Focus of the intervention
Mental health courts are designed to divert offenders suffering from severe mental health issues, such as schizophrenia, major depression and bipolar disorder towards treatment options rather than imprisonment. Customised sentences are given depending on the individual’s needs, with community supervision by a case manager to ensure that treatment is completed. The average length of treatment given as a sentence is 12-18 months. Upon satisfactory completion, graduates may be discharged from the programme and their criminal record for that offence may be removed.
This summary reports the findings of six systematic reviews:
- review one was based on 18 studies
- review two was based on 17 studies
- review three was based on 30 studies
- review four was based on 20 studies
- review five was based on 20 studies
- review six was based on 20 studies
The primary studies in the reviews identify a reduction in reoffending through a variety of measures, including rearrests, new charges, new convictions, time in prison, contact with the police and booking rates (booking refers to a process where suspected criminals are taken into police custody after arrest). All the primary studies are based on evidence from the USA, except for one study in review four which was conducted in Australia.
Effect – how effective is it?
Overall, the evidence suggests that mental health courts have reduced crime.
All the reviews found some evidence that mental health courts reduce reoffending. Reviews one, two, three and four found there was significantly lower levels of reoffending among participants in mental health courts compared to control groups, though the reviews did not test the effect based on different crime types. The rate at which mental health courts lowered levels of reoffending differed between reviews; review one found a 54% decrease and review two found a 20% decrease.
Review one found little difference in effect size between high-quality and low-quality studies, despite those published in academic outlets having stronger effect sizes than unpublished studies. Conversely, review two reported that low-quality studies produced statistically significant effects of mental health court participation on reoffending, while moderate and high-quality studies produced positive but non-significant effects. Similarly, academic dissertations yielded stronger effect sizes than peer-reviewed publications and reports.
Only review two found that a follow-up length of 12 months produced similar effects to those of longer follow-up periods, suggesting sustained reductions in reoffending over time.
How strong is the evidence?
Review one is a meta-analysis of 18 studies. The review was sufficiently systematic that most forms of bias that could influence the study conclusions could be ruled out.
Review one had a well-designed search strategy, included unpublished literature and used appropriate statistical methods in the analysis of effect size. However, biases remained within the primary studies, including dropout rates, the fact that some studies used non-completers as control groups, and how the study authors dealt with non-compliance or termination of programmes.
Review two is a meta-analysis of 17 studies and was also sufficiently systematic that most forms of bias that could influence the study conclusions could be ruled out. Reviews three and four accounted for many forms of bias. Reviews five and six were systematic but some forms of bias that could influence the study conclusions remain.
Mechanism – how does it work?
Reviews one, three, four and five suggest that mental health courts may work by focusing on rehabilitation. This includes the notion that the law can be applied in a therapeutic way and aid individuals (a non-therapeutic way can be unhelpful or even harmful).
In applying the law ‘therapeutically’, mental health courts focus on rehabilitation rather than punishment. By positively affecting the offender’s quality of life, and providing the treatment they need, this may prevent future crimes. Review two suggested that a compassionate judge-client relationship is the key to this, alongside treating participants with respect, engagement, and actively listening to them. None of the reviews tested whether this suggested mechanism had influenced the outcomes, or looked at whether the type or severity of the mental health condition makes any difference to the outcome.
Review five suggests that clinical and psychosocial factors can cause criminal behaviour, and that participating in monitored mental health treatment increases individuals’ connections and adherence to behavioural health services, thereby improving symptoms and reducing reoffending. However, the review was not able to show that mental health courts improve the frequency and consistency of behavioural health services. Further, of the 20 reviewed articles, just over a third assessed the impact of mental health courts on overall psychiatric functioning and only one study assessed quality of life. The lack of research measuring these outcomes prevents any conclusions about the effectiveness of mental health courts for these outcomes.
Review three suggests that mental health courts support offenders by offering individualised treatments which address their underlying needs, and connect them with services which treat, rather than worsen their mental health issues.
Moderators – in which contexts does it work best?
Review three examined several moderators including participant age, race and gender. Mental health courts show similar effects for youths and adults, and similar effect sizes were seen when studies controlled for race and gender.
Reviews one, four and five highlighted several potential moderators including the potential importance of participant race. However, all three reviews noted that most of the participants in the programmes were white males in their thirties, which was unrepresentative of the US prison population with over one third African-American males. In terms of gender, only one study in review four found a statistically significant difference between participants assigned to the treatment and control groups.
Reviews one, four and five also suggested that programme completion might be a moderator as treatment dosage could be important. Review six mentioned that four out of seven primary studies reported significantly lower re-arrest rates among offenders who successfully completed a mental health court programme compared to those who were partially successful or terminated from the programme and returned to a traditional criminal court.
Review four also noted that participants with multiple disorders (for example, substance misuse problems) are significantly less likely to graduate from treatment programmes than those with a single disorder.
Review two noted a moderating factor relating to outcome measures – primary studies that measured reoffending after mental health court exit compared to after enrolment showed statistically significant and stronger effects on reducing reoffending.
Review five states that the following additional characteristics may predict the likelihood of lower reoffending rates: having a more serious offence upon mental health court admission, having lower pre–mental health court arrest and imprisonment days, mental health treatment on intake, no substance abuse history, and a diagnosis of bipolar disorder rather than schizophrenia or depression.
Review five suggests the following additional characteristics may predict poorer reoffending outcomes: the presence of a substance abuse disorder, having higher lifetime arrest or imprisonment day rates, engaging in more serious criminal behaviours historically, being less educated, and reliance on disability benefits.
However, none of the reviews tested the effect of these moderators on reoffending.
Implementation – what can be said about implementing this initiative?
There is no consensus about what constitutes a mental health court, and the implementation varies depending upon the procedures in their location and the available treatments in the area. Nevertheless, there are some common components discussed by the reviews.
- Participants must report to the court on a regular basis to discuss their progress or revise their treatment plans, with sanctions being applied for noncompliance (reviews one and four).
- All participation is voluntary, and graduation happens once participants meet their individual goals and are able to handle their mental illness. In some cases, monthly drug screening is required for participants who also have a substance related disorder (reviews one and four).
- Review six suggested that mental health courts are characterised by three key components: screening offenders with a mental illness, an assessment done by a mental health professional, and negotiation between court diversion and criminal justice staff on reduced or waived charges.
Reviews one and six noted several factors that were associated with the successful implementation of mental health courts:
- Review one suggested good relationships between judges, court personnel and the participants, and the fact that judges acted more like case managers than traditional judges, were also enablers to successful implementation. Consistency was critical throughout the process, and success was related to the quality of services provided to participants.
- Review six suggests that case managers or access to vocational and housing services may be important components of mental health courts which successfully reduced offender re-arrest rates.
One potential barrier noted by review one was that a lack of staffing affected compliance rates by participants.
Economic considerations – how much might it cost?
While none of the reviews conducted a full cost-benefit analysis, some mention of costs was reported in a primary study in review one. This estimated that a mental health court programme in Pennsylvania saved approximately $3.5 million over a period of two years ending in 2007. This primary study specifically noted that mental health courts had the potential to decrease the cost of the most expensive forms of treatment that participants would otherwise face, such as hospitalisation.
General considerations
- Overall, the evidence suggests that mental health courts have reduced crime.
- All studies included in the reviews were conducted in the USA (except one primary study conducted in Australia), so care must be taken when transferring findings to the UK context.
- There is some overlap of studies included in each of the reviews. For example, review four has some primary studies in common with review two, and review five lists review one as one of the included studies.
- Mental health courts require participants to plead guilty, and there are questions as to whether individuals with mental health issues are able to fully understand the implications of this, and provide informed consent to their participation.
- Mental health courts are difficult to research due to the issue of a high degree of confidentiality in participant records.
- There were difficulties with how to deal with individuals who refused to participate in mental health courts – arguably those who chose to participate were more likely to wish to receive treatment and were therefore not necessarily comparable to those who refused. Despite this, some primary studies used those who refused treatment as part of a control group.
- Bias caused by the primary study designs was a common finding across the reviews – a majority implemented either post-test only control group, static group comparison, or one group pre-test post-test designs.
Summary
Overall, the evidence suggests that the intervention has reduced crime. Participants in mental health courts had significantly less reoffending than control groups, though questions remain about the composition of these control groups in the primary studies. Successful mental health courts were those where good relationships were maintained between the judge and participants.
Reviews
Review one
Reference
Sarteschi CM, Vaughn MG and Kim K. 2011. 'Assessing the effectiveness of mental health courts: A quantitative review', Journal of Criminal Justice, 39, pp 12–20
Review two
Reference
Lowder EM, Rade CB and Desmarais SL. 2018. ‘Effectiveness of mental health courts in reducing recidivism: A meta-analysis’, Psychiatric Services, 69(1), pp 15–22.
Review three
Reference
Fox B and others. 2021. ‘Assessing the effect of mental health courts on adult and juvenile recidivism: A meta-analysis’, American Journal of Criminal Justice, 46(4), pp 644–664.
Review four
Reference
Cross BL. 2011. ‘Mental health courts’ effectiveness in reducing recidivism and improving clinical outcomes: A meta-analysis’, Masters dissertation, University of South Florida.
Review five
Reference
Honegger LN. 2015. ‘Does the evidence support the case for mental health courts? A review of the literature,’ Law and Human Behavior, 39 (5), pp 478–488.
Review six
Reference
Loong D and others. 2019. ‘The Effectiveness of mental health courts in reducing recidivism and police contact: A systematic review’, Community Mental Health Journal, 55, pp 1073–1098.