Implementing evidence-based supported employment within specialty mental health probation: A hybrid type I implementation-effectiveness trial protocol

Abstract

Background:

People with serious mental illnesses who are involved in the criminal justice system face significant challenges in obtaining meaningful employment. Given the importance of employment in reducing recidivism, the field needs effective interventions to increase employment rates, address mental health needs, and reduce recidivism for justice-involved people with serious mental illnesses. Individual Placement and Support – Supported Employment (IPS-SE) improves employment outcomes among individuals with serious mental illnesses and has shown promising results when implemented with individuals with histories of justice involvement; however, IPS-SE has only been implemented in mental health service settings. Given lower levels of treatment engagement and completion among justice-involved populations, implementation of IPS-SE in specialty mental health probation (SMHP) is an opportunity to increase reach and engagement among justice-involved people with serious mental illnesses.

Methods:

This article describes a hybrid type 1 implementation-effectiveness study that aims to: (1) assess the implementation enablers and barriers, as well as the feasibility, appropriateness, and acceptability of IPS-SE embedded within SMHP; (2) identify the multi-level factors (i.e., implementation determinants) that influence IPS-SE implementation within the context of a probation setting; and (3) assess the impact of IPS-SE on employment – our primary endpoint – and the impact of IPS-SE on a number of secondary outcomes and potential treatment mechanisms. The efficacy arm of the study will be a randomized controlled trial of 130 adults on community supervision who will either receive treatment as usual or IPS-SE. The implementation arm of the study will examine implementation determinants and implementation outcomes using qualitative methods.

1. Introduction

In 2021, there were 1.7 million individuals incarcerated in our nation’s jails and prisons [1] and another 3.7 million on community supervision (i.e., probation or parole) [2]. Of those under community supervision, between 704,000 and 1.2 million had a mental health condition or serious mental illness, with best estimates suggesting that between 15% and 27% of people on probation have serious mental illnesses [3,4]. Evidence suggests that individuals with serious mental illnesses have chronic or cyclical patterns of reincarceration, resulting in higher recidivism rates compared to justice-involved individuals who do not have mental illnesses [5–7]. Repeated incarcerations disrupt housing and social support [8,9] – which are key components to maintaining mental health and stability in the community [10] – and can negatively impact individuals’ ability to sustain employment. Consequently, disrupting patterns of reincarceration by addressing the risk factors that lead to criminal justice involvement is essential to improve the well-being and psychosocial functioning of adults with serious mental illnesses.

Individual Placement and Support – Supported Employment (IPS-SE) is an evidence-based intervention shown to improve employment outcomes among individuals living with serious mental illnesses [11,12]. IPS-SE has also shown promising results when implemented with individuals with histories of criminal justice involvement [12], which is important given the fact that work is a protective factor against recidivism [13–16], the extremely high unemployment rate among individuals with serious mental illnesses [17,18], and evidence of evidence of financial insecurity among those with serious mental illnesses [19].

Despite the success of the IPS-SE model and the potential for improving outcomes for justice-involved individuals, IPS-SE has nearly always been implemented in community mental health settings. Given lower levels of treatment engagement and completion among justice-involved populations [20], implementation of IPS-SE solely within mental health service settings may result in limited reach with justice-involved populations. Thus, a likely intersection of social service and criminal justice interventions for individuals with serious mental illnesses are specialty mental health probation (SMHP) approaches.

Prototypical SMHP approaches include five primary elements: (a) probation caseloads consisting exclusively of people with mental illnesses; (b) reduced caseload sizes; (c) ongoing mental health training for SMHP officers; (d) a problem-solving supervision orientation; and (e) collaboration with internal and external resources to link people with supports [21]. The integration of a reduced caseload, problem-solving supervision orientation, and ongoing training have the potential to meet the unique and specific needs of individuals with serious mental illnesses and criminal justice involvement [22].

Evidence of SMHP’s effectiveness at increasing mental health and substance use treatment engagement and improving mental health symptoms is promising [23–25] and some studies have shown that SMHP is related to a decrease in violations, re-arrests, and jail days [23–26]. Intuitively, adding IPS-SE to SMHP has the potential improve employment outcomes for justice-involved adults with serious mental illnesses; however, there is no available evidence examining the feasibility or effectiveness of embedding IPS-SE within criminal justice settings.

To address this gap, this study focuses on three primary aims: (1) assess the implementation enablers and barriers, as well as the feasibility, appropriateness, and acceptability of IPS-SE embedded within SMHP; (2) identify the multi-level factors (i.e., implementation determinants) that influence IPS-SE implementation within the context of a probation setting; and (3) assess the impact of IPS-SE on employment – our primary endpoint – and the impact of IPS-SE on a number of secondary outcomes and potential treatment mechanisms. Although no studies to date have been conducted in mental health probation settings, IPS-SE has proven to be effective in a wide range of setting, increasing confidence in our study hypothesis.

2. Method

2.1. Design

The research team will implement a hybrid type 1 implementation effectiveness study [27] examining the effectiveness of IPS-SE embedded within SMHP and examining implementation determinants and outcomes. Hybrid effectiveness-implementation designs are meant to advance the research evidence for a given intervention while also focusing on the implementation context and dissemination. A hybrid type 1 effectiveness-implementation design means that a study has both effectiveness aims and implementation aims; however, the effectiveness aims are the primary outcome of study and the implementation aims often focus on factors that impact the implementation of the intervention. In the effectiveness arm of this study, eligible individuals who are engaged in SMHP will be recruited to participate in a randomized controlled trial of IPS-SE integrated with SMHP in two urban counties. Consenting participants will be randomly assigned to either SMHP or SMHP + IPS-SE. The effectiveness arm of the study will examine the impact of IPS-SE on client-level outcomes across multiple domains, including employment, mental health treatment, and criminal justice involvement. Data collection will occur for both the treatment and control groups at baseline, three, and six months. The implementation arm of the study assesses the primary implementation outcomes of acceptability, appropriateness, and feasibility [28] via data collection with study participants and those tasked with implementing IPS-SE. See Table 1 for measures and data collection time points.

Table 1.

Implementation and effectiveness measures (Aims 1, 2 and 3).

Measures	Source	T1	T2	T3	Alphas	Refs
Aim 1: Implementation
Screening and Enrollment Logs	Research staff	X			–	–
Field Notes	IPS-SE	X	X		–	–
Intervention Attendance	IPS-SE	X	X	X	–	–
Data Collection Interview Attendance	Research staff	X	X	X	–	–
Appropriateness, Acceptability, Feasibility	Subjects, IPS-SE, SMHP		X	X	0.79–0.85	[40]
Implementation Determinants	Subjects, IPS-SE, SMHP		X	X
Aim 2: Primary Outcomes
Employment	Subjects	X	X	X	–
Days Employed	Subjects	X	X	X	–
Exploratory Aim 3: Secondary Outcomes and Potential Treatment Mechanisms
Psychological Distress (Symptom Check List- 10R)	Subjects	X	X	X	0.77–0.90	[31]
Social Support (Duke- UNC FSSQ)	Subjects	X	X	X	0.90–0.94	[32]
Stigma (Internalized Stigma of Mental Illness; ISMI)	Subjects	X	X	X	0.92–0.96	[33]
Motivation to Work (researcher-adapted)	Subjects	X	X	X	–	[34]
Service Utilization (researcher-created)	Subjects	X	X	X	–	–
Activities of Daily Living (SF-12)	Subjects	X	X	X	–	[35]
Self-esteem (Rosenburg Self-esteem Scale; SES)	Subjects	X	X	X	0.86–0.92	[36]
Quality of Life (Lehman’s Quality of Life Scale)	Subjects	X	X	X	0.63–0.72	[37]
Demographic Data	Subjects	X			–	–

2.2. Regulatory approvals

This study was registered on ClinicalTrials.gov on July 2, 2021 (NCT04948853). The University of North Carolina at Chapel Hill (IRB) serves as the single IRB. The study was approved on May 24, 2021 (IRB # 21–0972).

2.3. Trial arms

2.3.1. Control arm – treatment as usual

The control group will receive usual care, which entails a county-specific employment referral resource guide which will be distributed by probation officers in each study county. Typical of a probation setting, subjects in the study’s control arm will be responsible for accessing employment resources and/or obtaining jobs on their own.

2.3.2. Experimental arm 1 - Individual Placement Support-Supported Employment (IPS-SE)

Subjects in this condition will receive services from a 1.5 FTE IPS-SE team that will work to provide one-on-one, person-centered services to help obtain employment, including but not limited to career profiling, resume assistance, job placement, training and other activities. IPS-SE is an individual, person-centered intervention designed to help individuals living with serious mental illnesses obtain and sustain meaningful employment. We will implement full-fidelity IPS-SE with a 1.5 FTE team (full-time supported employment specialist and a half-time team lead) such that fidelity standards of a 20:1 client-to-staff ratio is maintained at all times. The IPS-SE supervisor, consistent with fidelity standards, will carry a small caseload. The IPS-SE team lead will be a master IPS-SE training and fidelity specialist with significant years of experience with IPS-SE. The team lead will be responsible for hiring, training and providing ongoing supervision for a full-time employment specialist.

2.4. Study recruitment and sample

2.4.1. Recruitment for the randomized controlled trial

For the effectiveness arm of the study, subjects (n = 130) will be individuals on probation who have serious mental illnesses and who are supervised by SMHP officers in the two selected counties. Subjects will be identified and recruited via existing SMHP officers (n = 5). Each SMHP officer carries a caseload of up to 40 individuals with serious mental illnesses. In our SMHP settings, probationers who have diagnoses of schizophrenia, bipolar disorder, and/or depression are prioritized for SMHP, and SMHP officers confirm these diagnoses through brief mental health screening, evaluation of prison mental health records, when applicable, and confirmation with community mental health treatment providers. SMHP officers will refer individuals on their caseloads who are interested in employment and who express interest in participating in the study to the research staff. Eligibility criteria include: (1) on probation and supervised by a SMHP officer in one of the two selected counties; (2) aged 18–99; (3) interested in obtaining employment; and (4) competent to provide informed consent. Participants will receive a $10 gift card for participating in screening and a $30 gift card for enrolling in the study, as well as additional $30 gift cards for participating in follow-up data collection at 3 and 6 months.

In both of the selected county probation agencies, individuals on probation with serious mental illnesses will be referred to the study team by their mental health probation officers. Probation officers will be asked to distribute study fact sheets and alert potential participants that, if interested, members of the research team are available to discuss the study in greater detail. Research staff will work with probation officers to ensure the officers are not coercive and that the officers clearly explain that the study is voluntary, has no bearing on a potential participant’s legal or probation status, and that they can quit the study at any time without negative consequences. The research team will coordinate with the probation officers to be on site when potential participants appear for their regularly scheduled probation visits.

For those who are eligible for and interested in the study, study fact sheets will be used to explain the study and provide more information. The research team will explain that the study is voluntary and that participation can be discontinued at any time without consequences. Potential participants will be given a brief competency screen to assess eligibility. For those who are not eligible or who do not want to participate in the study, a $10 gift card to a local vendor will be offered. Reasons for refusing to participate will be documented at this point; however, no identifying information will be collected. For those who are eligible to participate and agree to volunteer for the study, informed consent will be obtained, demographic information will be collected and a set of standardized measures will be completed with each subject. Research staff will administer all data collection instruments to ensure standardization. Baseline and follow-up research interviews will be conducted in probation settings, unless otherwise directed by the study participant. To minimize burden on study subjects and optimize convenience, research interviews and data collection will be aligned with subjects’ routine probation officer visits (Fig. 1).

Fig. 1.

Enrollment, intervention allocation, follow-up and data analysis.

We will use a sophisticated recruitment and informed consent process, which we have developed and strengthened through years of research with high-risk populations of individuals with mental illnesses in criminal justice settings. Our protocol will entail detailed study description and information with all materials written at a 5th grade reading level (to the extent possible); consent protocols that will be conducted privately and are free of coercion, which is critical given our study setting and population; ample time for potential study subjects to make an informed decision; well-trained research staff who will assess competence to provide informed consent and who have full authority to discontinue the consent process in instances where competence is in question; and we will use a well-established study competency quiz that potential subjects must pass, which indicates to the research team that the subject has a full understanding of the risks, potential benefits and all study activities, i.e., random assignment, follow-up, in order to be consented.

We anticipate our three-year study will begin in late 2019 or early 2020 and end in the fall of 2022. Recruitment will commence once all institutional review board approvals are obtained and our study procedures are approved by probation. Rolling recruitment, data collection and follow-up will last 30 months, barring unforeseen delays and our subject recruitment period will begin in early 2020 and end in the spring of 2022, ensuring that all subjects can be followed for six months after baseline data collection.

The IPS-SE service will be delivered by an independent community mental health agency external to the probation setting. After recruitment, consent, enrollment and randomization to SMHP+IPS-SE, the probation officer will be informed of the randomization result and informed, in the case the probationer is randomized to IPS-SE, that the IPS-SE team will immediately be in contact with the subject to begin services. A member of the research team will attempt to contact the IPS-SE team prior to concluding the research interview to facilitate a “warm hand-off,” so the IPS-SE employment specialist can begin to develop a rapport with the subject. Thus, the probation officer will not be asked nor will be required to initiate services with the IPS-SE staff. Individuals who are randomized to receive IPS-SE will be retained in services as long as the participant wishes to engage with services, per fidelity to the model; however, the duration of study participation is six months. Also, per fidelity to the model, IPS-SE participants can choose to terminate once if/when suitable employment is obtained.

2.4.2. Randomization

A locked box with pre-assigned envelopes each containing either IPS-SE or TAU assignment cards ordered by a table of random numbers will be used to assign subjects to each condition. Simple randomization procedures will be employed to ensure an equal distribution of allocations to each condition. Once informed consent is obtained for a subject and baseline data collection has been completed, research staff will open the next envelope in the box to assign the participant to a study condition.

2.4.2.1. Recruitment for the implementation study.

For the implementation arm of the study, SMHP officers, chief probation officers and administrators, employment providers, intervention staff, and participants in the treatment arm will be invited to participate in semi-structured interviews pertaining to implementation determinants (i.e., multi-level factors that impact intervention implementation [29,30], acceptability, appropriateness, and feasibility [28]. Participants will be selected using purposive sampling based on their involvement in intervention implementation.

To examine appropriateness and acceptability from the participant perspective, the research team will invite study participants from the treatment group to answer additional qualitative questions during their 3-month follow-up appointment. Questions pertain to experiences and barriers seeking employment prior to the intervention, description of work with the employment specialist, including which supports were or were not helpful, ways in which the intervention was or was not beneficial, perceptions of communication between the employment specialist and probation officer, and any changes recommended to the employment intervention.

2.5. Measures

Implementation measures, effectiveness measures, and exploratory measures and potential clinical mechanisms are presented in Table 1 below.

The implementation study will examine feasibility, acceptability, and appropriateness of the IPS-SE intervention, as well as implementation determinants (i.e., factors that impacted the successful implementation of the model). Table 1 provides an overview of the constructs measured, description of data collection measures, and data collection time points.

2.5.1. Implementation determinants

The research team will use the Exploration, Preparation, Implementation, and Sustainment (EPIS) [38] framework to guide the methods for assessing implementation determinants associated with the inner, outer, and inter-organizational context of IPS-SE. To examine the implementation challenges associated with IPS-SE, the research team will conduct semi-structured qualitative interviews (n = ~20) with individuals involved in the implementation of IPS-SE. Qualitative interviews will be conducted at two time points, first during the study’s preparation phase [38] prior to implementation and again within approximately eight months to one year during the implementation phase of the study.

Questions on the interview guides will assess implementation determinants [29,30] – multi-level (e.g., individual, organizational, and policy) factors that impact implementation outcomes. Given the inter-organizational and multi-level nature of SMHP, the interview guides will incorporate questions related to factors associated with the outer context (e.g., service environment), the inner context (e.g., leadership, staffing), bridging factors (e.g., state and local policy), and characteristics associated with the IPS-SE intervention [38]. Results from the interviews will also be used to describe how IPS-SE was implemented within this setting and the role of the employment specialist.

2.5.2. Implementation outcomes

Proctor et al.’s [28] taxonomy of implementation outcomes will be used to guide data collection and analysis to assess acceptability, appropriateness, adoption, and feasibility. To examine acceptability, appropriateness, and feasibility, the research team will conduct a series of interviews (n = 55) with potential employers (n = 20), individuals on probation (n = 20), and probation officers as well as middle- and upper-level managers (n = 15). Acceptability is defined as the degree to which IPS-SE is agreeable or satisfactory among potential employers, people with serious mental illnesses on probation, and SMHP officers. Appropriateness is defined as the compatibility of IPS-SE within the study setting (i.e., probation). Feasibility is the degree to which IPS-SE can be successfully embedded within SMHP. Interview guides will be tailored to the subsample (i.e., employers, officers, people with severe mental illnesses who are on probation) and will examine perceptions related to: (1) partnership and collaboration between mental health services, probation, and employment services; (2) integration of employment, criminal justice, and mental health domains within a probation setting; and (3) satisfaction with the model.

2.5.3. Primary endpoint – effectiveness study

As shown in Table 1, our primary endpoint is employment, which will be assessed at 3- and 6-month intervals using the following items and response patterns: (1) current employment status - employed full time (35+ hours per week), employed part-time, unemployed (but looking for work), unemployed (disabled); unemployed (volunteer work), unemployed (retired), unemployed (not looking for work), and other (specify); (2) employed at or above the minimum wage (Y/N); (3) monthly income; (4) number of days worked in the past month; and (5) level of job satisfaction - very dissatisfied, dissatisfied, neither satisfied nor dissatisfied, satisfied, and very satisfied.

2.5.4. Secondary outcomes and potential mechanisms

Also shown in Table 1, we will collect data on a number of secondary outcomes and potential treatment mechanisms to gain a better understanding of the impact the IPS-SE on mental health well-being, social support, internalized stigma, motivation to work, service utilization, activities of daily living, self-esteem and quality of life.

2.5.5. Data safety and monitoring

This is a randomized controlled trial with a relatively small number of subjects; thus, in lieu of a Data and Safety Monitoring Board, the principal investigator will perform the monitoring function as part of the general oversight and scientific leadership of the study. The risk of the intervention is relatively low. Any unexpected effects of study participation will readily become apparent to the principal investigator through close monitoring of individual study subjects and communication with North Carolina Department of Adult Corrections (NCDAC) and IPS-SE staff. The principal investigator will comply with all requirements for prompt reporting of any serious and unexpected adverse unanticipated problems involving risks to subjects or others. The principal investigator will report any adverse events (AEs) to the local IRB at the University of North Carolina at Chapel Hill within 24 h of their occurrence. The IRB will evaluate these AEs and determine whether any additional information or protocol changes are needed.

2.6. Data analysis and statistical power

Initial analyses will include calculation of means, standard deviations, and the distributions of variables, as well as the correlations among them. Then, for the effectiveness arm of the study, the research team will use bivariate analyses to determine whether randomization resulted in balanced groups (i.e., that the treatment group and the control group are balanced on all relevant measures). If the groups are not balanced, the research team will use regression analyses to hold constant demographic and/or other variables, as needed.

Next, to examine our primary outcome, we will use the chi-square test to examine between-group differences in the proportion of subjects who are employed at 3- and 6-month intervals, separately at each interval. More sophisticated analyses for this feasibility study and a subsequent study will require analytic methods for longitudinal binary and continuous data, which will require the need to address correlated errors and dependence of observations.

To examine our secondary outcomes (e.g., scores on standardized measures), repeated-measures one-way ANCOVAs and repeated-measures multiple ANCOVAs will be used to examine group, time, and group X time interaction effects. Here, hierarchical linear models (HLM) for continuous variables will be used. In addition, our analysis plan will include the comparison of study completers vs. dropouts on baseline measures as well as dropout rates among treatment and control subjects.

The point of our effectiveness aim is to test our hypothesis that SMHP+IPS-SE results in superior employment outcomes, compared to those who receive SMHP and usual care, for individuals on mental health probation with serious mental illnesses. In addition, we will assess the feasibility of a randomized controlled trial of IPS-SE in a probation setting; collect data on available sample size, sample attrition and sample retention and the feasibility of collecting the data and information necessary assess the impact of IPS-SE on our primary endpoint; assess the impact of IPS-SE on potential clinical mechanisms; and generate the information needed to conduct comprehensive power calculations for a larger R01 study. Our sample for a pooled analysis has the potential to be large given that, assuming a 20% attrition rate, based on the research team’s experience with the study population, averaged across the outcome assessments at every three months, the HLM and HGLM analyses described above would include approximately 317 observations (130 baseline +104 + 83). Given evidence that IPS-SE results in employment rates of up to 50% for individuals with serious mental illnesses, compared to 15% or lower for individuals who participate in traditional vocational services or no employment services, a traditional two-tailed power calculation for a two-sample comparison of proportions suggests that overall, with our anticipated 6-month sample, we would have approximately 80% power to detect an effect size over 0.45.

There are two research objectives for the implementation arm of the study: (1) assess the feasibility, appropriateness, and acceptability of IPS-SE embedded within SMHP; and (2) identify the multi-level factors (i.e., implementation determinants) that influence IPS-SE implementation within the context of a probation setting. To address these objectives, the research team will use directed content analysis [39] guided by EPIS [38] and Proctor et al.’s taxonomy of implementation outcomes [28]. First, the coding team will read five transcripts to become familiar with the data. Next, the coding team will develop a priori codes informed by the two guiding frameworks. Specifically, definitions for the relevant implementation outcomes – feasibility, appropriateness, and acceptability – will be derived from Proctor et al.’s taxonomy of implementation outcomes. These definitions will guide the development of an initial codebook pertaining to implementation outcomes. Next, the research team will develop an initial set of codes for the second implementation research objective pertaining to implementation determinants, specifically the multi-level factors that influence IPS-SE implementation. The coding team will develop an initial set of codes informed by the EPIS constructs – outer context, inner context, bridging factors, and innovation factors. The coding team will then independently code a small sample of interviews, meet to discuss potential changes to the codebook, and continue to revise until agreement about the codebook is reached. Then, the coding team will independently code a sample of interviews and a senior research team member will calculate inter-rater reliability. Once IRR of 0.9 is reached, the coding team will continue to independently code transcripts. Analyses will be conducted using Dedoose (Dedoose, 9.0.9).

3. Discussion

People with serious mental illnesses who are involved in the criminal justice system face significant challenges in obtaining long-term, gainful employment. Given the importance of employment and financial security in reducing recidivism, and the challenges faced by individuals with serious mental illnesses in these areas, the field needs effective interventions to increase employment rates, address mental health needs, and reduce recidivism. IPS-SE is an evidence-based practice that, if embedded within a criminal justice context, has the potential to reach the individuals most in need of employment supports.

To our knowledge, this study will be the first to embed IPS-SE within a probation setting. This study will provide the foundation for subsequent IPS-SE studies within probation and other criminal justice settings. Specifically, using these results will help inform additional adaptations needed to enhance feasibility, acceptability, and appropriateness of the model. Given that employment is a primary requirement for individuals who are on probation and the difficulties with which individuals living with serious mental illness have in obtaining and sustaining meaningful employment, our study, if successful, will illuminate the implementation enablers and barriers to implementing IPS-SE in a criminal justice setting and provide preliminary evidence that IPS-SE embedded within a SMHP setting has the potential to improve employment, mental health, and criminal justice outcomes for a high-risk population.

Moreover, the criminal justice system is an important, albeit unfortunate, setting within which mental health services research must be conducted in order to serve those who are most vulnerable among those living with mental illnesses, and a setting in which to equip providers who work at the nexus of the mental health and criminal justice systems with the practices and skills needed to optimize outcomes for justice-involved individuals with mental illnesses. To our knowledge, there have been no studies of any interventions integrated with specialty mental health programs, which, in our opinion, positions our hybrid type I implementation effectiveness study well to contribute to the field. Moreover, we selected an employment intervention given one of the primary requirements of probation is employment, independent of mental illness, and employment is an area where individuals with serious mental illnesses have considerable difficulty.

In addition, knowledge regarding implementation determinants will inform the development of implementation strategies aimed to enhance the uptake of core components of IPS-SE within the probation setting. In subsequent studies, the research team will test the feasibility of the implementation strategies within the context of a larger hybrid type II study in which IPS-SE effectiveness outcomes and implementation strategy efficacy will be co-primary aims [27].

4. Limitations

Potential problems of our study could include failed randomization, missing data, differential attrition and treatment contamination given we are operating in the same probation agency, albeit across two counties, with the same probation officers serving individuals in both our treatment and control groups. If randomization should fail, as stated above, we will use multivariate statistical techniques that will allow us to mitigate group imbalance. Missing data can occur for a variety of reasons and could be a serious potential source of bias for our future work. We will closely assess missing data and address missing data as needed in our analyses. If applicable, multiple imputation strategies will be used.

There is a potential threat to validity in our study design (i.e., probation officers providing supervision to subjects who are in treatment and control conditions in the same setting), in that probation officers could attend more to employment issues for their probationers whom they are supervising but who are in the usual care arm of the study, i.e., not receiving IPS-SE. Given our study team will have close contact with our participating probation officers, we will gain an ongoing understanding of how officers are supervising their probationers with regard to employment. The intervention requires close work with the IPS-SE team, which could not be replicated through any other means. Moreover, with respect to the qualitative interviews with probation officers, which are a part of our implementation study activities, we will explore if and how supervision behaviors with regard to employment changed as a result of their probation participating in the study. This information will be critical to our future work and research regarding the implementation of IPS-SE and other evidence-based practices in criminal justice settings.

Moreover, our primary limitation here is the use of a small local sample, in one state, in one probation agency to conduct the proposed study; thus, the generalizability of our findings will be uncertain. However, one of our primary goals here is to establish the feasibility of our approach and information needed for the design of a practical RCT for future research, which can be accomplished with a single-site study. It is important to note that key aspects of employment were omitted from our employment outcomes, including: (a) time to first job; (b) hours worked per week; (c) total weeks worked; and (d) number of jobs held. This is our initial study of IPS-SE integrated within a mental health probation setting and our future work will include a more robust assessment of employment consistent with the IPS-SE literature. In addition, the IPS-SE team lead, who is an experienced IPS-SE trainer and conducts IPS-SE fidelity assessments nationwide, will be closely involved in the conceptualization of the study and the implementation of IPS-SE throughout. The team lead will ensure and monitor implementation fidelity and will provide information about IPS-SE fidelity and the implementation of IPS-SE in a mental health probation study during the qualitative interviews at the study’s conclusion; however, we do not have an external fidelity review built into this phase of our research and the lack of quantitative fidelity data will be a significant limitation, which we will address when study results are presented, and flaw in our design, which will be addressed in our future research. Replications will be essential for establishing generalizability in any research and look forward to our future research, and in supporting others who want to engage in similar work, to verify and extend our findings and their implications for mental health services research.

The expected outcomes of this hybrid clinical trial include: (1) an understanding of the implementation barriers and facilitators associated with providing IPS-SE in a probation setting; (2) an understanding of how IPS-SE needs to be adapted for people with mental illnesses who are on probation; and (3) evidence that SMHP+IPS-SE can improve employment outcomes among people with mental illnesses who are on probation. This research has the potential to promote the adoption and dissemination of best practices at the interface of the mental health and criminal justice systems and promises to grow the evidence for interventions that improve employment and reduce economic inequalities among a vulnerable population.

Data availability

Data will be made available on request.