Cross-Sectional Study of Homeless High Service Utilizers in Los Angeles County Jails: Race, Marginalization and Opportunities for Diversion

Abstract

Objective

To describe the demographic, legal, and clinical characteristics of a cross-section of incarcerated homeless individuals with the highest utilization of Los Angeles (LA) County public services in order to increase opportunities for focused jail diversion.

Methods

The “5% list” (N=5,905 in February 2018), LA County’s list of homeless individuals with the highest 5% utilization of public services, was cross-matched with the total jail population to obtain a sample of 333 homeless high-utilizing individuals. This sample was compared with the overall jail population (N=17,121) from publicly available aggregate data by Chi-square testing.

Results

84% of the high-utilizing sample were male, 38% Black, 37% Hispanic, 21% White. 67% were prescribed psychiatric medication. Compared with the overall jail population, the sample was significantly older, with a greater proportion of Black and White persons, and a lesser proportion of Hispanic individuals relative to the overall jail population. A significantly greater proportion of high-utilizing individuals faced misdemeanor charges.

Conclusion

These data highlight the compounding effects of homelessness, race, and mental illness on carceral and social vulnerability. Findings suggest homeless high utilizers in jail with mental illness are likely to benefit from court-based diversion efforts aimed at housing and treatment.

Introduction

Nationwide, the mutually reinforcing risks of homelessness, mental illness, and incarceration are increasingly recognized1 and considered a pressing public health concern. Commonly referred to as one of the country’s largest mental health institutions, Los Angeles (LA) County jails have an average population of 17,000 incarcerated individuals at any given time,2 and approximately 5,000 of those are receiving mental health treatment.2 Relatedly, Los Angeles has one of “the most acute homelessness problems,”3 with approximately 59,000 homeless individuals at any point in time.3 Incarceration and homelessness are particularly salient when understanding social vulnerability in racially marginalized communities affected by a nexus of structural racism, poor mental health care access, housing insecurity, and increased vulnerability to criminalization. In light of growing concern for these matters, the Office of Diversion and Reentry (ODR) was created by the Los Angeles County Board of Supervisors in 2015. In part, ODR’s mission is to implement criminal justice diversion for homeless individuals with mental health and/or substance use disorders in order to halt the cycle of incarceration and improve health outcomes.4 Given that these issues disproportionately affect communities of color in Los Angeles, ODR’s work has the potential to be instrumental in promoting racial and health equity.

Since 2016, LA County has compiled a list of homeless single adults who represent the top 5% of utilizers of public services4 as measured by cost (namely, through the Homeless Management Information System, the Department of Health Services, Department of Mental Health, Department of Public Health, Department of Public and Social Services, LA Sheriff’s Department and Probation, as captured in county administrative data systems). This “5% list” is updated semiannually and made available to ODR so it can identify high-need individuals in jail for prioritized diversion.

While descriptive studies of high utilizers and frequently incarcerated individuals have been published in other major cities in the country, no publications to our knowledge have described the characteristics of this population in LA, although national attention on LA County’s jail mental health crisis has grown.5 In light of the Board of Supervisors’ 2019 decisions to take an approach that moves  “away from incarceration and toward treatment, rehabilitation and diversion,”5 these data are requisite for the expansion of ODR’s programming, as well as for the creation of further targeted interdisciplinary interventions to support the most vulnerable in the county.

In this study, we describe the sociodemographic, legal, and clinical characteristics of individuals on the “5% list” incarcerated in LA County in February 2018. Additionally, we evaluate LA County courts in the context of neighborhood-based poverty to examine locales with the highest concentration of arrests of this population to identify opportunities for increased systematic intervention. This study represents an important contribution to the sparse literature on incarcerated homeless individuals in Los Angeles and their public health needs.

Methods

ODR cross-matched all 17,391 persons in LA County jails on February 27, 2018 with all persons on the county’s most recent “5% list” (n=5905), yielding a cross-sectional, purposive sample of n=333. Sociodemographic and legal data, prescribed medications, and receipt of Jail Mental Health (JMH) services (consisting of high and moderate observation units, an inpatient psychiatric unit, and/or persons in the general jail population prescribed psychotropic medications) among this sample were compared with the overall county jail population using publicly available aggregate data from the first quarter of 2018 reported on the LA Sheriff’s Department website.2 Observed proportions for categorical measures in the study sample were compared with the reported (expected) proportions in the overall jail population (N=17,121) by Chi-square testing with P≤.05 level of significance. Bonferroni correction was applied for post-hoc pairwise testing of individual categories within non-dichotomous measures. Descriptive data were collected for psychiatric medications prescribed in jail within 30 days of data collection and for primary courthouses where incarcerated individuals in the sample had active cases. Stata version 14 was used for all statistical analyses.

Results

As summarized in Table 1, of the “5%” study sample (N=333), 279 (84%) were male, 127 (38%) were Black, 123 were Hispanic (37%), 71 White (21%), and 12 (4%) represented other races. Of note, while the term Hispanic in fact denotes linguistic ability, the jail utilizes the term interchangeably with Latino or Latinx. The median age was 36 years old (interquartile range=18-54). 211 individuals (63%) were charged with a felony on arrest, and 208 (63%) received JMH services. 223 individuals (67%) were prescribed a psychiatric medication within 30 days of data collection, the majority of which were antipsychotics (58%), followed by antidepressants (27%) and mood stabilizers (21%). 40 individuals (12%) received a long-acting injectable (LAI) antipsychotic. 124 high-utilizing individuals (37%) had an active case at a downtown courthouse, followed by Mental Health Court (indicating active competency to stand trial proceedings) with 54 persons (16%).

Table 1. Sociodemographic characteristics, legal status, and receipt of Jail Mental Health housing and prescribed medication(s) among homeless high utilizers of Los Angeles County public services (“5%” sample; n=333) compared with overall jail population means from first quarter 2018 (N=17,121).

Characteristic	“5%” study sample n=333, n (%)	Overall jail population N=17121, n (%)		P (χ2)
Sex				.09 (2.8)
Female	54 (16)	2239 (13)
Male	279 (84)	14882 (87)
Age (years) – median (IQR)	36 (18–54)	n/a
By age group				<.0001 (43.2)
18–25 years	43 (13)	4159 (24)		<.001 (23.1)a
26–34 years	98 (29)	5718 (33)		.0004 (12.8)a
35–39 years	65 (20)	2371 (14)		<.001 (47.3)a
40–44 years	28 (8)	1499 (9)		.02 (5.2)
45 years or older	99 (30)	3374 (20)		<.0001 (20.6)a
Race				<.0001 (27.8)
Black	127 (38)	5091 (30)		.009 (11.0)b
Hispanic	123 (37)	8694 (51)		<.0001 (25.0)b
White	71 (21)	2630 (15)		.003 (8.9)b
All other races combined	12 (4)	706 (4)		.6 (0.2)
Charged with felony on arrest	211 (63)	15355 (91)c		<.0001 (270.6)
Legal status				.3 (1.0)
Fully or partially sentenced	194 (58)	9500 (56)
Not sentenced	139 (42)	7621 (44)
In Jail Mental Health housing	208 (63)	4695 (27)		<.0001 (198.5)
Additional characteristics			“5%” sample (n=333), n (%)
Prescribed any psychiatric medication(s)d			223 (67)
Antipsychotic (oral or long-acting injectable)			193 (58)
Long-acting injectable (LAI) antipsychotic			40 (12)
Mood stabilizer			71 (21)
Antidepressant			91 (27)
Types of psychiatric medicationse prescribedd
One			46 (14)
Two			120 (36)
Three			51 (15)
Four			6 (2)
Primary courthouse
Central District (Downtown)			124 (37)
Municipal or Superior			94 (28)
Revocation Court			30 (9)
Los Angeles Superior, Department 95 (Mental Health Court)			54 (16)
Long Beach Municipal or Superior			23 (7)
LAX (Los Angeles International Airport) Superior			20 (6)
Van Nuys Municipal or Superior			14 (4)
North Valley Superior			12 (4)
Torrance Municipal or Superior			12 (4)
Antelope Valley Municipal or Superior			11 (3)
Norwalk Municipal or Superior			11 (3)
Compton Municipal or Superior			10 (3)
All other courthouses combined			42 (13)

a. Statistically significant at P≤.01 level in post-hoc testing with Bonferroni correction for multiple hypotheses.

b. Statistically significant at P≤.013 level in post-hoc testing with Bonferroni correction for multiple hypotheses.

c. Mean value based on average overall inmate census performed weekly from January to March 2018 (N=16,950).

d. Within 30 days of primary data collection.

e. Including sedatives and anxiolytics.

Relative to the overall jail population (N=17,121), individuals in the sample were significantly older by Chi-square testing, particularly in the 35-39 years (20% vs 14%) and 45 years and above (30% vs 20%) categories. The sample also significantly differed by racial status, with greater Black (38% vs 30%), greater White (21% vs 15%), and lower Hispanic (37% vs 51%) proportions of incarcerated persons. A significantly lower proportion of individuals in the “5%” sample were charged with felonies (63% vs 91%), whereas significantly more received JMH services (63% vs 27%). There were no significant differences in sex and current sentencing status.

Discussion

These study findings offer insight into the population of homeless, incarcerated individuals with high service utilization in Los Angeles. These data are especially relevant given recent movement toward non-carceral approaches to addressing mental illness and homelessness in the county.5

In addition to the clinical and legal implications of the presented data, which will be discussed below, the demographic characteristics and disparate racial representation among the study samples warrant examination. Extensive literature highlights the structural racism embedded in various stages of the American carceral system, rendering Black individuals disproportionately vulnerable to incarceration.6 While Latinos are also more vulnerable to incarceration, this “5%” sample of homeless high-service utilizers contains a lower proportion of Hispanics than the general jail population. This finding may reflect broader patterns of underrepresentation of Hispanics in homeless populations,7 as well as lower health service utilization,8 decreasing Hispanic representation on the “5% list.” In contrast, the “5%” sample reflects a greater representation of Black individuals relative to the overall jail population (in which Black people are already over-represented compared with LA County as a whole),9 echoing the impacts of interlocking stigma and vulnerabilities on incarceration risk. These findings stress the limitations of a “colorblind” approach10 to mental health care, serving as an impetus for clinicians and policymakers to understand the centrality of racism as a determinant of mental health when treating, advocating for, and creating programming to support communities of color.

Literature demonstrates that being homeless, diagnosed with a psychotic disorder, and/or being Black in America increases individual-level risk of incarceration.11 Furthermore, incarceration itself increases both the risk of homelessness1 and the severity of psychiatric symptoms. While being White (with its associated structural privilege)6 and/or aged >45 years are usually related to disproportionately lower incarceration risk,11 our “5%” sample demonstrates increased representation of White individuals and those aged >45 years relative to the general jail population. Despite this, White individuals on the “5% list” were still disproportionately underrepresented relative to the general LA county demographic racial composition. This reduction in underrepresentation among homeless high service utilizers suggests that the protective function of White privilege may be abated by homelessness and serious mental illness (SMI). The mutually reinforcing impacts of societal disenfranchisement and their unique consequences for various communities highlight a need to move beyond single-issue approaches to public health interventions toward more intersectional ones, acknowledging “the interlocking nature of co-occurring social categories… and the forms of social stratification that maintain them.”10

The data presented here warrant an examination of physician treatment practices and their downstream impact on vulnerability to incarceration. Literature demonstrates that in real-world settings, LAI antipsychotic medications significantly delay time to psychiatric hospitalization and incarceration/reincarceration12 compared with equivalent oral formulations. Given the high rate of antipsychotic use (58%) but relatively low rate of LAI use (12%) in our sample (Table 1), these data highlight a potential opportunity for increased LAI use in vulnerable populations to reduce risk of hospitalization and reincarceration when treating underlying mental illness. Without dismissing the history of how long acting injectable medications have been used coercively in communities of color, given their efficacy in reducing psychiatric destabilization and subsequent incarceration risk, they may represent one of many tools that can reduce the burden of disease in communities with structural barriers to regular mental health care access, and increased vulnerability to criminalization.

With regard to legal status, the “5%” sample had significantly lower prevalence of felony offenses and more misdemeanor charges relative to the overall jail population (Table 1.) This difference reinforces existing evidence suggesting an inverse relationship between the seriousness of mental illness and severity of legal charges13; defendants found incompetent to stand trial are more often charged with minor “public nuisance” offenses and tend to have more serious psychiatric illness than those with felony charges.13 Moreover, the finding that significantly fewer high utilizers with SMI were facing felony charges also challenges public perceptions of dangerousness among homeless individuals with SMI, supporting the viability of scaled-up diversion and community integration.

Finally, these data suggest the importance of targeted interventions based on geographic patterns of incarceration. Courts with the highest number of active cases in our sample were located downtown (Table 1), where poverty and LA’s homeless population are most concentrated.14 This geographic pipeline from homelessness to the carceral system underscores the link between poverty and criminalization (particularly for Black individuals). Furthermore, nearly one-quarter of downtown cases (in these neighborhoods of high homelessness) are made through Revocation courts, indicating arrests for violations of probation or parole rather than new criminal charges. This interplay between areas of poverty and incarceration, particularly for low-level violations, stresses the need for increased accountability, as well as mental health, case management, and housing interventions via individual courthouses with high arrest profiles of homeless defendants with SMI.

Since its inception in 2015, ODR has created and operated various programs in the service of its mission: to implement and develop criminal justice diversion for individuals with mental health and/or substance use disorders and to provide re-entry support services to vulnerable populations affected by the carceral system.15 These endeavors have included community-based competency restoration programs for clients with misdemeanor charges (1,238 clients have been removed from jail and engaged since October 2016)15 and a supportive housing program for homeless individuals with serious mental disorders diverted from LA County jails,15 the majority of which have been diverted via Downtown LA courthouses. ODR’s supportive housing program, which couples housing with case management and mental health treatment, has served 1,728 clients since 201615 and has a 91% rate of housing retention after six months.3 In 2018, a novel community-based competency restoration program for individuals with felony charges was launched, which has served 133 clients to date.15 Additionally, ODR offers programs to divert incarcerated pregnant women, alternatives to state hospital diversion, and programming to promote efforts in critical intervention training and mental evaluation teams.15 The majority of ODR’s programs serve individuals from LA County’s “5% list,” and all of these programs necessitate inter-agency collaboration and increased data to appropriately address the complex social, medical, legal, and mental health needs of the clients served.

The study results reported here highlight the need to scale up existing diversion programs in districts where homelessness and high service utilizers are concentrated, as well as to increase education and intervention in courts with the highest arrest rates of vulnerable individuals on the “5% list.” Additionally, these data have significant implications for community and carceral treatment practices, legal policy, and the need for applied critical race theory in legal and health paradigms (Table 2). Given that the intersection of homelessness risk, untreated mental health burden, and criminalization disproportionately impacts Black communities in Los Angeles, mental health diversion programs such as ODR are likely to benefit from further research and improved understanding of the manifestations of structural racism. In turn, this understanding can allow for the systematic dismantling of structural racism in the promotion of health equity.

Table 2. Key findings and possible implications.

Findings	Possible implications
Latinx, White, and Black American communities in Los Angeles are distinctively impacted by the nexus of homelessness, incarceration, and mental illness.	Applied critical race theory: These data illustrate the need to move beyond “colorblind” approaches10 to equity promotion and acknowledge intersectionality-- “the interlocking nature of co-occurring social categories… and the forms of social stratification that maintain them.”10
Long-acting injectable (LAI) antipsychotic medications can delay time to psychiatric hospitalization and incarceration/ reincarceration12 compared with equivalent oral formulations, yet only 12% of this vulnerable population were prescribed LAIs.	Prescribing practices: This highlights a potential opportunity for increased LAI prescribing for vulnerable and willing patients with serious mental illness.
High service utilizers have a significantly lower prevalence of felony charges, in line with existing evidence that an inverse relationship exists between the seriousness of mental illness and severity of legal charges.13	Diversion viability: This may challenge lay-community fears about mental illness and serious crime risk, supporting the viability of scaled-up jail diversion and community integration.
Courts with most active cases for this sample were located in LA neighborhoods with the most concentrated poverty and homelessness,14 suggesting a geographic pipeline from homelessness to the carceral system, particularly for Black individuals.	Court-based interventions: This pattern highlights the need for increased mental health, case management, housing interventions, and institutional accountability-- particularly in individual courthouses with high arrest profiles of homeless defendants with SMI.
The intersection of homelessness risk, untreated mental health burden, and criminalization disproportionately impacts Black communities in Los Angeles.	Centrality of antiracism research: Public health and mental health diversion programs such as ODR are likely to benefit from further research on the manifestations of structural racism in order to inform its effective and systematic dismantling.

SMI, serious mental illness; ODR, Office of Diversion and Reentry

Despite the implications of these findings, some limitations to the available data and possible analyses include the cross-sectional study design (restricting causal inference), the unavailability of individual-level data, substance-use data, and extensive descriptive measures from the control population data, restricting assessment of potential interactions and confounding. In addition, the racial categories used reflect those documented by the Sheriff’s department when collecting data, which limits the ability to discern heterogeneity among these designations or racial self-identification of sample individuals. Finally, while psychiatric medication data were available for our sample, specific diagnoses were not, and medication and homelessness data were not available for the control overall jail population. Nonetheless, despite these limitations, these findings represent a crucial opportunity to inform interventions for a highly marginalized and often-overlooked population in Los Angeles County and for similarly marginalized populations nationwide.