areas of the county or because of wait times for out­ tele-psychiatry, providing salary raises to psychiatrists, patient care. It is also possible that people are being integrating mental health into primary care, and col­ transferred from the ED to other facilities outside the laborating with medical schools to provide residency county or to crisis facilities. placements to students. In addition, a mental health clinic in the county reported that it hires unlicensed social workers, marriage and family therapists, and Table A11. E mergency Services Utilization for Mental Health Diagnosis, Contra Costa County, professional counselors to provide services because Selected Years of the difficulty recruiting licensed professionals. The clinic provides the required supervision hours for 2010 2016 CHANGE each provider type so they are eligible for licensing. Visits to Psychiatric EDs 7.0 9.2 +32% The county is also using MHSA funds to provide paid (per 1,000 persons) 2012 2017 CHANGE internships to mental health providers and is imple­ menting a student loan repayment program as an Total ED Visits for 14.1 16.9 +19% Mental Health Diagnosis incentive to recruit qualifying professionals to work in (per 1,000 persons) the public mental health system.16 Admitted ED Visits for 1.4 1.3 –3% Mental Health Diagnosis (per 1,000 persons) ED Discharges to Psychiatric 2.2 2.0 –8% Care (per 1,000 persons) California Health Care Foundation www.chcf.org 24 MHSA Funding Used to Expand Intermediary Care and Crisis Services Overall, MHSA spending increased 43% in Contra Costa County from 2013 to 2017. The county report­ edly uses these funds to operate and contract with mental health service providers to support full-service partnerships in an effort to provide clients with the full spectrum of mental health and social services. MHSA funds have been used to add a new mobile cri­ sis response team to serve adults, and to expand the hours during which the child mobile crisis response team operates. Additionally, Contra Costa Behavioral Health Services is funding an assertive community treatment program for transition-age youth (ages 15 to 26) and is considering adding a short-term resi­ dential treatment program.17 These services will be available in the fall of 2020 and may help fill gaps in care for this population. Conclusion Although quantitative analyses suggest that, overall, Contra Costa County had sufficient beds to meet its needs as of 2016, respondents pointed to a need for more beds, particularly for adolescents and indi­ viduals involved in the criminal justice system. From 2010 to 2016, inpatient care decreased, but use of ED services rose, suggesting that barriers to access­ ing outpatient services exist. Respondents identified intermediary care and especially partial hospitaliza­ tion and assertive treatment services as key gaps in the county’s mental health delivery system. Efforts to increase the mental health workforce may increase capacity to provide outpatient care and reduce wait times for outpatient appointments. Factors Affecting Inpatient Psychiatric Bed Availability in California: Findings from San Diego, Fresno, and Contra Costa Counties 25 Appendix B. Methods The quantitative component of this study uses publicly available data for 2010–2017 (where available) from the California Office of Statewide Health Planning and Development (OSHPD), Mental Health Services Oversight and Accountability Commission, the US Census Bureau’s US Census and American Community Survey, the Health Resources and Services Administration’s Area Health Resources Files, and the UCLA Center for Health Policy Research’s California Health Interview Survey. All data are aggregated to the county level for each of the three counties in the study — San Diego, Fresno, and Contra Costa — as well as to the statewide level for purposes of comparison to each county trend. State psychiatric hospitals are excluded from the analyses. DATA TYPE YEARS VARIABLES SOURCE(S) OSHPD Financial 2010– $ Psych ED visits per 1,000 persons FY 2012–2013 through Data 2016 FY 2016–2017 $ Licensed psychiatric inpatient beds per one million persons FY 2009–2010 through $ Available psychiatric inpatient beds per one million persons FY 2011–2012 $ Psych discharges per 1,000 persons $ Psych inpatient days per 1,000 persons $ Average length of stay for psych inpatient stays* $ Percentage of licensed bed days filled* $ Active psychiatric hospital and non-hospital staff per one million persons $ Licensed acute care children’s psychiatric inpatient beds per one million persons $ Available acute care children’s psychiatric inpatient beds per one million persons $ Percentage of licensed children’s bed days filled* $ Active acute care children’s psychiatric hospital staff per one million persons $ Partial hospitalization days per 1,000 persons $ Acute care children’s psych discharges for per 1,000 persons $ Acute care children’s psych inpatient days per 1,000 persons $ Average length of stay for acute care children’s psychiatric inpatient stays* OSHPD Clinic Data 2010– $ Primary care clinic mental health patients per 1,000 persons 2012–2017 Primary Care 2016 Clinic Utilization Data $ Specialty care clinic psych patients per 1,000 persons 2010–2012 Primary Care $ Total clinic visits for mental health (primary care and specialty) Clinic Utilization Data per 1,000 persons* 2012–2017 Specialty Care $ Psych clinicians in primary care clinics per one million persons Clinic Utilization Data 2010–2012 Specialty Care Clinic Utilization Data OSHPD Emergency 2012– $ Total ED visits for mental health diagnoses per 1,000 persons 2012–2017 Emergency Department Data 2017 Department Utilization $ Admitted ED visits for mental health diagnoses per 1,000 persons Mental Health 2013– $ Total mental health expenditures per capita FY 2012–2013 through Services Oversight 2017 FY 2016–2017 $ Total MHSA expenditures per capita and Accountability Commission California Health Care Foundation www.chcf.org 26 DATA TYPE YEARS VARIABLES SOURCE(S) US Census 2010– $ County and state population Population Totals 2017 Percent Rural Population American 2010– $ Percentage of population below the poverty line factfinder.census.gov Community Survey 2017 Area Health 2010, $ Psychiatrists (MDs) per one million persons Health Resources & Resources Files 2015, Services Administration 2016 “Data Downloads: Area Health Resources Files” (county level) California Health 2011– $ Likely has had serious psychological distress during past month AskCHIS Interview Survey 2017 $ Ever seriously thought about committing suicide *Calculated from variables in the data. Factors Affecting Inpatient Psychiatric Bed Availability in California: Findings from San Diego, Fresno, and Contra Costa Counties 27 Endnotes 1.Paul Sisson, “Is San Diego Headed for a Psych Bed Crisis?,” 13.Coffman et al., California’s Current and Future Behavioral San Diego Union-Tribune, July 23, 2018. Health Workforce. 2.California Hospital Association, California’s Acute Psychiatric 14.Lynne Ashbeck, “Fresno County Is Transforming How It Helps Bed Loss (PDF) (Sacramento: California Hospital Association, People with Mental Illness,” Fresno Bee, September 26, March 28, 2018), accessed August 31, 2018. 2018, accessed July 19, 2019. 3.For a map of inpatient psychiatric facilities in the state, 15.Coffman et al., California’s Current and Future Behavioral please see the Substance Abuse and Mental Health Services Health Workforce. Administration’s Behavioral Health Treatment Services 16.Contra Costa Health Services, Contra Costa County: Mental Locator. Health Services Act Three Year Program and Expenditure Plan 4.Margaret Tatar and Richard Chambers, Medi-Cal and Update, Fiscal Year 2019–2020 (Martinez, CA: March 2019). Behavioral Health Services (Oakland: California Health Care 17.Contra Costa Health Services, Contra Costa County: Mental Foundation, February 2019). Health Services Act Three Year Program and Expenditure Plan 5.Janet Coffman, Tim Bates, Igor Geyn, and Joanne Spetz, Update, Fiscal Year 2019–2020 (Martinez, CA: March 2019). California’s Current and Future Behavioral Health Workforce (San Francisco: Healthforce Center at UCSF, February 12, 2018). 6.For a detailed explanation of these funding sources, see Deborah Reidy Kelch, Locally Sourced: The Crucial Role of Counties in the Health of Californians (Oakland: California Health Care Foundation, October 2015), accessed October 13, 2019. 7.Logan Kelly, Allison Hamblin, and Steve Kaplan, Behavioral Health Integration in Medi-Cal: A Blueprint for California (Oakland: California Health Care Foundation, February 2019). 8.David Wright, “Clarification of Ligature Risk Policy,” memorandum to State Survey Agency directors, Centers for Medicare & Medicaid Services, December 8, 2017. 9.The Centers for Medicare and Medicaid Services defines ligature risk as anything that could be used to attach a cord, rope, or other material for the purpose of hanging or strangulation, such as shower rails, coat hooks, pipes, bedsteads, ceiling fittings, and hinges. The focus on a “ligature resistant” or “ligature free” environment is primarily intended for psychiatric units/hospitals. 10.These metrics may overstate the actual bed capacity across the county because they reflect the proportion of licensed acute inpatient psychiatric beds that are occupied. Some physical beds may not have available psychiatric staff, such as psychiatrists or other mental health professionals, and therefore may not be operational. 11.California Department of State Hospitals, 2018 Annual Report (PDF) (Sacramento: California Department of State Hospitals, 2018), accessed October 10, 2019. 12.Ha Tu, Lara Converse, Annie Doubleday, and Paul Ginsburg, San Diego: Major Providers Pursue Countywide Networks and New Patient Care Models (Oakland: California Health Care Foundation, June 2016). California Health Care Foundation www.chcf.org 28