The Availability and Utility of Services to Address Risk Factors for Recidivism among Justice-Involved Veterans

Abstract

The availability and utility of services to address recidivism risk factors among justice-involved veterans is unknown. We explored these issues through qualitative interviews with 63 Specialists from the Department of Veterans Affairs’ (VA) Veterans Justice Programs. To guide the interviews, we utilized the Risk-Need-Responsivity (RNR) model of offender rehabilitation. Specialists reported that justice-involved veterans generally have access to services to address most RNR-based risk factors (substance abuse; lack of positive school/work involvement; family/marital dysfunction; lack of prosocial activities/interests), but have less access to services targeting risk factors of antisocial tendencies and associates and empirically-based treatments for recidivism in VA. Peer-based services, motivational interviewing/cognitive-behavioral therapy, and Veterans Treatment Courts were perceived as useful to address multiple risk factors. These findings highlight potential gaps in provision of evidence-based care to address recidivism among justice-involved veterans, as well as promising policy-based solutions that may have widespread impact on reducing recidivism in this population.

Introduction

Justice-involved veterans, defined as those detained by or under the supervision of the criminal justice system, comprise a non-trivial proportion of the U.S. correctional population. Eight percent of individuals who are incarcerated in federal and state prisons and city and county jails are veterans of the U.S. military–approximately 181,500 veterans (Bronson, Carson, Noonan, & Berzofsky, 2015). Notably, incarcerated veterans likely represent only a fraction of the total number of justice-involved veterans nationally, given that approximately 75% of the correctional population in the U.S. resides in the community on probation or parole (Glaze, 2011).

Many veterans are caught in a cycle of contact with the criminal justice system. For example, the majority of incarcerated veterans have at least one prior episode of incarceration (Bronson et al., 2015; Noonan & Mumola, 2007), and, in fiscal year 2012, justice-involved veterans who were in contact with the Department of Veterans Affairs’ (VA) Veterans Justice Programs reported an average of eight prior arrests in their lifetime (Department of Veterans Affairs, 2012). Consequently, reducing justice-involved veterans’ risk for recidivism is a top priority for a number of national organizations, including the VA (Blue-Howells, Clark, van den Berk-Clark, & McGuire, 2013), the Bureau of Justice Statistics (Bronson et al., 2015), and the Substance Abuse and Mental Health Services Administration (Peters, Bartoi, & Sherman, 2008). Nonetheless, key questions remain regarding the extent to which empirically-supported risk factors for recidivism are routinely addressed in the care of justice-involved veterans, and whether these veterans have access to empirically-based treatments (EBTs) for reducing recidivism (Blonigen et al., 2014; Timko et al., 2014).

Empirically-supported principles for reducing recidivism among justice-involved adults

The empirical literature on offender rehabilitation indicates that successful reintegration of justice-involved adults into the community, particularly in terms of reducing risk for recidivism, is maximized through adherence to the principles of the Risk-Need-Responsivity (RNR) model (Andrews & Bonta, 2010a, 2010b; Andrews & Dowden, 2006; Blanchette & Brown, 2006; Ward, Melser, & Yates, 2007). The Risk principle states that level of service for justice-involved adults should be matched to their level of risk for reoffending. The Need principle states that case management efforts and intervention targets for justice-involved adults should focus on dynamic (i.e., modifiable) risk factors of recidivism (Andrews & Bonta, 2010a, 2010b). The Responsivity principle states that interventions to reduce recidivism should be tailored to the unique characteristics of each individual, such as their learning style, intellectual ability, or other relevant demographic factors.

Research related to the Risk and Need principles has outlined the Central Eight risk factors for recidivism (“criminogenic needs”), which are divided into the Big Four and Moderate Four. The Big Four–criminogenic needs with the strongest evidence of predicting future recidivism among justice-involved adults–include history of antisocial behavior (the frequency and variety of offenses); antisocial personality pattern (e.g., poor impulse control; sensation seeking); antisocial cognitions (e.g., blame externalization); and antisocial associates (close affiliations with law-breaking peers). The Moderate Four–criminogenic needs that significantly predict future recidivism, but less strongly than the Big Four–include substance abuse; family/marital dysfunction; lack of positive school or work involvement; and lack of prosocial/recreational activities. With exception of history of antisocial behavior, the Central Eight are conceptualized as modifiable risk factors of recidivism among justice-involved adults (Andrews & Bonta, 2010a, 2010b). Regarding justice-involved veterans, the Central Eight also appear applicable to the prediction of recidivism risk in this subgroup of offenders (Blonigen et al., 2014); however, there is some evidence of veteran-specific criminogenic needs, which are not included in the RNR model, but which are highly prevalent among veterans – e.g., homelessness (Greenberg & Rosenheck, 2008); PTSD (Sadeh & McNiel, 2015).

In terms of the Responsivity principle, interventions aimed at “general responsivity” utilize cognitive-behavioral treatments and strategies, which have been shown to be the most efficacious in terms of reducing risk for recidivism among justice-involved adults (Milkman & Wanberg, 2007; Wilson, Bouffard, & MacKenzie, 2005), particularly interventions that target Big Four risk factors related to antisociality (Aos, Miller, & Drake, 2006). Three cognitive-behavioral interventions have demonstrated empirical support for reducing risk for recidivism among justice-involved adults: Moral Reconation Therapy (MRT; Little & Robinson, 1988); Reasoning and Rehabilitation (Ross, Fabiano, & Ross., 1986); and Thinking 4 a Change (T4C; Bush, Glick, Taymans, & Guevara, 2011). In general, all three focus on development of social and cognitive skills such as self-control and problem solving and therefore aim to modify antisocial personality patterns and cognitions, but are also theorized to address the full range of criminogenic needs from the RNR model (Blodgett, Fuh, Maisel, & Midboe, 2013). Of these interventions, MRT has the most extensive evidence base, with meta-analyses demonstrating consistently greater reductions in rates of recidivism among MRT participants than control participants (Aos et al., 2006; Ferguson & Wormith, 2012; Little, 2005), as well as reductions in substance abuse and improved employment outcomes (Anderson, 2002; Fuller, 2003). Meta-analyses of Reasoning & Rehabilitation studies, which include some randomized trials, have demonstrated significant positive effects for reducing recidivism (Aos et al., 2006; Tong & Farrington, 2006; Wilson et al., 2005). Finally, T4C has received less empirical examination than MRT or Reasoning & Rehabilitation, but has been shown to reduce risk for recidivism in multiple studies (Lee et al., 2012; Lipsey & Cullen, 2007). Importantly, prior reviews have highlighted other cognitive-behavioral therapies (e.g., social skills training and relapse prevention) with demonstrated effectiveness for improving offender outcomes (Milkman & Wanberg, 2007). In our examination of “EBTs” in the current study, we focused on MRT, R&R, and T4C, given that a recent structured evidence review of justice-involved veterans cited these three interventions as having the most empirical support for reducing recidivism per se (Blodgett et al., 2013).

The present study

Recent reviews have highlighted gaps in our understanding of how to best reduce risk for criminal recidivism among justice-involved veterans (Blonigen et al., 2014; Timko et al., 2014). Chief among these gaps is a lack of knowledge regarding (1) the availability of services for justice-involved veterans that directly address empirically-supported risk factors of recidivism, as operationalized by the Central Eight (Andrews & Bonta, 2010b); (2) the types of treatment options and resources that are perceived as being most helpful in addressing these risk factors; and (3) whether, and in what contexts, justice-involved veterans have access to EBTs for recidivism risk (i.e., MRT, Reasoning & Rehabilitation, and T4C). From a policy and resource allocation standpoint, such information is critical in order to identify potential gaps in the provision of services and evidence-based care for recidivism risk factors among justice-involved veterans and, in turn, build more effective service delivery mechanisms for this vulnerable population (Taxman, 2014). Relative to other groups of justice-involved adults, the availability and nature of services to address recidivism risk may differ for justice-involved veterans, given that a substantial and growing number of these veterans receive services through the Veterans Health Administration (Blue-Howells et al., 2013; Finlay & Rosenthal, 2015).

The present study sought to address these gaps through qualitative interviews with Specialists from the VA’s Veterans Justice Program (VJP). VJP is a national VA-based outreach and linkage service for justice-involved veterans (Blue-Howells et al., 2013). The primary role of VJP Specialists is to link justice-involved veterans to VA and community-based services to address the health (physical and mental) and psychosocial needs of these veterans. The ultimate goal of the VJP is to facilitate the successful reintegration of justice-involved veterans into the community and end their cyclical contact with the criminal justice system (Clark, McGuire, & Blue-Howells, 2010). VJP comprises two programs: Veterans Justice Outreach and Health Care for Reentry Veterans. The Veterans Justice Outreach program operates at VA medical centers nationwide and provides outreach, assessment, linkage to care services, and time-limited case management to justice-involved veterans in city and county jails and drug and mental health treatment courts (e.g., Veterans Treatment Courts; Clark et al., 2010). The Health Care for Reentry Veterans program provides outreach, assessment, re-entry planning, linkage to care services, and time-limited case management with veterans in state or federal prisons, and up to four months after release from these facilities. This program has served justice-involved veterans in 81% of US prisons. In order to effectively accomplish their duties, VJP Specialists work very closely with justice system personnel. This collaboration reflects both the clear distinction between the role of the justice system and that of the VA, as well as the necessary collaboration across these systems. With respect to role distinction, it is the justice system that determines what will be required of each veteran based on their legal status, whereas VA makes available those services deemed to meet the needs of and requirements for the veteran once released from custody. In accomplishing this, there is frequent and ongoing communication between VJP Specialists and justice system personnel (e.g., corrections staff re-entry planners; court team case managers; parole and probation officers) who are similarly involved in the assessment of and linkage to needed services for a justice-involved veteran once out of custody.

Given their national presence, interface with veterans at multiple points within the criminal justice system (Blue-Howells et al., 2013), and knowledge of both VA and non-VA services, VJP Specialists are uniquely positioned to report on the availability and utility of services for justice-involved veterans that address key risk factors for recidivism. Accordingly, a qualitative approach was chosen so as to obtain in-depth knowledge of this process and the services used to target recidivism risk among justice-involved veterans.

Methods

Study design

VJP Specialists nationwide were recruited to participate in the study. At the outset of recruitment (April 2014), 220 VJP Specialists were employed by the VA (171 and 49 from the Veterans Justice Outreach and Health Care for Reentry Veteran programs, respectively). The names and email addresses of Specialists were identified through online, publicly-available databases. To obtain information from a representative sample of VJP Specialists nationally, we sought to recruit three Specialists from each of the Veteran Health Administration’s 21 regional networks (target N of 63), which equates to 29% of all VJP Specialists nationwide. Further, to better represent the ratio of Specialists between the two VJP programs, we randomly selected 2 Veterans Justice Outreach Specialists and 1 Health Care for Reentry Veteran Specialist from each VA regional network.

A total of 80 Specialists were emailed over three stages of recruitment until the recruitment goals were met and 63 Specialists agreed to participate (79%). To be included in the study, the participant had to currently hold a position as a Veterans Justice Outreach and/or Health Care for Reentry Veteran Specialist, and to have been in their current position for at least six months. Participation in the study involved a one-time, semi-structured interview by phone, which focused on Specialists’ practices and perspectives regarding treatment of recidivism risk factors among justice-involved veterans. Interviews were conducted and audio-recorded by the second author (AR), and lasted approximately 1.5 hours, on average. Interviews were conducted from April – October 2014. All procedures were approved by the Stanford University Institutional Review Board and the VA Palo Alto Research & Development Committee.

Sample

The final sample comprised 41 Veterans Justice Outreach Specialists, 13 Health Care for Reentry Veteran Specialists, and 9 “hybrid” Specialists who served justice-involved veterans through both VJP programs. Participants were predominantly female (n=48; 76%); Caucasian (n=48; 76%) or African American (n=10; 16%); 45 years old, on average (SD=10.3); had a Masters in social work or psychology (n=62; 98%); and had worked as a VJP Specialist for 40.6 months, on average (SD=23.3).

Measures: Qualitative interview guide

Data for this study were drawn from Specialists’ responses to questions from an interview guide, which was developed by the authors. Broadly, the interview sought to elicit a discussion with Specialists regarding their practices and perspectives in the treatment of recidivism risk among justice-involved veterans. Questions focused largely on modifiable risk factors from the RNR model (i.e., substance abuse; family/marital dysfunction; lack of positive school or work involvement; lack of prosocial activities/interests; antisocial personality patterns and cognitions; and antisocial associates) and EBTs for reducing recidivism among justice-involved individuals (i.e., MRT, T4C, and Reasoning & Rehabilitation). A sampling of relevant questions from the interview guide is provided in Table 1. For each of the aforementioned risk factors, participants were asked how they or other providers in the VA, the justice system, or in the community addressed this risk factor among justice-involved veterans; specifically, what types of services were most helpful.

Table 1.

Questions from the qualitative interview guide on availability of services and empirically-based treatments to address recidivism risk factors among justice-involved veterans.

“How have you or others addressed [INSERT RISK FACTOR BELOW] among the justice-involved veterans you serve?” “What has worked well to address this issue?” [Substance abuse] “…problems with alcohol or drugs” [Family/marital dysfunction] “…marital or family problems” [Lack of positive school or work involvement] “…limited education or employment problems” [Lack of prosocial activities/interests] “…limited leisure or recreational activities” [Antisocial tendencies] “…antisocial tendencies or traits,” “…criminogenic thinking,” “…antisocial thinking” [Antisocial associates] “…close relationships with others who engage in criminal activity”

“What is your experience or familiarity with the following treatments for recidivism?” “…Moral Reconation Therapy, or MRT?” “…Thinking 4 Change, or T4C?” “…Reasoning & Rehabilitation?”

“Do the veterans you work with receive any of these interventions?”

[If ‘yes’ to prior question] “Which ones and by whom?”

“In what context or settings are these interventions provided?”

Prior to recruitment, the interview guide was piloted with the National Coordinators of the Veterans Justice Outreach and Health Care for Reentry Veterans programs, as well as two former VJP Specialists. Pilot participants had difficulty distinguishing between questions related to antisocial personality patterns and antisocial cognitions, which are listed as two distinct risk factors in the RNR model, and provided essentially identical responses in terms of how these risk factors are addressed. Consequently, questions about these risk factors were combined into a single risk factor of “antisocial tendencies.”

Qualitative data analysis

Audiofiles of the interviews were de-identified by research staff (AR, LM), transcribed verbatim by a professional transcription service, and then reviewed for accuracy by research staff. Transcripts were then imported into the ATLAS.ti v7.5.9 (2015) software package. Subsequently, we used an adapted direct content analysis approach (Hsieh & Shannon, 2005) to develop a priori codes for the six RNR-based risk factors described above, and then used pile-sorting strategies (Bernard, 2005) to refine/subdivide the coding categories. The goals for analysis were to determine the extent to which treatment options or referral sources were reported as being available to address RNR-based risk factors for criminal recidivism among justice-involved veterans; how these risk factors are addressed, in terms of the treatment options and resources that Specialists found most helpful; and whether justice-involved veterans have access to EBTs for recidivism (i.e., MRT, T4C, and Reasoning & Rehabilitation).

First, two authors (AR, LM) coded each transcript to identify sections pertaining to discussion of the six RNR-based risk factors (see above). Next, the risk factor codes were extracted separately and coded further by the same authors in terms of whether a “treatment option” was or was not described by the Specialist. Specifically, a treatment-option code was applied when a Specialist reported at least one specific group, treatment program, or resource that they would provide and/or link justice-involved veterans to, or that was available to justice-involved veterans through other means, which directly addressed the risk factor in question.

To ensure that the risk factor and treatment-option codes could be applied reliably, the same authors (AR, LM) independently coded the textual data from six randomly chosen transcripts (approximately 10% of the total), and then discussed their independent ratings to reconcile any differences that emerged. This process was repeated until at least 80% agreement was reached on these codes. The remaining interviews were then divided equally between the two coders, who independently coded their assigned transcripts and then met weekly to discuss and reach consensus on any outstanding questions or issues. Following this, the first author independently coded all transcripts, and then met with the two coders to discuss and reach consensus on any discrepancies. Prior to this final consensus process with the first author, Cohen’s Kappas to estimate the inter-rater reliability of the treatment-option codes were: substance abuse (1.0), family/marital dysfunction (.65), lack of positive school or work involvement (1.0), lack of prosocial activities/interests (.74), antisocial tendencies (.70), and antisocial associates (.75).

Next, three authors (DB, AR, LM) used pile-sorting techniques to review and refine the treatment-option codes for each risk factor (Bernard, 2005). Pile-sorting is a qualitative data analysis technique where textual data are printed on paper and then sorted into piles based on common topics or themes (Bernard & Ryan, 2010). For each risk factor, an iterative process was used to review each quotation (i.e., the coded text) and sort into piles (i.e., sub-themes). Each quotation was discussed with the team until reaching a consensus about which pile to place the quotation. If an existing pile did not appear to fit the content of a quote, new piles were created. Once this process was complete, two authors (DB, CT) identified broader “meta-themes” among the specific treatment-option sub-themes for each risk factor. Exemplar quotes were then identified by the research team to illustrate each theme.

Finally, all transcripts were coded (by AR and LM) to identify sections pertaining to discussion of MRT, T4C, or Reasoning & Rehabilitation. The codes for these EBTs were extracted and coded further (by DB and JB) on whether the Specialist indicated that justice-involved veterans had access to one of these EBTs in a VA setting, and a non-VA setting (e.g., court or corrections). Of note, no Specialist reported knowledge of Reasoning & Rehabilitation being available to justice-involved veterans in any context; thus, codes for this EBT were not applied to any transcripts. All transcripts were coded independently (by DB and JB), who then discussed and reached consensus on any discrepancies for the MRT and T4C codes. Prior to this consensus process, Cohen’s Kappas for initial inter-rater reliabilities of these codes were: MRT in VA (.73), T4C in VA (.64), MRT in non-VA (.89), and T4C in non-VA (.89).

Results

Availability and utility of services to address recidivism risk factors among justice-involved veterans

Table 2 provides a summary of the number and percentage of Specialists who reported that treatment services and resources were available to address RNR-based risk factors (these risk factors are organized according to rate of availability, from highest to lowest). Table 2 also lists the themes regarding the helpfulness of services that were in common across risk factors (i.e., peer-based services; brief/informal motivational interviewing and/or cognitive-behavioral therapy; and structured cognitive-behavioral interventions), along with sample quotes to illustrate those themes separately for each risk factor. In addition to these cross-cutting themes, Veterans Treatment Courts were frequently reported by Specialists as being a helpful forum for justice-involved veterans to access a wide range of services to address recidivism risk factors. A more detailed summary of all themes that were identified for each risk factor is presented below. Of note, for each risk factor, co-occurrence of themes within interviews was permitted such that a given Specialist’s report of what is helpful to address that risk factor could have included more than one of the themes listed.

Table 2.

Availability and utility of services and resources to address recidivism risk factors among justice-involved veterans.

		Helpful services and resources (cross-cutting themes and sample quotes)
Risk factor	Number (%) of Specialists reporting that services and resources were available	Peer-based services	MI/CBT techniques (brief, informal)	Cognitive-behavioral interventions (structured groups)
Substance abuse	63 (100%)	“We have three peer support specialists, which I think is important for our justice-involved veterans and veterans with substance use disorders to have somebody to help them navigate through the recovery process and to be a good, pro-social kind of peer mentor to look up to and just to be there for support through the recovery process.”	“Implementing MI techniques and interventions with justice-involved veterans with substance use disorders have been really helpful to help me build rapport with these veterans, assess treatment motivation and treatment needs and help get them engaged in treatment.”	“It just seems that with substance abuse we see a lot of the stuff that goes with the Moral Reconation. We see the same kinds of world view and ways the guys are, that’s just where they’re at, is kind of with the personality traits. And so we definitely try to steer guys into that, the Moral Reconation Treatment group.”
Lack of positive school or work involvement	63 (100%)	“We have a Peer Support Program where they can be referred there to help them… if they have a job fair coming up, transport them, make sure they get there, make sure they have their resume. Anytime they have community resources or people who are currently hiring they’re able to help them as far as filling out applications and getting to interviews, things like that.”
Family/marital dysfunction	59 (94%)
Lack of prosocial activities/interests	54 (86%)	“I think here again, that peer support connection is a positive piece, because they can talk on that vet-to-vet level and maybe share some of the things that have been helpful for him. And connect them to those service organizations and those activities, and help them explore some leisure type of things.”
Antisocial tendencies	36 (57%)	“Even though a lot of the literature talks about some of those factors like personality are really hard to change, the truth is we know that the majority of folks do change and that is takes time, and that folks need help and encouragement with it, so…And I think that’s another way to incorporate that peer support specialist, specifically if we got somebody who had actually served time themselves”	“I think that the motivational interviewing techniques really help out a lot to get you to get past all of the resistance and the blaming to really what are their goals, what are their aspirations, their desires, their warmth, and how can I help.” “I have previously been trained in CBT, so I use a lot of CBT intervention and homework assignments and changing thoughts, and looking at cognitive distortions and things like that.	“The best evidence-based practice for that is MRT or corrective thinking. Or Thinking for a Change or some of those others corrections- based programs.” “I do think that there are some DBT skills that they can benefit from. But I definitely think in kind of looking at evidence-based practices that with the court and with people who are high recidivists who have these traits, that it’s really imperative to have some type of group, like MRT or Thinking for a Change.”
Antisocial associates	32 (51%)	“I also talk about the recovery process. It’s a lifestyle change. It’s not just I stopped drugs; you have to stop associating yourself with people who use drugs. And so we really highlight that piece to it. I also utilize peers in that process because we can have the conversation, but I find that it has far more meaning when it comes from a peer than it does from a clinician.”	“Sometimes in MRT when people have talked about that, about how it’s difficult to be able to let go of some relationships that they know are probably harmful to them, we would do motivational interviewing, I think it’s the veteran’s choice who they decide to be with. But you know, I think providing alternatives, talking about social situations that may be different.” “It’s never a good idea to be around somebody who’s engaging in criminal activity when you're trying to do something different… So I would probably attack kind of those thought processes… I would use motivational and obviously the cognitive behavioral [approaches] and try and help them see that that’s probably not a good idea.”	“MRT really is about decision- making and about where you have been, what decisions have you made up to this point in your life, and what decisions do you want to make? What kind of life do you want to have for yourself as you look forward in the future? And a lot of that is about the people that are in your life. So it’s about repairing damaged relationships with the people that you love. And oftentimes it involves severing ties with people who do drugs and commit crimes and do those kinds of things.”

Notes. n = 63. MI = motivational interviewing; CBT = cognitive-behavioral therapy; MRT = Moral Reconation Therapy; DBT = Dialectical Behavior Therapy.

Substance abuse

All Specialists reported that some type of treatment option or resource was available to address substance abuse among justice-involved veterans. In terms of what is helpful to address this risk factor, responses by Specialists focused largely on (a) referring justice-involved veterans to substance abuse outpatient and residential treatment programs and self-help groups in the VA and community, (b) VJP Specialists’ use of motivational interviewing or brief motivational enhancement therapy to increase veterans’ motivation for substance abuse treatment, (c) the utility of Peer Support Specialists, available either through the VA or a Veterans Treatment Court, who help to increase a veteran’s engagement in and motivation for treatment services, and (d) structured cognitive-behavioral groups, including MRT, available through either the VA or a Veterans Treatment Court.

Lack of positive school or work involvement

All Specialists reported that some type of treatment option or resource was available to address lack of positive school or work involvement for justice-involved veterans. In terms of what is helpful to address this risk factor, responses by Specialists focused on themes of (a) referrals to specialized employment services (e.g., Vocational Rehabilitation programs; Supported Employment; Compensated Work Therapy) in the VA or community; (b) referrals to Veterans Service Organizations or a Veterans Benefits Administration representative to assist the veteran with determining his/her eligibility for benefits to subsidize education and training, and (c) peer groups and/or Peer Support Specialists, available through either the VA or a Veterans Treatment Court, who can assist the veteran with developing their job skills, networking, and/or increasing motivation to seek gainful employment.

Family/marital dysfunction

Fifty-nine Specialists (94% of the total sample) reported that some type of treatment option or resource was available to address family/marital dysfunction among justice-involved veterans. In general, these responses focused on (a) couples counseling and other specialized family-based services through VA behavioral health services and/or Vet Centers in the community, or (b) interpersonal violence programs through VA, Vet Centers, or Veterans Treatment Courts.

Lack of prosocial activities/interests

Fifty-four Specialists (86% of the total sample) reported that some type of treatment option or resource was available to address lack of prosocial activities/interests among justice-involved veterans. In general, these responses entailed (a) referrals to recreational/volunteer groups in the VA or the community (e.g., Veteran Service Organizations; Vet Centers); (b) referrals to residential behavioral health programs that explicitly focused on enhancing recreational/leisure activities as part of a veteran’s recovery programming; and (c) the utility of Peer Support Specialists, available through the VA or Veterans Treatment Courts, who facilitated recreational and volunteer opportunities for veterans.

Antisocial tendencies

Twenty-seven Specialists (43% of the total sample) did not report use of any treatment options or resources to address antisocial tendencies among justice-involved veterans. When asked to describe how such tendencies are addressed in these veterans in the absence of treatment options, these Specialists reported either (a) resistance to working with veterans with these tendencies due to perceived lack of treatability; (b) the importance of setting limits and maintaining clear boundaries; (c) the use of sanctions and rewards from the justice system as a means of modifying behavior; and/or (d) having an informal conversation with the veteran regarding their behavior and choices, but not utilizing any specific therapeutic model or curriculum.

Thirty-six Specialists (57% of the total sample) reported that some type of treatment option or resource was available to address antisocial tendencies among justice-involved veterans. These responses focused on (a) informal use of motivational interviewing and/or cognitive-behavioral techniques to increase motivation for behavioral change and challenge pro-criminal attitudes among veterans; (b) the utility of Peer Support Specialists through either the VA or a Veterans Treatment Court to help model effective problem-solving skills and self-control; and (c) structured, cognitive-behavioral groups through the VA or a Veterans Treatment Court that specifically targeted antisocial tendencies. This included specific mention of MRT and T4C to address antisocial cognitions, as well as anger management training to address aggression and Dialectical Behavior Therapy to address impulse control problems.

Antisocial associates

Thirty-one Specialists (49% of the total sample) did not report use of any treatment options or resources to address affiliations with antisocial peers among justice-involved veterans. When asked to describe how this risk factor is addressed in these veterans in the absence of treatment options, these Specialists reported either (a) use of sanctions and rewards from the justice system as a means of modifying behavior; (b) having an informal conversation with the veteran regarding his/her behavior and choices, but not utilizing any specific therapeutic model or curriculum; and (c) identifying alternative housing options for the veteran to remove him/her from a high-risk neighborhood.

Thirty-two Specialists (51% of the total sample) reported that some type of treatment option or resource was available to address affiliations with antisocial peers among justice-involved veterans. These responses focused on (a) referring veterans to substance abuse and/or mental health treatment programs in VA or community, which explicitly employed either a relapse prevention model to modify veterans’ peer networks, and/or social skills training to improve veterans’ ability to develop relationships with prosocial peers; (b) informal use of motivational interviewing and/or cognitive-behavioral techniques to address the veterans’ affiliations with antisocial peers; (c) the utility of a Peer Support Specialist, available through either the VA or a Veterans Treatment Court, and/or facilitating veterans’ engagement with a sponsor through a mutual-help group; (d) linking veterans to recreational or volunteer groups in the VA and/or the community that could assist them with building a larger network of prosocial peers; (e) “re-entry” groups through the VA or a Veterans Treatment Court that explicitly focused on veterans’ peer networks, but did not utilize a manualized curriculum; and (f) structured, cognitive-behavioral groups through the VA or a Veterans Treatment Court that directly targeted antisocial tendencies and included modules that focused on changing one’s peer networks (e.g., MRT or T4C).

Justice-involved veterans’ access to empirically-based treatments to reduce risk for recidivism

Table 3 provides a descriptive summary of the percentage of VJP Specialists who reported that justice-involved veterans had access to EBTs to reduce risk for recidivism. As noted above, no Specialist reported knowledge of Reasoning & Rehabilitation being available to justice-involved veterans. A little less than one-third of VJP Specialists reported knowledge of either MRT or T4C being currently available at a VA medical center in their catchment area, with a larger percentage of Specialists reporting the availability of MRT (25%) than T4C (5%) at VA.

Table 3.

VJP Specialists who reported that justice-involved veterans had access to empirically-based treatments (EBTs) to reduce risk for criminal recidivism.

	n (%)
	EBTs provided in VA	EBTs provided in non-VA settings	EBTs provided in either VA or non-VA settings
MRT and/or T4C	18 (29%)	34 (54%)	43 (68%)
MRT	16 (25%)	19 (30%)	27 (43%)
T4C	3 (5%)	19 (30%)	21 (33%)

Notes. MRT = Moral Reconation Therapy. T4C = Thinking 4 a Change.

Relative to VA, access to an EBT for recidivism was reported to be higher for justice-involved veterans in non-VA settings (slightly more than half of all Specialists; n=34). Of these Specialists, the majority (65%) reported that one of the EBTs was available to Veterans through a Veterans Treatment Court and/or parole or probation services, with the remaining Specialists reporting access as occurring while veterans were incarcerated in jails and prisons. Further, the availability of MRT and T4C was comparable in non-VA settings. Across both VA and non-VA settings, the majority of VJP Specialists (68%) reported that justice-involved veterans had access to an EBT for recidivism, with the overall rate higher for MRT (43%) than T4C (33%).

We also explored whether access to EBTs for recidivism varied by regions. To increase our power to detect group differences, the 63 participants from the 21 regional networks that we sampled from were grouped into the five larger networks of the Department of Veterans Affairs: North Atlantic (n=18), Southeast (n=9), Midwest (n=15), Continental (n=9), and Pacific (n=12). Only one significant difference was observed: the percentage of VJP Specialists who reported that justice-involved veterans had access to MRT in either a VA or non-VA setting was significantly different across these regions, χ²(4) = 10.03, p=.04. Specifically, the percentage was highest for Specialists in the Midwest (67%), followed by the Pacific (58%), Continental (44%), Southeast (33%), and North Atlantic (17%).

Discussion

The objectives of the current study were to determine the availability of services for justice-involved veterans that address empirically-supported risk factors for recidivism that are targeted by the RNR model, the types of treatment options and resources that are perceived as being most helpful in addressing these risk factors, and whether justice-involved veterans have access to EBTs for recidivism risk. For most risk factors (substance abuse, lack of positive school or work involvement, family/marital dysfunction, and lack of prosocial activities/interests), responses from the majority of VJP Specialists in our sample included description of some type of treatment option or resource that was available to address these issues among justice-involved veterans. By comparison, only a little more than half of the Specialists’ responses included description of any treatment options or resources to address antisocial tendencies and antisocial associates. This is noteworthy, given that risk factors related to antisociality (i.e., the Big Four) have been identified as the strongest predictors of criminal recidivism (Andrews & Bonta, 2010a, 2010b).

The current findings are largely consistent with research on availability of services for offenders in general. For example, a recent review highlighted the range of offender reentry programs and services that target vocational training, substance abuse prevention, and other psychosocial needs of offenders (e.g., housing; James, 2015). By contrast, others have noted that extant programming for offenders tend not target antisocial tendencies, despite the centrality of these issues in the prediction of recidivism risk (Epperson et al., 2014; Wolff et al., 2013). This potential gap may be greater among justice-involved veterans, given that in the current study the availability of interventions that directly target antisocial cognitions and attitudes (e.g., MRT, T4C) were reported to be less prevalent in VA (vs. non-VA settings).

Potential limitations in the availability of treatments and resources to address risk factors related to antisociality represent a novel finding in the empirical literature on justice-involved veterans, and highlight a potential gap in the implementation of best practices to reduce recidivism in this population. To address this issue, it may be beneficial to provide more education and training to providers who work with justice-involved veterans about the evidence for the Big Four as being strong predictors of recidivism, and the existence of EBTs to address these risk factors. The latter may help to address potential misperceptions regarding the lack of treatability of antisocial tendencies (National Collaborating Centre for Mental Health, 2010), and encourage providers to focus not just on the unique re-entry challenges of justice-involved veterans (e.g., traumatic brain injury and/or PTSD due to combat exposure; Sreenivasan et al., 2013), but also on the risk factors and re-entry challenges that are often in common between veteran and non-veteran offenders. Such trainings could also clarify that antisocial traits and tendencies exist on a continuum and are not “set like plaster” throughout adulthood (Caspi, Roberts, & Shiner, 2005). Training and education in these areas may increase the extent to which clinical providers who work with justice-involved veterans assess for, and monitor change in, antisocial cognitions and attitudes over the course of treatment (Walters & Lowenkamp, 2015; Knight, Garner, Simpson, Morey, & Glynn, 2006).

Peer-based services

In terms of the types of treatment options and resources that were perceived as being most helpful to address RNR-based risk factors more generally, several cross-cutting themes emerged. Most notably, peer-based services were perceived by VJP Specialists as being a useful resource to address nearly all risk factors from the RNR model. The use of peer-based services within criminal justice settings has grown substantially in the past decade (Davidson & Rowe, 2008), and more recently has played a critical role in the care continuum of veterans with substance use and mental health disorders (Levardi, 2013). Peer Specialists can fill a range of functions for veterans including offering practical advice and guidance on how to navigate complex bureaucratic systems to identify housing, vocational, educational, and even recreational opportunities. For example, previous authors have asserted that peer-based support may be an especially valuable resource for justice-involved veterans by providing them with information about their eligibility for, and the availability of, VA and community-based services to support re-entry planning (Glynn et al., 2014; Rosenthal & McGuire, 2013). Peer Specialists can also provide a range of psychosocial functions such as instilling hope in veterans, motivating them to engage in treatment and other supportive services, and assisting veterans in their relapse prevention efforts related to drug and alcohol use (Landers & Zhou, 2011). Further, Peer Specialists may also help to address risk factors of antisociality by modeling effective problem-solving skills and self-control and more generally serving as a model of someone who was able to overcome a criminal lifestyle, and/or expanding justice-involved veterans’ networks of prosocial peers (Davidson & Rowe, 2008). Collectively, the current findings suggest that use of peer-based services for justice-involved veterans may have broad impact in terms of addressing a number of risk factors identified in the RNR model. This subjective evidence regarding the effectiveness of peer-based services to reduce risk for recidivism is also supported by some empirical studies. For example, O’Donnell and Williams (2013) reported that adolescent offenders assigned to a peer mentoring program (vs. a no-treatment control group) exhibited significantly lower rearrest rates when followed up into adulthood. In addition, among female offenders who participated in a peer-driven case management program through the San Diego Sheriff’s Department, the rate of recidivism in the 12 months post-release was only 23%, compared to 66% for all offenders released from the California Department of Corrections over the same time period (Goldstein, Warner-Robbins, McClean, Macatula, & Conklin, 2009). It should be acknowledged that neither of these studies involved justice-involved veterans per se. Nonetheless, policies and programs that support and expand peer-based services may be a promising strategy for reducing risk for recidivism among justice-involved veterans.

Brief/informal motivational interviewing and/or cognitive behavioral therapy

Another theme that emerged across multiple risk factors (i.e., substance abuse; antisocial tendencies; antisocial associates) was the use of motivational interviewing and/or cognitive behavioral techniques to help veterans modify their behavior and increase motivation for behavioral change and treatment engagement (Glynn et al., 2014). The use of cognitive-behavioral techniques and strategies, even when used informally by case managers, is consistent with “what works” for reducing risk for recidivism among justice-involved adults (Aos et al., 2006; Landenberger & Lipsey, 2005; Milkman & Wanberg, 2007). Further, in terms of treatment engagement, a number of randomized trials support use of brief forms of motivational interviewing and Motivational Enhancement Therapy to increase veterans’ motivation for substance abuse and mental health treatment engagement, including two that focused on justice-involved veterans (Davis, Baer, Saxon, & Kivlahan, 2003; Wain et al., 2011). Accordingly, with the requisite funding and resources, policies that support increased training in the use of brief motivational interviewing and cognitive-behavioral techniques among providers and other justice system personnel who work with justice-involved veterans may have a widespread, beneficial impact on multiple risk factors for recidivism. Such training could also incorporate ways to adapt these clinical techniques to the unique needs of this veteran population (Baer & Kivlahan, 2008; McMurran, 2009).

Cognitive-behavioral interventions (structured groups)

Structured, group-based cognitive-behavioral interventions, such as MRT and T4C, were also identified by VJP Specialists as a treatment option that can be effective for addressing multiple risk factors for recidivism. Although the content and theorized mechanisms of these interventions are largely focused on modifying “criminogenic thinking,” there is some evidence that they may have beneficial effects for a range of health-related outcomes including substance abuse (Anderson, 2002; Fuller, 2003). In terms of access to these EBTs, across VA and non-VA settings, the majority of Specialists (68%) reported that justice-involved veterans have access to either MRT or T4C. However, this estimate was driven largely by the availability of EBTs in non-VA settings, primarily through Veterans Treatment Courts and parole/probation services in the community. While this figure speaks well to the availability of EBTs for justice-involved veterans in general, access to an EBT for recidivism in the VA per se is lower. This potential disparity was noted by one of the VJP Specialists we interviewed (“The courts are hearing more about [MRT] and it’s becoming very popular. And some of the judges have made comments that the VA doesn’t provide this or why can’t the VA provide this?”). Due to the emergence of the VJP, the number of justice-involved veterans receiving health care from the VA has increased substantially over the last decade. Accordingly, efforts to increase implementation of MRT or T4C into VA settings represent an opportunity to expand justice-involved veterans’ access to best practices for reducing risk for recidivism. To this end, a veteran-specific manual for MRT was recently developed (Little & Robinson, 2013), and research is currently underway to examine both the effectiveness and implementation potential of MRT in mental health residential treatment programs across VA. Policy responses that further support these and other initiatives may only further the potential to reduce justice-involved veterans’ risk for recidivism.

Veterans Treatment Courts

Finally, as noted above, although it is not a treatment service or resource in the same vein as the other meta-themes described here, Veterans Treatment Courts were frequently highlighted by VJP Specialists as a forum in which justice-involved veterans have access to a wide range of resources that are helpful to target recidivism risk factors. These courts are a recent adaptation to the specialty mental health and drug treatment court model. They are designed to meet the needs of justice-involved veterans with criminal charges, divert them from incarceration, and facilitate their engagement in an array of mental health and psychosocial services (Clark et al., 2010; Russell, 2009). Importantly, many of these courts have strict eligibility requirements (Baldwin, 2013; Clark et al., 2010) and therefore serve only a specific subset of justice-involved veterans; however, for those who are eligible, the resources and services available through these courts were perceived by VJP Specialists to be useful for addressing a range of recidivism risk factors. Consistent with this perception, there is preliminary evidence for the efficacy of Veterans Treatment Courts to improve a number of outcomes related directly and indirectly to the risk factors in the RNR model (i.e., substance abuse, family functioning, relationships with others, and social connectedness; Knudsen & Wingenfeld, 2015).

Since their initiation in 2008, the number of Veterans Treatment Courts in the U.S. has grown rapidly and now reaches over 200 nationwide (National Association of Drug Court Professionals, n.d.). Nonetheless, many court districts have yet to establish a Veterans Treatment Court, and many limit participation in these courts to certain offenders (e.g., low-risk offenders, or those with non-violent offenses; Baldwin, 2013). Importantly, such exclusions run counter to the principles of the RNR model, which assert that programs that provide significant structure and intensive monitoring and supervision, such as Veterans Treatment Courts, should be provided to the highest risk offenders (Andrews & Bonta, 2010a, 2010b). Given these issues, it may behoove policymakers to support communities with a large veteran population that do not have a Veterans Treatment Court to begin such courts, as well as adjust eligibility requirements, to promote greater inclusion. Further, as future research begins to identify the characteristics of these courts that influence veterans’ outcomes, more formal guidelines and recommendations regarding the structure, policies, and services of Veterans Treatment Courts will emerge to help standardize the components of these specialty courts and guide policymakers in their efforts to expand justice-involved veterans’ access to these courts.

Limitations and future directions

The present findings should be interpreted within the context of some limitations. First, the present findings are limited to the perspectives of VJP Specialists. While VJP Specialists are uniquely positioned to provide information to address this study’s questions, other providers or justice system personnel may have differing perspectives on the availability and utility of treatment options and resources to reduce recidivism risk among justice-involved veterans. Second, although there are a number of advantages to our qualitative data such as gaining in-depth information from key informants in the care of justice-involved veterans, it will be critical for future research to utilize other data sources at the patient-level (e.g., surveys of justice-involved veterans and/or examination of clinical notes of these veterans who receive substance use and mental health treatment in the VA) to further evaluate the availability of services and resources to address recidivism risk factors. Comparison of such data with the qualitative data from this study would strengthen our understanding of whether (and where) gaps exist in the implementation of best practices to reduce recidivism risk among justice-involved veterans. Further, it should be noted that availability of a service does not equate to engagement; thus, the current results cannot speak to whether justice-involved veterans take advantage of available services or whether VJP Specialists and/or clinical providers appropriately match an intervention to a veteran’s assessed risk and need. Third, as noted in the introduction, although the Central Eight appears applicable to the prediction of recidivism among justice-involved veterans (Blonigen et al., 2014), other factors not included in the RNR model, such as homelessness (Greenberg & Rosenheck, 2008) and PTSD (Sadeh & McNiel, 2015), may also be relevant to the prediction of recidivism in this population. Thus, future research should continue to fill this gap in the literature to ensure that any policy responses aimed at reducing recidivism risk among justice-involved veterans are based on the strongest empirical evidence. Finally, VJP Specialists’ perceptions of what treatment options are most helpful to address a given risk factor do not substitute for empirical evidence. Nevertheless, from a policy standpoint, such information highlights the types of services and resources that should be prioritized for more rigorous testing and (if proven effective) implemented more widely across systems.

Summary and Conclusions

This study provides guidance to policymakers on potential gaps in the implementation of best practices to reduce recidivism among justice-involved veterans, and promising policy-based solutions that may have a widespread, beneficial impact on reducing recidivism in this population. It suggests that more systematic implementation of interventions and strategies that focus on risk factors of antisociality (e.g., antisocial cognitions, attitudes, and affiliations) is needed. In addition, expansion of peer-based services, training in motivational interviewing and cognitive behavioral therapy, Veterans Treatment Courts, and implementation of EBTs for recidivism in the VA are policy responses that may help to maximize reductions in the risk for recidivism among justice-involved veterans.