Introduction

Prevalence and incidence of mental health disorders are increasing in the US [1,2,3], but treatment utilization remains very low [4,5,6]. Individuals involved in or at risk for involvement in the criminal justice (CJ) system are at a particularly high risk of mental health problems [7,8,9]. Treatment uptake is disproportionately low in CJ-involved individuals, leading to people with unmet mental health treatment needs being over-represented in the justice system [10]. Importantly, the relationship between mental health treatment and CJ involvement is complex and bidirectional. Those with untreated mental health problems are more likely to interact with law enforcement and be arrested and incarcerated [11, 12], while the trauma and stress of navigating the CJ system can create or worsen mental health symptoms in communities where access to mental health treatment is scant [13,14,15]. Promoting treatment before incarceration may serve to slow or weaken the effects of this cycle. Mandated mental health treatment by a court that targets individuals with diagnosed disorders and historically low treatment use is one approach to promote treatment utilization in the short-term, in turn improving mental health symptoms and reducing incarceration risk in the longer-term.

Court-ordered treatment (COT) for mental health has been variously defined and applied in legal settings and is often in lieu of a pending jail or prison sentence. COT may consist of inpatient or outpatient therapeutic or correctional services which are generally prescribed by a judge during criminal proceedings [16]. COT can also refer to involuntary inpatient or outpatient commitment for individuals deemed to be a danger to themselves or others, and unable or unwilling to consent to behavioral treatment [17]. COT can be an alternative to incarceration for people living with mental illness who are disproportionately impacted by the criminal justice system. COT has also been defined as “coercive”, meaning those impacted are required to participate in a treatment program or face incarceration or other legal penalties. For some COT programs, participation and therapeutic progress are reportable to corrections or other CJ authorities, which contrasts with some clinical principles of confidentiality between patient and provider. Mental health courts are one form of COT and have been shown to address behavioral health, social service needs, and reduce recidivism using a comprehensive team approach led by a judge [18]. While definitions vary, the overall goal of COT is to increase treatment use among a low-use population at risk for continued and recurring interactions with law enforcement and incarceration.

Ample research has been conducted around court-ordered services, outcomes, and the merits and limitations of coerced treatment. However, not all mental health treatment providers offer services specifically tailored for court-ordered clients. Roughly half (53%) of outpatient mental health treatment facilities offered services for court-ordered clients in 2020, which varied from as low as 46% in the US Northeast region to as high as 58% in the Midwest [19]. During the same timeframe, the incarceration rate in the US surpassed 600 per 100,000 residents, the nation’s prison population included over 1 million people, and the disproportionate representation of Black Americans in prisons continued to widen [20]. Mental health is a driver of such a high rate of incarceration in the US, as people with a mental illness in the US are 10 times more likely to be incarcerated than hospitalized [21]. Mental health burden, incarceration rates, and treatment access all vary by state, which creates a complex network of intersecting determinants of health. As a result, scholars and community members have called for a re-examination of relying on prisons for safety-net mental health treatment while advocating for healthcare systems that can absorb an influx of justice-involved clients who have been diverted away from incarceration and admitted into formal treatment services [22].

Individuals involved in the CJ system have unique and complex behavioral healthcare needs, resulting in a growing need for treatment facilities to offer services tailored specifically to those ordered to mental health treatment by a court. Our national study was designed to assess changes in the availability of justice-tailored services over time and between states, as well as characteristics associated with offering these services. Our scientific approach was broadly informed by Link and Phelan’s Fundamental Cause Theory [23, 24], which helps to explain the association between the social and environmental conditions in which one lives and the effect on health outcomes. We apply the concept of “place matters” to our close examination of temporal changes within states and state-level characteristics associated with tailored mental health services for justice-involved clients. We discuss our findings within this framework and offer suggestions for health systems strengthening and future research.

Methods

Data source and sample

Data were sourced from the National Mental Health Services Survey (N-MHSS; 2016, 2018, and 2020), which is cross-sectional administrative data describing characteristics of treatment facilities in the United States. We restricted our sample to outpatient mental health treatment facilities in the US (n = 14,223 facilities; 2016 n = 4744; 2018 n = 4616; 2020 n = 4869). In short, N-MHSS is hosted by the Center for Behavioral Health Statistics and Quality, which is housed within the Substance Abuse and Mental Health Services Administration. N-MHSS includes all known public and private facilities in the US that provide specialty mental health treatment, which SAMHSA defines as any facility or entity that provides mental health treatment services to people with mental illness. Representatives from each facility completed an annual survey about services provided. In 2020, the response rate among facilities eligible to participate was 89% [19]. This study was a secondary analysis of publicly available administrative data describing characteristics of mental health facilities. No human subjects were involved in this study and no individual-level data was available or used.

Variables

Our dependent variable of interest was whether outpatient mental health treatment facilities offered court-ordered services. Respondents were asked to choose all that apply to the survey question, “Which of these services and practices are offered at this facility, at this location?” The response option identified for this study was “court-ordered treatment”. We considered several independent variables based on a priori understanding of facility- and state-level characteristics that are related to mental health, healthcare, and the criminal justice system. We included the year of data collection [2016 (reference), 2018, and 2020], facility owner type [private for-profit (reference), private non-profit, or public agency], whether a facility accepts Medicaid as payment [yes or no (reference)], whether a facility charges payment fees on a sliding scale [yes or no (reference)], whether a facility provides substance abuse treatment [yes or no (reference)], whether a facility provides trauma therapy [yes or no (reference)], whether a facility offers integrated primary care services [yes or no (reference)], whether a facility provides housing services [yes or no (reference)], and the number of total clients enrolled in April 2020 as an indication of the overall size of facility [1–100 clients (reference), 101–500, 501–1000, and more than 1000]. Of note, N-MHSS only reports data on the total number of clients on even years. As a result, we were not able to include data from 2017 or 2019 in our analyses. We also sourced state-level information externally and appended it to our analytic N-MHSS dataset. We used data from the US Census Bureau 2020 decennial census to define the percentage of a state’s residents living in a rural area [25]. Residence in a rural location is defined by the Census Bureau as anyone not residing in an urbanized area of 50,000 people or more, or in an urban cluster of at least 2500 and less than 50,000. To ease interpretation of our model results, we categorized rurality into three groups based on tertiles, resulting in classifications of more urban (5.8–12.6% of residents in a rural area), mid (12.7–25.9% of residents in a rural area), and more rural (26.0–64.9% of residents living in a rural area). We used publicly available data from The Sentencing Project to include a measurement of the state-level imprisonment rate per 100,000 residents [26]. We categorized imprisonment rates based on the tertile distributions, which resulted in low imprisonment (133–302 people in prison per 100,000 population), mid (303–381), and high (382–680). We included the US Census Bureau’s Diversity Index which is a useful indicator of the racial/ethnic composition of a state [27]. The Diversity Index is a percentage that indicates the probability that two people chosen at random will be from different racial and ethnic groups, with higher percentages representing a larger mix of racial/ethnic groups in a state. We categorized the Diversity Index based on tertiles which resulted in low diversity (18–44% probability that two people chosen at random will be from different racial/ethnic groups), mid diversity (45–63%), and high diversity (64–76%). Finally, we included whether facilities were located in a state that had expanded Medicaid under the Affordable Care Act by the year of data collection [28].

Analysis

We used SAS v9.4 for all analyses [29]. We described the distribution of each study variable, for the total sample and disaggregated by our dependent variable of whether a facility offered court-ordered services. We used a barbell graph to plot the proportion of facilities that offered court-ordered services by state and over time (2016 and 2020 only). To further assess trends over time, we calculated the proportion of facilities that offered court-ordered services for each study variable and within each year (all years). Finally, we used multiple logistic regression to model court-ordered services, adjusted for all study variables, and reported the adjusted odds ratio, 95% confidence interval, and p-value for each comparison. We specified that facilities are clustered within states to account for the correlation between facilities within states, under the assumption that facilities in one state are likely more similar to each other than to facilities in other states.

Results

Slightly more than half (55%) of outpatient mental health treatment facilities offered court-ordered treatment (Table 1). Court-ordered treatment availability was above average in 2020 (58%), in private for-profit facilities (65%), facilities that accepted Medicaid as payment (56%), charged payment fees on a sliding scale (61%), provided substance abuse treatment (68%), provided trauma therapy (59%), provided housing services (67%), and had a relatively large capacity serving 501–1000 clients (59%). Court-ordered treatment was also disproportionately higher in states that were more rural (61%), had high imprisonment rates (58%), and low racial/ethnic diversity index scores (59%), and had not expanded Medicaid (60%).

Table 1 Facility- and state-level characteristics of specialty outpatient mental health treatment facilities (N-MHSS; 2016, 2018, and 2020)
Full size table

The proportion of treatment facilities that offered court-ordered treatment varied by state, and court-ordered services increased in the majority of states between 2016 and 2020 (Fig. 1). Over 5 years, the proportion of facilities offering court-ordered services decreased in eleven states (NV, CT, MT, NC, TN, AZ, IN, AL, MO, SC, CO), with the largest decrease observed in South Carolina (from 69% of facilities in 2016 to 33% in 2020).

Fig. 1
figure 1

Proportion of specialty mental health outpatient treatment facilities offering court ordered services, by state, 2016 and 2020

Full size image

Overall, the percentage point change in facilities that offered court-ordered treatment between 2016 and 2020 was 6% (from 52 to 58%; Fig. 1). The change over time was highest among facilities that provided housing services (+ 10.1%) and were owned by a public agency (+ 9.0%).

Finally, the odds of offering court-ordered treatment were 16% higher in 2020 compared to 2016 (aOR = 1.16, 95% CI 1.06–1.27, p < 0.01) (Table 2). We identified several significant associations between our study variables and court-ordered services availability. Notably, the strongest associations identified were that outpatient mental health treatment facilities that also offered substance abuse treatment had nearly three times the odds of offering court-ordered services than facilities that did not offer substance abuse treatment (aOR = 2.95, 95% CI 2.70–3.22, p < 0.0001), and those that offered trauma therapy had two times the odds of offering court-ordered services than facilities that did not offer trauma therapy (aOR = 2.05, 95% CI 1.85–2.27, p < 0.0001).

Table 2 Temporal changes in the percentage of facilities offering court-ordered services, by study variable
Full size table

Discussion

Access to court ordered services in mental health treatment settings is growing nationally and in most states, although the rate of growth varied widely between states. Several states demonstrated low access and slow growth over 5 years, and eleven states decreased access to court ordered services. The mental health population burden in the US is increasing [1] and reports outlining the associations between mental health symptoms and individual and community exposure to the criminal justice system are growing [30,31,32]. Our findings complement the existing literature by describing national epidemiologic trends in the availability of court ordered services by state. Wide variation between states regarding both prevalence and rate of change suggest that further research is warranted to investigate state-level and policy factors that may inhibit or enable the promotion and growth of court ordered services as an alternative to incarceration for people with mental health disorders (Table 3).

Table 3 Multiple logistic regression modeling court ordered services in outpatient MH facilities
Full size table

The strongest association identified with court ordered service availability was whether a mental health treatment facility also provided substance abuse treatment. Roughly half of the facilities offered integrated substance abuse treatment. Dual substance use and mental health disorders are increasingly common. Roughly one in five adults with a substance use disorder also have a co-occurring mental illness, and less than 10% receive treatment for both [33]. Dual disorders are a risk factor for criminal justice involvement [34] and people with dual diagnoses are over-represented in prisons [35]. Our finding of increased availability of court-ordered treatment in facilities that also offer substance use treatment may reflect a growing recognition of the intersection between mental health, substance use, and criminal justice involvement. Individuals participating in court ordered treatment have complex healthcare needs—treating only mental health for individuals with a concurrent need for substance use treatment is not meeting the full needs of the individual and threatens the successful completion of court ordered mental health services, thereby increasing the risk of incarceration by failing to meet the criteria agreed upon with the court. We found that over 2000 facilities across all 50 states offered court-ordered services but not integrated substance use treatment. Public health and clinical interventions targeting justice-involved individuals should consider adding or strengthening substance use treatment services to improve health outcomes and minimize the risk of incarceration. Interventions addressing dual disorders among justice-involved populations can further advance healthcare equity by targeting low socio-economic populations and streamlining payment structures.

We also found a strong association between facilities offering trauma therapy and court-ordered services. Broadly, trauma therapy takes into account how sociocultural forces (like poverty and racism, for example) not only exacerbate the negative effects of trauma, but may act as forms of trauma themselves [36]. Exposure to the criminal justice system and incarceration are inherently harmful to people’s mental health [37], and many experiences in prison lead to symptoms of post-traumatic stress such as anxiety, depression, hypervigilance, and suicidality [38]. As a result, trauma-informed care is widely recommended for people with experience in the criminal justice system [39, 40]. We found that the majority of facilities in our study (83%) offered trauma therapy, which is a positive attribute of the nation’s mental health treatment system. More research is needed to better understand trauma therapy outcomes among mental health treatment clients engaged in court-ordered services.

Other notably strong associations identified with court ordered service availability were acceptance of Medicaid payment and offering a sliding scale fee based on income. The majority of all facilities (> 90%) accepted Medicaid as payment while two-thirds offered sliding scale fees, indicating that vulnerable and low socio-economic populations who are covered by Medicaid or are low income may have reduced burdens in accessing behavioral healthcare. Facilities with administrative and payment structures tailored to low income populations are likely located in and cater to low SES communities. Importantly, service quality varies widely across the US, and healthcare facilities with large low income clientele tend to offer less comprehensive services [41, 42]. Low income clients ordered to mental health treatment by a court may have less access to peripheral services like integrated primary care, social services, or flexible scheduling to accommodate work and family obligations. Future research should investigate differences in the quality of mental health services and treatment outcomes among low income individuals involved in the justice and court systems.

We also identified that facilities in more rural states were the most likely to offer court ordered services, but that the fastest growth in court ordered services over 5 years was among the more urban states. The mental health population burden is generally higher in more urban areas, although rural areas experience substantial disparities in treatment outcomes even though they demonstrate lower overall prevalence [43]. The highest jail incarceration rates in the US are predominately in rural counties, even though crime rates are disproportionately lower in rural areas [44]. Much of this difference is attributable to high rates of pretrial jail incarceration and low resources for mental health diversion alternatives such as crisis stabilization [45]. Between 2016 and 2020, our study identified the biggest increase in tailoring mental health services to individuals with criminal justice involvement in more urban states. Our findings suggest a need to continue supporting mental health system strengthening in rural areas, while also identifying strategies to bolster mental health treatment utilization and incarceration alternatives in urban areas.

Finally, our study has framed access to court-ordered services in mental health treatment settings as a positive characteristic and a useful tool for promoting mental healthcare and diverting individuals with mental health problems away from the carceral system. At the same time, differential access to court ordered services perpetuates disparities that negatively affect Black and other under-represented Americans. Black individuals are less likely than their White counterparts to be referred to a behavioral health court in the first place [46], meaning that they must start serving a sentence without the option of a mental health treatment alternative. Another national report demonstrated that referral to treatment by any criminal justice source was disproportionately low among Black treatment clients [47]. Once accepted into a drug or mental health court, multiple studies have identified racial disparities in graduation rates, with African Americans demonstrating lower success rates than most other racial/ethnic groups [48,49,50,51]. Our study was motivated by the Fundamental Causes Theory with the intention of contributing to the research narrative of upstream factors that influence disparate health outcomes. Access to court ordered services is a positive step toward improving health and minimizing the consequences of the mass incarceration of Americans with behavioral health disorders. At the same time, we must provide additional context around our findings, which includes how the overreliance on the carceral system as a safety net behavioral healthcare provider has wrought historical trauma on Black Americans specifically and disrupted opportunities for building generational wealth [52,53,54]. In light of this narrative, using courts to promote treatment uptake and divert people with mental health problems away from incarceration must be done so as to not further perpetuate structural racism in the healthcare and criminal justice sectors. Promising evidence-based interventions like mental health and drug courts much be availed equitably to all who qualify. Our study identified growth in court ordered services on average across the US, but also shortfalls and entry points for health system improvement. Federal and state policies must reflect the need for new and expanded services, while maintaining a focus on health promotion among marginalized populations at the highest risk for poor health and criminal justice outcomes.

Limitations

Our definition of court-ordered services was broad as it was sourced from the N-MHSS questionnaire. The types of services offered within each facility in our large national sample were likely different, but the extent to how services varied is unknown. Despite this limitation, our study was the first to identify national trends about this important issue, and may be interpreted as an approximation or proxy for mental health treatment facilities that have any form of special attention paid to clients ordered to treatment by a court. To our knowledge, no national data exists that has a more granular or detailed definition of court ordered services. Future research may investigate differences in the types of treatment availed to court ordered clients at more regional or local levels.

N-MHSS includes data on court-ordered service availability, but there is no data on the number of patients who used the service. We stress the importance of not equating offering a service with utilization a service, and encourage interpreting study results in this light.

Conclusions

We identified overall growth in the proportion of mental health treatment facilities that offered services tailored to clients ordered to treatment by a court. There was wide variation by state, however, such that the availability of tailored services decreased or was stagnant in many states with large mental health and incarceration burdens. Other factors, in particular substance use treatment and rurality, were strongly associated with tailored services. The healthcare needs for individuals at risk for exposure to the criminal justice system are complex. Court ordered treatment is a promising strategy to improve health and reduce reliance on the carceral system as a healthcare provider. At the same time, we express caution around disparities within behavioral health courts and advocate for equity in access to incarceration alternatives.