A systematic review of comprehensive interventions for substance abuse: Focus on victimization

Abstract

Violence victimization is common among men and women who use substances and is associated with co-occurring health issues such as PTSD, depression and HIV. Substance use interventions, therefore, should include integrated components that are designed to address co-occurring health issues among victimized substance-using individuals. This systematic review synthesized the evidence on efficacy of comprehensive, integrated, multicomponent interventions for victimized substance-using individuals. The efficacy of integrated multicomponent intervention strategies was assessed for the following syndemic conditions: mental health, substance misuse, violence, and HIV risk. Seventeen studies were identified. Examples of effective components were empowerment strategies for violence, mindfulness-based stress reduction for mental health, social cognitive skill building for addressing HIV risk and psychoeducation for substance misuse. Although in this review, some components were found to be effective, we identified methodological limitations of included studies which calls for more rigorous research in this area. Further, there is lack of evidence base for multicomponent interventions for victimized substance-using individuals in developing countries. Additional studies are needed to establish rigorous evidence base for multicomponent interventions for victimized substance using individuals that help them cope effectively with their trauma of violence and address their needs.

1. Introduction

People or men and women who use substances/drugs are more likely than the general population to have been exposed to violence (de Waal et al., 2018; Marshall, Fairbairn, Li, Wood, & Kerr, 2008; Vaughn et al., 2010). Violence victimization has been associated with cooccurrence of mental health and HIV among people who use substances (de Waal et al., 2018; Kessler, 2004; Marshall et al., 2008; Nayak, Lown, Bond, & Greenfield, 2012; Sabri, 2012; Taylor, 2011), greater impairment in functioning and poorer response to substance abuse treatment (de Waal et al., 2018; Sabri, 2012). Treatment and prevention of these problems are of critical importance for psychosocial wellbeing, prevention of human immunodeficiency virus (HIV) and viral hepatitis, and adherence to and engagement in substance abuse treatment (Buckingham, Schrage, & Cournos, 2013; Walkup et al., 2008; Wood, Kerr, Tyndall, & Montaner, 2008). This calls for comprehensive, integrated, multicomponent interventions designed to address and prevent co-occurring health issues among victimized substance-using individuals. Victimized substance-using individuals refer to substance using individuals with experiences of interpersonal violence victimization. The experiences of victimization include experiences of physical, sexual and/or psychological abuse in childhood and/or adulthood by any perpetrator (e.g., intimate partner, family member, community member).

Although victimized substance-using individuals have co-occurring health concerns such as HIV risk and mental health, the healthcare delivery system typically attends to these problems separately, thus creating siloed systems of care (McGovern & McLellan, 2008). An analysis of healthcare facilities in the United States (US) found that only 18% of substance use services met were designed to address both substance misuse and co-occurring disorders (McGovern, Lambert-Harris, Gotham, Claus, & Xie, 2014). Victimized substance-using individuals with multiple co-occurring health problems tend to derive less benefit from single disorder-focused interventions (McGovern, Lambert-Harris, Alterman, Xie, & Meier, 2011; Sacks, 2000; van Dam, Ehring, Vedel, & Emmelkamp, 2013). Data and experts tend to support multicomponent interventions to address co-occurring health problems (National Institute on Drug Abuse, 2010; Perron, Bunger, Bender, Vaughn, & Howard, 2010; Torrens, Rossi, Martinez-Riera, Martinez-Sanvisens, & Bulbena, 2012). It is possible that single-disorder focused interventions targeting either substance misuse, mental health problems, HIV, or violence are not as effective for victimized substance using individuals with co-occurring health problems because they do not meet the unique needs of this population. Comprehensive multicomponent interventions that integrate elements of treatments for substance misuse, violence, mental health, HIV risk and/or other health problems may address cooccurring needs and shortcomings of single-disorder focused interventions. Multicomponent interventions have been developed and tested for people with co-occurring substance misuse and other health problems. It is, however, unclear which components of interventions are most effective in improving mental health, reducing violence, HIV risk, and substance misuse among victimized substance using individuals.

According to the syndemic model, the epidemic of substance misuse, violence, HIV and mental health can be conceptualized as a syndemic, defined as interconnected health concerns that interact to account for health inequities in some populations (Gonzalez-Guarda, Florom-Smith, & Thomas, 2011; Sullivan, Messer, & Quinlivan, 2015) such as people who use substances. This calls for interventions that address at least some of these syndemic conditions simultaneously. Drawing from the syndemic model, the study aims to identify gaps in the current evidence-base for interventions for victimized substance-using individuals. Further, it aims to identify potential intervention components that could be considered for designing multicomponent interventions for addressing at least two or more syndemic conditions among victimized substance-using individuals. Current integrated interventions for co-occurring disorders (e.g., Frisman, Ford, Lin, Mallon, & Chang, 2008; Weiss et al., 2007) do not focus on victimized substance-using individuals. Further, systematic reviews on interventions for people who use substances with co-occurring disorders focus on post-traumatic stress disorder (PTSD) or general trauma (Roberts, Roberts, Jones, & Bisson, 2015; Simpson, Lehavot, & Petrakis, 2017; Trochalla, Nosen, Rostam, & Allen, 2012), and not additional syndemic conditions such as violence and HIV. Moreover, the focus has not been on the need for multicomponent integrated interventions that can address syndemic issues such as violence, HIV, mental health along with substance misuse among victimized substance-using individuals. We address this gap in the literature by focusing on the need for multicomponent interventions that address syndemic conditions specifically among victimized substance-using individuals.

1.1. Purpose of the study

This systematic review synthesized the evidence on comprehensive multicomponent interventions for victimized substance-using individuals. A multicomponent intervention was defined as an intervention with a combination of components that addressed two or more syndemic outcomes (i.e., violence, mental health, substance misuse, HIV risk) for victimized substance-using individuals. The purpose was to: a) Describe the characteristics and effectiveness of integrated intervention strategies for victimized substance-using individuals; and b) Determine the efficacy of various integrated interventions strategies on the following syndemic outcomes: violence, HIV risk, substance misuse and mental health.

2. Material and methods

2.1. Search strategy

We searched six academic literature databases, namely PubMed/MEDLINE, PsycInfo, the Cochrane Register of Controlled Trials (CENTRAL), Embase, ISI Web of Science and CINAHL. The search strategy contained key terms describing violence, mental health, HIV, substance misuse, and randomized controlled trials.

2.2. Eligibility criteria

2.2.1. Inclusion criteria

Studies were included if they were (1) randomized controlled trials (RCTs) since RCT is considered a gold standard, (2) evaluated a multi-component intervention on two or more of the following primary syndemic outcomes for victimized substance-using adults: violence, HIV, substance misuse, and mental health. We selected studies with minimum two conditions since we could not find an intervention evaluation study that addressed all of the aforementioned syndemic conditions simultaneously; (3) Since violence exposure/victimization among people who misuse substances was explicitly reported as an eligibility criterion in only 4 intervention studies (3 intimate partner violence (IPV) and 4 general histories of physical and sexual abuse, we also included studies which reported prevalence of violence victimization (i.e., 77% or more) among the participants who use substances (e.g., child abuse, IPV), 4) included studies that focused on men and women with histories of substance use, and (5) we included only studies that were published in English. The search criteria did not limit the studies to particular geographical locations, settings or specific time periods (Table 1).

Table 1.

Description of studies evaluating multicomponent interventions for survivors of abuse.

Authors	Name of the intervention; location	Description of the intervention/program	Eligibility regarding violence/trauma & substance misuse	Description of the control group	Follow up and assessment of outcomes	Results	Limitations
Barrett et al., 2015	Seeking Safety; Australia	Objective: To reduce trauma- and substance-related problems Program Components: Cognitive-behavioral therapy Psychoeducation Coping skills training Safety planning Individual sessions Provider: Clinical psychologist (masters level)	Eligibility: Violence not mentioned as an eligibility criterion, but more than 80% participants reported violence histories PTSD checklist screen positive Substance misuse - history of problematic substance use Sample size Intervention group N=15 Control group N =15 Recruited from correctional facilities	Usual care - included opioid substitution treatment, SMART Recovery, Narcotics Anonymous, non-pharmacological substance use treatment, mental health treatment	Follow-up: 2- and 6-months Outcomes assessed: intervention feasibility/acceptability; ability to resist substance use; PTSD	Demographics - Adult male prisoners; 23% Aboriginal Mental Health - Preliminary evidence suggesting reductions in PTSD symptoms in both groups Substance Use - Preliminary evidence suggesting increased confidence in ability to resist substances in intervention group	Statistical power/sample size; Attrition; Contamination; Shortened treatment protocol (25 to 8 modules); No statistical inferences on treatment outcomes
Choo et al., 2016	BSAFER; United States	Objective: Reduce drug use and IPV Program Components: Web-based program to be completed during a single ED visit + phone-based booster session 2-weeks post-ED visit Automated feedback on drug use and health Empowerment Goal setting Social support Referrals and advice Individual (web- and booster session) session Provider: Web-based + telephone booster	Eligibility: Intimate partner violence in past 3months Substance misuse - drug use in past 3 months Sample size Intervention group N= 21 Control group N =19 Recruited from adult EDs from a level I trauma center	Time-matched web—based program on home fire safety and included interactive components; Telephone booster with a brief conversation about fire safety	Follow-up: 1- and 3-months Outcomes assessed: Past-month drug use; physical, psychological and sexual violence	Demographics - Adult females in the emergency department; 50% White Substance Use - Preliminary evidence for modest decreases in drug use days in both groups at 3 months. The intervention group had a mean decrease of 0.7days per week compared to a mean decrease of 1.5days per week in the control group. Among those using drugs other than marijuana, the intervention group had a mean decrease of 2.5 using days per week compared with a decrease of 1.3 using days per week in the control group. Violence - Preliminary evidence for modest decreases in IPV in both groups. The IIPV scores decreased by a mean of 4.1 points in the intervention group compared with a mean decrease of 3.3 points among controls	Statistical power/sample size; Attrition; Implementation challenges; Selection bias; Generalizability
Garland et al., 2016	Mindfulness-Oriented Recovery Enhancement (MORE); United States	Objective: Reduce craving, PTSD symptoms and psychological distress Program Components: Mindfulness training Cognitive-behavioral therapy/positive reappraisal training, Group sessions Provider: Social worker (masters level)	Eligibility Trauma histories including violence More than 80% of participants reported violence experience Substance use and co-occurring psychiatric issues Sample size Intervention group N= 64 Control group N = 64 CBT, N= 52 TAU Recruited from a therapeutic community	Two control conditions: 10-session cognitive-behavioral therapy based on Seeking Safety; treatment as usual: psychoeducation, therapy, coping skills in therapeutic community	Follow-up: 10-weeks Outcomes assessed: Substance use craving, PTSD, Depression, Anxiety, Positive/Negative Affect	Demographics - Adult male; 42% White, 44% Black Mental Health - Greater reductions in PTSD symptoms relative to Seeking Safety; Marginally significant reductions in PTSD symptoms relative to treatment as usual (p = 0.05) Substance Use - Great reductions in craving relative to Seeking Safety; No differences between mindfulness and treatment as usual	Short follow-up precluding evaluation of sustained treatment effects; may not be generalizable to patients without co-occurring substance use and mental disorder
Ghee et al., 2009	Condensed Seeking Safety Intervention	Program Components: Coping skills training for both trauma and substance abuse with six topics: Introduction to safety, PTSD, detaching from emotional pain, setting boundaries in relationships, asking for help & commitment	Eligibility Histories of physical and/or sexual abuse Enrolled in residential treatment of substance use in a community-based alcohol and drug treatment center Sample size Intervention group N = 52 Control group N= 52	Standard treatment of substance use for participants in residential chemical dependence program	Follow-up: 30 days postcompletion of residential treatment Outcomes assessed: Sexual abuse trauma, PTSD, drug-screen and self-report of drug abstinence/relapse Overall trauma	Demographics: Adult women, Caucasian (51%), and African American (47%). Lower sexual abuse related trauma symptoms—not more advantageous in reducing overall trauma symptoms or relapse 30 days after treatment ended	Small sample; One treatment facility limits the generalizability of our findings-the small number of participants returned for the 30 days posttreatment assessment may have affected the power to show results; Longer time frame for assessing violence
Gilbert et al., 2006	Relapse Prevention and Relationship Safety (RPRS); United States	Objective: Reducing drug use and IPV Program Components: Social cognitive skill building Empowerment Social support Negotiation and boundary setting skill building appropriate to individual woman’s situation Negotiations skills related to condom use Raising awareness of the co-occurrence of IPV & drug use- Role playing self-regulatory, communication & negotiation skills & reinforcing selfefficacy-treatment content culturally specific to Black & Latina women Group and individual sessions Provider: Trained facilitator	Eligibility: Past 90-day intimate partner violence experience Substance use-past 90-day drug use Sample size Intervention group N= 16 Control group N= 18 Recruited from methadone maintenance treatment programs	Informational control condition: 1-hour didactic session presenting a variety of community services	Follow-up: 3-months Outcomes assessed: drug use, IPV, depression, PTSD, sexual HIV risk behaviors	Demographics - Adult female; 59% Latina, 21% White HIV-risk - Less likely to have sex while high on drugs, but no difference in number of unprotected sexual occasions or number of sexual partners Mental health - The RPRS group was significantly more likely than control to report decrease in depression. Although, the RPRS group showed a decrease in PTSD avoidance symptoms than control, the difference was not significant at 0.05 level (p = 0.06) Substance use - No difference between the groups in use of heroin or marijuana, However, women in RPRS were more likely than control group to report decrease in any illicit drug use, binge drinking or crack cocaine, although the effects were not significant Violence - RPRS participants were more likely than controls to show decrease in minor physical, sexual or injurious IPV and any severe physical IPV as well as minor or severe psychological IPV.	Statistical power/sample size
Gilbert et al., 2015	Project WINGS (Women Initiating New Goals of Safety); United States	Objective: Disclose IPV and receive IPV services; improve IPV-related selfefficacy, social support and drug abstinence Program Components: Psychoeducation Motivation enhancement Safety planning & danger assessment Social support Goal setting Referrals and advice Individual session Provider: Web based	Eligibility: Being in an intimate relationship (~77% sample reported intimate partner violence) Substance use - past 6-month illicit drug use, drinking or substance use treatment Sample size Intervention group N= 94 Control group N= 97 Recruited from probation or community court-administered alternative-to-incarceration programs	SBIRT provided by a case manager (1 session)	Follow-up: 3-month follow-up Outcomes assessed: IPV, drug use	Demographics - Adult female; 67% Black, 30% Latina Substance Use - Decrease in drug use days in both groups, but no significant between-group differences Violence - Increase in IPV services utilization, but no between-group differences	No inactive control group; Measurement issues for IPV
Hien et al., 2009, 2010* (*Both tested the same intervention)	Seeking Safety; United States	Objective: Reduce PTSD symptoms and substance use Program Components: Cognitive behavioral therapy Skill building Individual + group sessions Provider: Community-based counselors and supervisors	Eligibility: Trauma histories including violence; More than 80% experienced violence in their lifetime (adulthood & childhood) Substance use - current substance misuse Sample size: Intervention group N =176 Control group N= 177 Recruited from community-based substance use treatment programs	Women’s Health Education Group (psychoeducation; 12 sessions) - Attention Matched Control Group	Follow-up: weekly during treatment, 1-week, 3-, 6- and 12-months posttreatment Outcomes assessed: PTSD, substance use, unprotected sexual occasions	Demographics - Adult female; 46% Caucasian, 34% Black HIV-Risk - Seeking safety associated with reduction of HIV sexual risk among women with higher levels of unprotected sex (Hien et al., 2010) Mental Health - PTSD symptoms reduced in both groups, but not significant between-group differences; PTSD improvement associated with subsequent substance use improvement; Substance Use - No changes in abstinence rates	No inactive control group; generalizability; attrition; statistical limitations
McGovern et al., 2011	Integrated Cognitive Behavioral Therapy; United States	Objective: Reduce PTSD symptoms and substance use Program Components: Cognitive behavioral therapy Relapse prevention Coping strategies Social support Individual sessions Provider: Community-based counselors	Eligibility: PTSD screen positive; More than 80% experienced violence in their lifetime (adulthood & childhood) Substance use - enrolled in outpatient substance abuse treatment Sample size: Intervention group N = 32 Control group N= 21 Recruited from community-based intensive outpatient or methadone maintenance programs	10–12 weekly individual addiction counseling sessions with the following modules: treatment initiation, early abstinence, maintaining abstinence, recovery, termination	Follow-up: Baseline, 3- and 6-months postbaseline Outcomes assessed: Substance use days, substance use severity, PTSD, depression	Demographics - 43% female; Adult; 91% Caucasian Mental Health - Reductions in depressive symptoms over follow-up, but no differences between groups. Significantly greater reduction in PTSD re-experiencing symptoms and PTSD diagnosis in the intervention group. Substance Use - Reductions in substance use severity, alcohol and drug use days over time in both groups; significantly greater reduction in drug use days in integrated CBT group	Power/small sample size; Attrition
Mills et al., 2012, 2017* (*Both tested the same intervention)	Concurrent Treatment of PTSD and Substance Use Disorders Using Prolonged Exposure (COPE); Australia	Objective: Reduce PTSD symptoms Program Components: Cognitive behavioral therapy Motivational enhancement Psychoeducation In vivo and imaginal exposure Individual sessions Provider: Clinical psychologist	Eligibility: Past-month PTSD; More than 80% experienced violence in their lifetime (adulthood & childhood) Substance use-Past-month substance misuse Sample size: Intervention group N= 55 Control group N= 48 Recruited from substance use treatment services, provider referral and media advertisements	Usual care: Access to community-based substance use treatment	Follow-up: 6-weeks, 3-months and 9-months post-baseline Outcomes assessed: PTSD symptom severity, severity of substance use disorder, depression, anxiety	Demographics - 62% female; Adult; 6% Aboriginal Mental Health - Reductions in anxiety, depressive and PTSD symptoms over follow-up, but no differences between groups for depression and anxiety. Significantly greater reduction in PTSD in the intervention/COPE group. Substance Use - No significant differences in number of drug classes used or abstinence rates between groups.	Power/small sample size
Reed et al., 2015; Wechsberg et al., 2013* (*Both tested the same intervention)	Women’s Health CoOp (WHC); South Africa	Objective: Address alcohol and other drug use risks, sexual risks for HIV, violence and gender inequality Program Components: Skill building Psychoeducation Safety planning Group sessions Provider: Peer educator	Eligibility: Violence not a criterion but participants had high prevalence of non-partner and partner violence (84% reported in Reed et al.; 51% non-partner and 33% partner) Substance use-Past 3-month drug use Sample size: Intervention group N= 360 Control group N= 181 Nutrition; N= 179 HCT Recruited by peer outreach workers	Two control conditions: Nutrition: Healthy eating and food preparation HIV Counseling and Testing (HCT)	Follow-up: 3-, 6-, 9- and 12-months Outcomes assessed: Substance use, IPV, non-partner violence	Demographics - Adult female; 57% Coloured, 43% Black African HIV-Risk - No difference in unprotected sexual occasions, sex while high or casual sexual partners Substance Use - Increased abstinence at 12-months compared to combined control conditions (Among violence-affected participants at baseline, those no longer reporting violence at follow-up did not differ significantly in drug abstinence compared with those who reported violence at follow up (Reed et al., 2015)) Violence - No difference in physical partner violence	Group imbalances; Statistical power for secondary analyses; attrition
Reif et al., 2002	North Carolina CoOperative (NC CoOp); United States	Objective: Reduce HIV-related risk behaviors and cooccurring distress Program Components: HIV education and counseling Safety planning & danger assessment Skill building Referrals & advice Individual sessions Provider: Case managers	Eligibility: Violence not a criterion but participants had high prevalence of violence victimization (2/3rd reported experiences of violence) Substance use-past 30-day injection drug or crack use Sample size: Intervention group N =122 Control group N= 84 Recruited from a metropolitan area; restricted to women out of treatment	2 risk reduction sessions including HIV education and counseling	Follow-up: 3-months post-treatment Outcomes assessed: Depression, anxiety, violence victimization, substance use	Demographics - Adult female; Predominantly Black Mental Health - Significantly greater reductions in anxiety and overall distress in intervention group. No significant differences in reductions in depression, traumatic stress. Among participants with high cooccurring distress, the intervention significantly reduced depressive symptoms. Violence - No significant differences in victimization between groups.	Attrition and selection bias; Generalizability
Sacks et al., 2008	Dual Assessment and Recovery Track (DART); United States	Objective: Improve functioning related to social and emotional functioning (psychiatric severity, trauma, housing). Substance use intervention effects hypothesized to be comparable in intervention and control condition Program Components: Psychoeducation Skill building Referrals Modified therapeutic community features Group sessions Provider: not specified	Eligibility: Violence not the criterion but all participants had histories of community violence; 97.5% had histories of interpersonal violence Substance use - Substance misuse problems Sample size Intervention group N =126 Control group N= 114 Recruited from outpatient substance use programs	Usual care - intensive outpatient substance use treatment	Follow-up: 12-months Outcomes assessed: Psychiatric severity, trauma, housing, substance use, criminal activity	Demographics - 43% male; Adult; 79% Black Mental Health - Significant reductions in psychological symptoms and depressive symptoms over follow-up period; significant differences between groups in psychological symptoms, but no difference in depressive symptoms Substance Use - Significant reductions in substance use problems and days using drugs for entire sample over follow-up period, but no between-group differences Violence - Significant reductions in community and interpersonal violence over follow-up period, but no between-group differences	Multiple comparisons; Inconsistent results across similar outcomes; attrition; no inactive control group
Tirado-Munoz et al., 2015	Women’s Wellness Treatment; Spain	Objective: Reduce IPV and depressive symptoms Program Components: Cognitive behavioral therapy Motivation enhancement Skill building Group sessions Provider: Clinical psychologist	Eligibility: Past-month intimate partner violence Substance use - Currently receiving treatment for substance misuse Sample size Intervention group N = 7 Control group N= 7 Recruited from a community outpatient substance use treatment program	Usual care (substance use focused) – motivational interviewing, relapse prevention, counseling, medication management	Follow-up: 1-, 3- and 12-months posttreatment Outcomes assessed: IPV, depressive symptoms, quality of life	Demographics - Adult females Mental Health - Significant reductions in depressive symptoms, but no between-group differences Substance Use - No significant differences in drug use between groups; Reduction in alcohol use was significantly greater in experimental group Violence - Significant reductions in IPV in both groups, non-significantly favoring intervention condition. Significant difference in reduction of psychological maltreatment	Statistical power/sample size; Differential attrition; Participation and compliance to study condition activities; No inactive control
Zlotnick et al., 2009	Seeking Safety + TAU; United States	Objective: Reduce substance use, legal problems and psychopathology Program Components: Psychoeducation Coping skills- Empowerment Group sessions Provider: Substance use counselors in the prison	Eligibility: Violence not an eligibility criterion but 94% lifetime sexual violence and 90% lifetime physical violence Full or sub-threshold PTSD Substance use - Substance misuse issues Sample size Intervention group N= 27 Control group N= 22 Recruited from a residential substance use program in a women’s prison	Usual care - individual and group residential prison-based treatment; 12-step model	Follow-up: 12-weeks after intake, 3- and 6-months post-prison release Outcomes assessed: Psychopathology, substance use, legal problems	Demographics - Incarcerated adult females; 47% White; 33% Black Mental Health - Reduction in PTSD symptoms over follow-up, but no significant between-group differences Substance Use - Reduction in substance use severity over follow-up, but no significant between-group differences	No masking – information bias; contamination; differential follow-up protocols between groups; Statisti cal power

CBT: Cognitive-behavioral therapy; ED: Emergency department; HCT: HIV counseling and testing; ICBT: Integrated cognitive behavioral therapy; IPV: Intimate partner violence; PTSD: Post-traumatic stress disorder; SBIRT: Screening, brief intervention and referral to treatment; SUD: substance use disorder; TAU: Treatment as usual; USO: unprotected sexual occasions.

2.2.2. Exclusion criteria

Studies were excluded if they (1) were single-disorder or single component interventions (i.e. evaluated an intervention with only one component); (2) were not full RCTs or just described the protocol of the intervention; (3) did not evaluate two or more of the following syndemic outcomes in combination: violence, HIV risk, substance misuse, and mental health; (4) were not published in English; (5) focused on participants under 18 years of age; and (6) did not include participants who reported victimization by violence or focused only on perpetrators of violence (e.g., IPV batterers).

2.3. Study selection

We first reviewed the titles and abstracts from all search results. Studies that described an intervention for victimized substance-using individuals were retained for full text reviews. Full texts were reviewed against the pre-specified eligibility criteria. We screened the references from articles retained for full text review for reports that were missed in the search strategy. Articles that met eligibility criteria were included in the synthesis

3. Results

3.1. Description of studies

3.1.1. Search results

Electronic searches identified 13,948 results from academic literature databases (PubMed n = 2790, PsycINFO n = 3653, CENTRAL n = 688, Web of Science n = 2785, Embase n = 3217, CINAHL n = 815) and 9 articles identified through cross-referencing included articles. Once duplicates were removed (n = 5922), we reviewed the titles and abstracts of 8035 articles. Sixty-one of these were retained for full text review. After the full texts of these articles were reviewed against our eligibility criteria, we determined that 17 articles (including 14 interventions) were relevant (Table 1).

3.1.2. Study setting

Most trials were conducted in the US (n = 13) or other high-income countries (Australia, n = 2; Spain, n = 1) (e.g., Barrett et al., 2015; Choo et al., 2016; Garland, Roberts-Lewis, Tronnier, Graves, & Kelley, 2016; Reif, Wechsberg, & Dennis, 2002; Sacks, McKendrick, Sacks, Banks, & Harle, 2008; Tirado-Munoz, Gilchrist, Lligona, Gilbert, & Torrens, 2015; Zlotnick, Johnson, & Najavits, 2009). One intervention was tested in South Africa, an upper-middle-income country (n = 1) (Reed, Myers, Novak, Browne, & Wechsberg, 2015; Wechsberg et al., 2013). Participants were recruited from correctional facilities or the criminal justice system (Barrett et al., 2015; Gilbert et al., 2015), emergency departments (Choo et al., 2016), substance abuse treatment and community services (e.g., Garland et al., 2016; Ghee et al., 2009; Gilbert et al., 2006; Hien et al., 2009, 2010), or community-based outreach (Reed et al., 2015; Wechsberg et al., 2013).

3.1.3. Sample characteristics

Eleven studies exclusively enrolled females (n = 9) or males (n = 2) (e.g., Barrett et al., 2015; Choo et al., 2016; Garland et al., 2016; Ghee et al., 2009; Gilbert et al., 2006). Caucasian and African American represented the racial/ethnic majority in almost half of the trials (e.g., Choo et al., 2016; Garland et al., 2016; Reif et al., 2002; Sacks et al., 2008).

The sample sizes ranged from 14 to 720 participants. Substance use was variably defined in the studies, including history of problematic substance use (Barrett et al., 2015), recent substance use (Choo et al., 2016; Gilbert et al., 2006, 2015;Reed et al., 2015; Reif et al., 2002; Wechsberg et al., 2013), substance abuse disorder(Garland et al., 2016; Hien et al., 2009, 2010; McGovern et al., 2011; Mills et al., 2012, 2017; Sacks et al., 2008; Zlotnick et al., 2009), or currently receiving treatment for substance use (Ghee et al., 2009;McGovern et al., 2011; Tirado-Munoz et al., 2015).

3.2. Description of integrated multicomponent interventions

All interventions in the experimental condition were multicomponent, integrated interventions addressing two or more health outcomes: violence, mental health, substance misuse and HIV risk. The interventions were psychological (e.g. integrated cognitive behavioral therapy (CBT), CBT combined with cognitive processing therapy) (Barrett et al., 2015; Hien et al., 2009, 2010; McGovern et al., 2011; Zlotnick et al., 2009), mindfulness (Garland et al., 2016), relapse prevention and relationship safety(Gilbert et al., 2015), therapeutic communities(Sacks et al., 2008), peer- and case manager-administered brief interventions(Reed et al., 2015; Reif et al., 2002; Wechsberg et al., 2013), and web-based brief interventions(Choo et al., 2016; Gilbert et al., 2015).

Seeking Safety was the most common intervention and was evaluated as the experimental condition in five studies(Barrett et al., 2015; Ghee et al., 2009; Hien et al., 2009, 2010; Zlotnick et al., 2009) and as the control condition in one study(Garland et al., 2016). Seeking Safety represents an integrated CBT intervention that aims to reduce trauma related problems and substance use through psychoeducation and coping skills training (Najavits, 2002). Components of other integrated CBT interventions included managing negative cognitions and emotions (McGovern et al., 2011; Tirado-Munoz et al., 2015), relaxation (McGovern et al., 2011), motivational interviewing (Mills et al., 2012, 2017; Tirado-Munoz et al., 2015), in vivo and imaginal exposure (Mills et al., 2012, 2017), managing substance use triggers and relapse prevention (McGovern et al., 2011; Tirado-Munoz et al., 2015), healing from violence, sexual boundaries, and HIV/HCV risk (Tirado-Munoz et al., 2015).

The mindfulness intervention included elements of CBT, positive psychology, targeting automatic habit behavior, non-reactivity, positive reappraisal and emotion regulation (Garland et al., 2016). The relapse prevention and relationship safety intervention included social cognitive skill building, and empowerment (Gilbert et al., 2006). The Dual Assessment and Recovery Track (DART) Program represented a therapeutic community that provided psychoeducation, trauma-informed substance abuse treatment, and case management services (Sacks et al., 2008). One of the two brief peer/provider-delivered interventions was a two-session group-based intervention and included a discussion of topics such as health information, skills training to reduce health risks, relationship power and negotiation and violence prevention (Reed et al., 2015; Wechsberg et al., 2013). The other brief intervention was an individual five-session program with HIV education and counseling, risk assessment, problem solving and communication components (Reif et al., 2002). Lastly, the web-based psychoeducation programs included feedback on substance misuse, motivation, empowerment, violence, safety planning, goal setting and social support (Choo et al., 2016; Gilbert et al., 2015). One of these interventions provided a brief follow-up call two weeks after completing the web-based intervention to further discuss goals and other topics introduced in the program (Choo et al., 2016).

3.2.1. Description of control conditions

Ten RCTs compared the experimental intervention to another active substance abuse treatment program, which was most commonly a CBT or a standard outpatient treatment as usual (Barrett et al., 2015; Garland et al., 2016; Ghee et al., 2009; Gilbert et al., 2015; McGovern et al., 2011; Mills et al., 2012, 2017; Sacks et al., 2008; Tirado-Munoz et al., 2015; Zlotnick et al., 2009). Six studies included a time/attention-matched control condition including a web-based fire safety program (Choo et al., 2016), informational sessions/groups (Gilbert et al., 2006; Hien et al., 2009, 2010), a nutrition program (Reed et al., 2015; Wechsberg et al., 2013), and HIV counseling and testing (Reed et al., 2015;Reif et al., 2002; Wechsberg et al., 2013).

3.2.2. Cultural considerations

Three trials mentioned efforts to adapt the experimental intervention to the culture and context relevant to their target population: a web-based intervention with tailored relapse prevention program to reflect the realities of African American and Latina women such as gender inequalities, attitudes towards safe sexual practices (Gilbert et al., 2006), an integrated CBT intervention adapted to meet the health and safety needs (e.g., reduced IPV and depressive symptoms) of people who inject drugs in Spain (Tirado-Munoz et al., 2015), and a brief intervention adapted through formative focus groups with women that use drugs in South Africa (Reed et al., 2015; Wechsberg et al., 2013). Cultural factors (e.g., values, beliefs and cultural practices) are important considerations for effective intervention strategies with diverse groups of people who use substances (Table 2.)

Table 2.

Summary of results.

Authors	Name of the intervention	Control	Significance of syndemic outcomes
			Violence reduction/future revictimization	Mental health	HIV risk (Sexual HIV risk and drug related HIV risk)	Substance misuse
Barrett et al., 2015; Hien et al., 2009, 2010; Zlotnick et al., 2009	Seeking Safety	Treatment as Usual (Opioid Substitution Treatment & Prison-based residential substance use treatment) -Women’s Health Information Groups	NA	vs. Treatment as Usual: Reduction in PTSD but no significant between group differences vs. Women Health Information Group Reduction in PTSD but no differences between Seeking Safety & control	vs. Women Health Information Group Reduction of sexual HIV risk	vs. Treatment as Usual: Increased confidence in ability to resist substances; Reduction in substance use severity but no between group differences vs. Women Health Information Group Not significant for abstinence
Choo et al., 2016	BSAFER	Web-based program on fire safety	Reduction in IPV in both groups	NA	NA	Reduction in drug use days in both groups
Garland et al., 2016	Mindfulness-Oriented Recovery Enhancement	Seeking Safety Treatment as Usual	NA	vs. Seeking Safety: Reduced PTSD vs. Treatment as Usual: Reduced PTSD	NA	vs. Seeking Safety: Reduced cravings vs. Seeking Safety: Not significant
Ghee et al., 2009	Condensed Seeking Safety Intervention	Treatment as Usual	NA	Reduced sexual abuse related trauma but not more advantageous in reducing overall trauma or relapse 30 days posttreatment	NA	Not significant in reducing relapse 30days post treatment
Gilbert et al., 2006	Relapse Prevention and Relationship Safety (RPRS)	Informational Control	Reduction in minor physical, sexual or injurious IPV, any severe physical IPV as well as minor or severe psychological IPV	Reduced depression; Reduced PTSD but no significant differences between RPRS & control	Reduction in sex while high on drugs; Not significant for number of unprotected sexual occasions or multiple sex partners	Decrease in drug use, binge drinking or crack cocaine but effects not significant
Gilbert et al., 2015	Computer-Administered Project WINGS (Women Initiating New Goals of Safety; SBIRT); United States	Case Manager-Administered Project WINGS (Women Initiating New Goals of Safety; SBIRT)	Increase in IPV services utilization, but no differences between Project WINGS & control	NA	NA	Reduction in drug use days, but no significant differences between Project WINGS & control
Mills et al., 2012, 2017	COPE	Treatment as Usual	NA	Reductions in depression, PTSD & anxiety but no significant differences between COPE & control on depression & anxiety	NA	Not significant
Reed et al., 2015; Wechsberg et al., 2013	Women’s Health CoOp (WHC)	Nutrition HIV Counseling & Testing	Not significant	NA	Not significant	Increased abstinence
Reif et al., 2002	North Carolina Cooperative (NC CoOp)	Risk reduction intervention	Not significant	Reduced overall distress & anxiety	NA	NA
Sacks et al., 2008	Dual Assessment and Recovery Track (DART)	Intensive outpatient substance use treatment	Reduction in exposure to community or interpersonal violence but no differences between DART & control group	Reduced psychological/emotional problems	NA	Reduced substance problems & drug using days but no significant difference between DART & control
Tirado-Munoz et al., 2015	Women’s Wellness Treatment; Spain	Outpatient substance use treatment	Reduction in IPV but only significant difference between intervention and control group was psychological IPV	Reduced depression but intervention & control group did not differ significantly	NA	Reduction in alcohol use; Intervention & control group did not differ in use of drugs

3.2.3. Interventionists

The providers included clinical psychologists (Barrett et al., 2015; Mills et al., 2012, 2017; Tirado-Munoz et al., 2015), social workers (Garland et al., 2016), community-based and substance abuse counselors (Hien et al., 2009, 2010; McGovern et al., 2011; Zlotnick et al., 2009), trained facilitators (Gilbert et al., 2006), peer educators (Reed et al., 2015; Wechsberg et al., 2013), and case managers (Reif et al., 2002). Two trials used web-based programs (Choo et al., 2016; Gilbert et al., 2015).

3.3. Effective components and syndemic outcomes

3.3.1. Violence

Seven trials aimed to evaluate interventions that would reduce violence (Barrett et al., 2015; Choo et al., 2016; Gilbert et al., 2006, 2015; Reed et al., 2015; Sacks et al., 2008; Tirado-Munoz et al., 2015; Wechsberg et al., 2013). The most common form of violence that was targeted was IPV, which was operationalized as the presence of any IPV or specific subtypes (e.g. physical, psychological, sexual) using the Revised Conflict Tactics Scale (CTS2) (Gilbert et al., 2006, 2015), the Composite Abuse Scale (Choo et al., 2016; Tirado-Munoz et al., 2015), or the Psychological Maltreatment of Women Inventory (Gilbert et al., 2015; Tirado-Munoz et al., 2015). Other forms of violence measured were community violence (Sacks et al., 2008), sexual abuse (Ghee et al., 2009) aggressive behavior (Reif et al., 2002), general victimization and interpersonal violence (Reif et al., 2002; Sacks et al., 2008). Two trials found that an individual and group-based relapse prevention program, which integrated elements of relationship safety, and a group-based integrated CBT intervention were effective in reducing experiences of psychological abuse, but did not find a treatment effect for physical, sexual or any IPV (Gilbert et al., 2006; Tirado-Munoz et al., 2015). Similarly, two brief peer-led interventions in South Africa and the U.S. were not found to reduce the prevalence of physical IPV (Reif et al., 2002; Wechsberg et al., 2013). A comparison of a web-based vs. provider-delivered brief intervention found that both groups were more likely to utilize IPV services postintervention, but there were no between-group differences (Gilbert et al., 2015). A web-based brief intervention found preliminary evidence of a reduction in IPV in both the intervention and treatment as usual condition (Choo et al., 2016). Sacks and colleagues evaluated the effect of an enhanced outpatient substance abuse treatment program as compared to treatment as usual and found that participants in both study conditions were less likely to experience community or interpersonal violence over the follow-up period, but this change did not differ between groups (Sacks et al., 2008).

3.3.1.1. Components of interventions that included violence as an outcome for victimized substance-using individuals

The interventions for violence that showed improved outcomes for victimized substance-using individuals were both online and in-person interventions and were delivered in an individual and/or group format. The online interventions that had positive outcomes for violence, specifically IPV, included elements such as psycho education, motivational enhancement to improve relationship safety, empowerment, safety planning, social support, goal setting, and referrals and advice (Choo et al., 2016; Gilbert et al., 2015). The effective components for in-person interventions were goal setting to reduce or prevent violence, assessment of risk and safety planning, psychoeducation, enhancing social support, empowerment, motivational enhancement for relationship safety, coping skills to deal with trauma, and identification of service needs and referrals or linkage to community resources.

3.3.2. HIV risk

HIV risk outcomes included condom use with main partner, multiple sexual partners, casual sexual partners, sex while intoxicated/high and unprotected sex. These outcomes were measured using self-reported indicators or measurement tools (e.g. Sexual Risk Behavior Questionnaire; Risk Behavior Survey) in three studies (Gilbert et al., 2006; Hien et al., 2010; Wechsberg et al., 2013). Two of these trials found that integrated CBT and relapse prevention and relationship safety interventions were effective at reducing HIV risk. More specifically, one study found that an integrated CBT intervention reduced the number of unprotected sexual occasions among women with a high number of unprotected sexual occasions at baseline (Hien et al., 2010). An integrated relapse prevention and relationship safety intervention delivered to women receiving methadone maintenance treatment was found to significantly reduce the proportion of women who reported engaging in sex while high but did not find a significant effect of treatment on number of unprotected sexual occasions or number of sexual partners (Gilbert et al., 2006). The third trial, a brief peer-delivered intervention did not find a treatment effect on unprotected sexual occasions, sex while high on substances or casual sexual partners (Wechsberg et al., 2013). Out of three interventions that aimed to reduce HIV risk, the two that were found to be effective included mental health components such as CBT and skill building components surrounding HIV and sexual health risk reduction (Gilbert et al., 2006; Hien et al., 2010).

3.3.2.1. Components of interventions that included HIV risk as an outcome for victimized substance-using individuals

The multicomponent interventions that showed promising findings for reducing HIV risk among victimized substance using individuals included components such as increasing knowledge about sexual risk for HIV, social cognitive skill building: building self-efficacy, safe coping strategies, boundary setting with a high risk partner, negotiation and communication skills related to condom use, and safety planning with HIV prevention an aspect of relationship safety (Gilbert et al., 2006; Hien et al., 2010).

3.3.3. Substance use

Ten trials tested interventions that were intended to improve substance abuse outcomes (e.g.,Barrett et al., 2015; Choo et al., 2016). Most studies included in this review identified reductions in substance misuse across intervention types, but the magnitude of these reductions did not differ between groups possibly due to the numerous comparative effectiveness designs. These interventions included integrated CBT, brief interventions, mindfulness and relapse prevention. The most common intervention aimed to improve substance use outcomes, Seeking Safety, was tested in six trials in the U.S. and Australia, once as the control condition.

Substance abuse outcomes included abstinence, days of alcohol/drug use, alcohol/drug use severity and number of drug classes used. Severity was measured using the Addiction Severity Index (McGovern et al., 2011; Zlotnick et al., 2009), the Drug and Alcohol Use Behavior Questionnaire (Gilbert et al., 2006), and the GAIN Substance Problem Index (Sacks et al., 2008). Substance misuse and abstinence were measured by self-report sometimes assisted by structured measurement tools (e.g. Timeline Followback) (Choo et al., 2016; McGovern et al., 2011; Zlotnick et al., 2009) or were biologically verified (Ghee et al., 2009; Hien et al., 2009; McGovern et al., 2011; Wechsberg et al., 2013).

Three trials did not find significant differences between groups in days of substance misuse, severity or abstinence between study conditions over the follow-up period (Gilbert et al., 2006; Hien et al., 2009; Mills et al., 2012). Four studies found that number of days using substances, substance misuse problems and severity of substance misuse post-treatment was significantly lower for both groups post-treatment (Gilbert et al., 2015; Hien et al., 2009; Sacks et al., 2008; Zlotnick et al., 2009); however, at later follow-up periods (e.g. 9- or 12-months), the number of days using drugs, abstinence rates or severity of substance misuse moved towards baseline levels in three studies (Hien et al., 2009). One study of integrated CBT incorporating relapse prevention, coping strategies and social support components compared to individual addiction counseling found similar between-group decreases in severity of substance misuse, toxicology measures and alcohol use days over the follow-up period; however, the decrease in number of drug use days was significantly larger in the integrated CBT condition (McGovern et al., 2011). Two pilot studies of integrated CBT (vs. treatment as usual) and a brief web-based intervention (vs. time/attention-matched control) provide preliminary evidence of greater confidence in one’s ability to resist substances and fewer drug use days, respectively, in the intervention group (Barrett et al., 2015; Choo et al., 2016). Evaluation of a peer-delivered brief intervention found that participants in the experimental condition had 1.5-fold greater odds of abstinence from substances relative to participants in either of the time/attention-matched control conditions (Wechsberg et al., 2013).

One study found an effect of integrated CBT on reduced alcohol, but not drug use. This intervention included motivational enhancement and skill building delivered by a clinical psychologist through group sessions (Tirado-Munoz et al., 2015). The effect of a mindfulness intervention on substance use craving was evaluated in an RCT with two control groups, integrated CBT (i.e., Seeking Safety) and treatment as usual. The mindfulness intervention was found to reduce craving to a significantly greater extent than the integrated CBT condition. There were, however, no significant differences observed between the mindfulness and treatment as usual conditions (Garland et al., 2016).

3.3.3.1. Components of interventions that included substance abuse as an outcome for victimized substance-using individuals

The integrated intervention trials that were effective for substance misuse included trials delivered in both web-based and in-person formats. The elements of web-based interventions were goal setting, psychoeducation, empowerment, social support, automatic feedback on drug use and health, motivational enhancement and referral and advice (Choo et al., 2016; Gilbert et al., 2015). The group-based or a combination of individual and group-based interventions used components such as psychoeducation on substance use disorders, motivational enhancement, mindfulness training to target areas such as automaticity in addiction, regulating craving, cognitive behavioral therapy, cognitive restructuring with attention to maladaptive thoughts related to substance misuse, skill building to prevent substance misuse, increasing social support or developing effective communication skills to build healthy support networks, empowerment, and access to services to reduce substance misuse (e.g., Garland et al., 2016; Gilbert et al., 2006; Hien et al., 2010; Mills et al., 2012; Tirado-Munoz et al., 2015; Zlotnick et al., 2011). Psychoeducation, cognitive behavioral therapy, motivational enhancement, coping skills training, and social support were also common elements of interventions delivered in individual sessions (Barrett et al., 2015;Mc-Govern et al., 2011; Mills et al., 2012)

3.3.4. Mental health

Nine multicomponent interventions targeted mental health among victimized substance-using individuals using a combination of approaches: web-based or in-person individual and/or group sessions. These RCTs included evaluations of mindfulness interventions, integrated CBT, brief interventions, and therapeutic communities (Garland et al., 2016; Ghee et al., 2009; Hien et al., 2009, 2010; McGovern et al., 2011; Mills et al., 2012, 2017; Reif et al., 2002; Sacks et al., 2008; Tirado-Munoz et al., 2015; Zlotnick et al., 2009). Five of these interventions were group based, four were delivered on an individual basis, and two were combined (group + individual sessions). MH outcomes included anxiety, depression, PTSD and general distress.

3.3.4.1. Depression and anxiety

Anxiety symptomatology was most commonly measured using the Brief Symptom Inventory (BSI) and the State-Trait Anxiety Inventory (Garland et al., 2016; Mills et al., 2012, 2017). Depressive symptoms were commonly assessed using the Brief Symptom Inventory (Garland et al., 2016; Gilbert et al., 2006), as well as the Beck Dépression Inventory (McGovern et al., 2011; Mills et al., 2012, 2017; Sacks et al., 2008; Tirado-Munoz et al., 2015). In five studies evaluating the efficacy of mindfulness interventions, CBT or therapeutic communities with active control conditions, depressive and/or anxiety symptoms improved over the ten weeks of treatment, but there were no significant differences between study conditions (Garland et al., 2016; McGovern et al., 2011; Mills et al., 2012; Sacks et al., 2008; Tirado-Munoz et al., 2015). One study of a brief intervention delivered by a case manager found that compared to two sessions of risk reduction, the addition of three supplementary sessions was associated with significantly greater reductions in anxiety and overall distress but did not impact depression. No significant reductions in mental health symptoms were observed in the standard 2-session control condition (Reif et al., 2002). Only one intervention - a relapse prevention program that focused on social cognitive skill building, safety planning, empowerment and social support – was found to reduce depressive symptoms relative to the comparison condition (Gilbert et al., 2006).

3.3.4.2. Post-traumatic stress disorder (PTSD)

PTSD, including traumatic stress, was measured in some trials using the Clinical-Administered PTSD Scale (Barrett et al., 2015; Hien et al., 2009;McGovern et al., 2011; Mills et al., 2012, 2017), PTSD Checklist – Civilian Version (Garland et al., 2016; Gilbert et al., 2006), and PTSD Symptom Scale – Interview (Hien et al., 2009). Two studies, one with an active control condition and the other with a time/attention-matched control women’s health education group, found clinically significant reductions in PTSD symptoms over the follow-up period; however, the magnitude of this reduction was similar between study conditions (Hien et al., 2009; Zlotnick et al., 2009). A study of integrated CBT found that PTSD severity and diagnosis decreased over the follow-up period for both the integrated CBT as well as the addiction counseling control group; however, there was also a significant group by time interaction such that PTSD re-experiencing symptoms and PTSD diagnosis were lower in the integrated CBT group relative to the individual addiction counseling control group (McGovern et al., 2011).Comparative effectiveness studies indicated that participants randomized to a mindfulness intervention (vs. Seeking Safety) or integrated CBT (vs. individual addiction counseling or treatment as usual) experienced significantly greater reductions in PTSD symptoms relative to participants assigned to the control condition on one or more PTSD indicators or time points (Garland et al., 2016; Reif et al., 2002). These interventions included mindfulness stress-reduction techniques, CBT strategies, motivational enhancement, psychoeducation, and in vivo and imaginal exposure components (Garland et al., 2016; McGovern et al., 2011; Mills et al., 2012, 2017).

3.3.4.3. Components of interventions that targeted mental health among victimized substance-using individuals

The common promising elements of interventions for improving mental health of victimized substance-using individuals included psychoeducation for improving understanding of symptoms of mental health, CBT, coping skills training, managing negative moods, motivational enhancement, and referrals for interventions delivered in group based (Sacks et al., 2008; Tirado-Munoz et al., 2015; Zlotnick et al., 2011), individual (McGovern et al., 2011; Mills et al., 2012, 2017) and a combination of individual and group sessions format (Hien et al., 2009, 2010). For instance, a group-based intervention used cognitive restructuring/training in positive reappraisal to regulate negative emotions (Garland et al., 2016), Interventions delivered in an individual format used psychoeducation about mental health, and cognitive therapy techniques (McGovern et al., 2011; Mills et al., 2012) for improved outcomes in mental health. For example, breathing retraining for anxiety reduction, cognitive restructuring for identifying stressful activating situations, beliefs/thoughts and consequences, and skill development for disrupting negative beliefs and generating alternative emotions or behaviors (McGovern et al., 2011).

4. Discussion

Co-occurring problems can synergistically contribute to serious health concerns and inequities among victimized substance-using individuals. This co-occurrence is associated with more frequent substance misuse, can affect responsiveness to substance abuse interventions and maintenance of post-intervention abstinence of drugs (Sabri, 2012). Thus, victimized substance using individuals with cooccurring health issues need comprehensive, integrated, multicomponent interventions that are designed to concurrently address multiple health issues simultaneously. Interventions that include one component and target one outcome do not account for multiple epidemics faced by survivors of violence (Sabri & Gielen, 2018) (e.g., mental health issues, risky sexual behaviors, more frequent use of substances to cope with trauma). Since multiple syndemic health issues among victimized individuals are inter-related, interventions focusing on any one of the issues in isolation may be less effective than a comprehensive integrated approach (Kimerling & Goldsmith, 2000). A strong evidence base, therefore, is needed for multicomponent interventions that target syndemic conditions among victimized substance-using individuals. Addressing violence/trauma in their lives is of critical importance, since addressing violence and related mental health symptoms can also address other co-morbid issues such as HIV and risk for future violence. We, however, found only 17 trials that evaluated integrated interventions incorporating components to address needs of victimized substance-using individuals. Almost all studies were conducted in high income countries. This highlights a critical gap in evidence for comprehensive, integrated, multicomponent interventions for victimized substance-using individuals, particularly those in developing countries.

In our review of interventions, we identified intervention components (e.g., skills training, psychoeducation) that targeted specific outcomes and were found to be effective in improving multiple outcomes for victimized substance-using individuals. For instance, if the intervention study reported reduced frequency of drug use, the component of the intervention that addressed addiction problems was considered effective. Despite variability in the interventions, there were many common components across studies such as CBT strategies, skill building, psychoeducation, safety planning and/or danger assessment, referrals and advice, empowerment and goal setting.

The effective intervention components for violence were goal setting, psychoeducation, social support, empowerment strategies, risk assessments, safety planning and referrals/linkage to community resources. These effective interventions primarily focused on IPV. For mental health, particularly PTSD, the effective components were CBT, coping skills training with other skill building components, psychoeducation, social support, and referrals (Garland et al., 2016; McGovern et al., 2011; Mills et al., 2012, 2017). For HIV risk reduction, substance abuse interventions that addressed trauma and included components such as CBT, building self-efficacy, negotiation skills, social cognitive skill-building with individual and group sessions were found to reduce HIV risk (Gilbert et al., 2006; Hien et al., 2010). The substance use outcomes were improved for interventions that were brief, delivered by peer educators, integrated CBT (traditional CBT elements, relapse prevention, coping strategies and social support), mindfulness strategies, skill building, psychoeducation, and motivational enhancement (McGovern et al., 2011; Reed et al., 2015; Wechsberg et al., 2013). Regarding cultural considerations, few studies cited culture as an important element in delivery of the intervention (Gilbert et al., 2006; Reed et al., 2015; Tirado-Munoz et al., 2015; Wechsberg et al., 2013).

Although some interventions in this review appeared promising in addressing multiple outcomes for victimized substance-using individuals, the limitations in design and methods highlight the need for stronger interventions evaluated with more rigorous methods. The trials in this review had several limitations that may affect the results. Some studies had sample sizes of less than 60 participants, which suggests that these studies may have been underpowered to detect a clinically meaningful treatment effect. Several of these small studies did not report inferential statistics due to low power. Study follow-up was generally short, ranging from 1 to 12-months, and seven studies suffered from high rates of attrition, further reducing power and potentially introducing emigrative selection bias. Additionally, studies often targeted specific populations, including persons that were incarcerated, homeless, veterans, and patients in the emergency room, which limits the generalizability of their findings to other populations.

The study contributes to the literature by providing the first systematic synthesis of multicomponent interventions targeting syndemic conditions among victimized substance-using individuals. Furthermore, it describes and identifies the elements of these interventions that may serve as critical components for effectively addressing co-occurring healthcare needs of victimized substance-using individuals. It highlights the need to establish a strong evidence base for multicomponent interventions that address multiple syndemic conditions among victimized substance using individuals. Further, interventions are needed that are culturally informed and meet the needs of diverse groups of victimized individuals who use substances.

Appendix A.

PubMed

(“mental disorder” OR “mental health” OR “mental illness” OR depression OR anxiety OR.

“post-traumatic stress” OR psychological OR psychiatric OR violence OR abuse OR HIV) AND ((randomized controlled trial[pt] OR controlled clinical trial[pt] OR randomized[tiab] OR randomised[tiab] OR placebo[tiab] OR randomly[tiab] OR trial[tiab] OR groups[tiab]) NOT (Animal[mesh] NOT human[mesh])) AND (“injection drug use” OR “injection drug” OR

“people who inject drugs” OR PWID OR IDU OR drug OR heroin OR opiate OR opioid) AND (violence OR assault OR trauma OR “intimate partner violence” OR “spouse abuse” OR “spousal abuse” OR “abused spouse” OR “abused spouses” OR “battered women” OR “battered woman” OR “domestic violence” OR “Sexual abused” OR “sexually abused” OR “sexual abuse” OR

“Physical abuse” OR “Child abuse” OR “abused in childhood” OR “child sexual abuse” OR rape OR beating OR traumatized OR violence-affected OR victimized)

PsycINFO

(“mental disorder” OR “mental health” OR “mental illness” OR depression OR anxiety OR

“post-traumatic stress” OR psychological OR psychiatric OR violence OR abuse OR HIV) AND (“injection drug use” OR “injection drug” OR “people who inject drugs” OR PWID OR IDU OR drug OR heroin OR opiate OR opioid) AND (violence OR assault OR trauma OR “intimate partner violence” OR “spouse abuse” OR “spousal abuse” OR “abused spouse” OR “abused spouses” OR “battered women” OR “battered woman” OR “domestic violence” OR “Sexual abused” OR “sexually abused” OR “sexual abuse” OR “Physical abuse” OR “Child abuse” OR “abused in childhood” OR “child sexual abuse” OR rape OR beating OR traumatized OR violence-affected OR victimized) AND (PT randomized controlled trial OR PT controlled clinical trial OR AB randomized OR AB randomised OR AB placebo OR AB randomly OR AB trial OR AB groups NOT AB animals)

Cochrane CENTRAL

(“mental disorder” OR “mental health” OR “mental illness” OR depression OR anxiety OR

“post-traumatic stress” OR psychological OR psychiatric OR violence OR abuse OR HIV) AND (“injection drug use” OR “injection drug” OR “people who inject drugs” OR PWID OR IDU OR drug OR heroin OR opiate OR opioid) AND (violence OR assault OR trauma OR “intimate partner violence” OR “spouse abuse” OR “spousal abuse” OR “abused spouse” OR “abused spouses” OR “battered women” OR “battered woman” OR “domestic violence” OR “Sexual abused” OR “sexually abused” OR “sexual abuse” OR “Physical abuse” OR “Child abuse” OR “abused in childhood” OR “child sexual abuse” OR rape OR beating OR traumatized OR violence-affected OR victimized)

ISI Web of Science

(“mental disorder” OR “mental health” OR “mental illness” OR depression OR anxiety OR

“post-traumatic stress” OR psychological OR psychiatric OR violence OR abuse OR HIV) AND

(“injection drug use” OR “injection drug” OR “people who inject drugs” OR PWID OR IDU OR drug OR heroin OR opiate OR opioid) AND (violence OR assault OR trauma OR “intimate partner violence” OR “spouse abuse” OR “spousal abuse” OR “abused spouse” OR “abused spouses” OR “battered women” OR “battered woman” OR “domestic violence” OR “Sexual abused” OR “sexually abused” OR “sexual abuse” OR “Physical abuse” OR “Child abuse” OR “abused in childhood” OR “child sexual abuse” OR rape OR beating OR traumatized OR violence-affected OR victimized) AND (“randomized controlled trial” OR “controlled clinical trial” OR randomized OR randomised OR placebo OR randomly OR trial OR groups NOT animals)

Embase

(“mental disorder” OR “mental health” OR “mental illness” OR depression OR anxiety OR

“post-traumatic stress” OR psychological OR psychiatric OR violence OR abuse OR HIV) AND(“injection drug use” OR “injection drug” OR “people who inject drugs” OR PWID OR IDU OR drug OR heroin OR opiate OR opioid) AND (violence OR assault OR trauma OR “intimate partner violence” OR “spouse abuse” OR “spousal abuse” OR “abused spouse” OR “abused spouses” OR “battered women” OR “battered woman” OR “domestic violence” OR “Sexual abused” OR “sexually abused” OR “sexual abuse” OR “Physical abuse” OR “Child abuse” OR “abused in childhood” OR “child sexual abuse” OR rape OR beating OR traumatized OR violence-affected OR victimized) AND (randomized:it AND controlled:it AND trial:it OR (controlled:it AND clinical:it AND trial:it) OR randomized:ab,ti OR randomised:ab,ti OR placebo:ab,ti OR randomly:ab,ti OR trial:ab,ti OR groups:ab,ti NOT animals:ab,ti)

CINAHL

(“mental disorder” OR “mental health” OR “mental illness” OR depression OR anxiety OR

“post-traumatic stress” OR psychological OR psychiatric OR violence OR abuse OR HIV) AND (“injection drug use” OR “injection drug” OR “people who inject drugs” OR PWID OR IDU OR drug OR heroin OR opiate OR opioid) AND (violence OR assault OR trauma OR “intimate partner violence” OR “spouse abuse” OR “spousal abuse” OR “abused spouse” OR “abused spouses” OR “battered women” OR “battered woman” OR “domestic violence” OR “Sexual abused” OR “sexually abused” OR “sexual abuse” OR “Physical abuse” OR “Child abuse” OR “abused in childhood” OR “child sexual abuse” OR rape OR beating OR traumatized OR violence-affected OR victimized) AND (PT randomized controlled trial OR PT controlled clinical trial OR AB randomized OR AB randomised OR AB placebo OR AB randomly OR AB trial OR AB groups NOT AB animals)