Stephens-Lewis et al.4 provides more information.
Background
Reviews of IPV perpetrator interventions for men have shown mixed results and small effect sizes,93–97 with better outcomes when interventions included substance abuse or trauma components.96 No single approach can be definitively supported.98 Perpetrators receiving interventions with motivational strategies were almost twice as likely to be retained.99
Men who misuse substances are rarely referred to perpetrator programmes,100–102 as many are considered unsuitable.102 They are also most likely to drop out of perpetrator programmes100,103–105 and may require alternative approaches.94
Aims
We conducted a systematic narrative review with meta-analysis to determine the effectiveness of interventions to reduce IPV perpetration by men who use substances.
Methodology
The review was conducted following PRISMA protocols46 and prospectively registered (PROSPERO 2017: CRD42017056596).
Search strategy
Search terms related to ‘IPV’ AND ‘substance use’ AND ‘intervention’. Searches were performed to May 2018 in MEDLINE, EMBASE, Cumulative Index to Nursing and Allied Health Literature, PsycInfo, Social Sciences Citation Index, International Bibliography of the Social Sciences and Social Services Abstracts, with an update in MEDLINE to April 2019. Clinical trial databases were also searched (National Institute for Health Research Register, www.who.int/ictrp/ and National Health; and Medical Research Council, Australia).
Inclusion criteria and screening
We applied population, intervention, comparison, outcome and design to develop the inclusion criteria and used the hierarchy assessment method of eligibility.106 Randomised controlled trials (RCTs) or non-RCTs (Design), published in any language, were eligible for inclusion if the:
- intervention targeted IPV or relationships among adult heterosexual males (Population)
- sample included > 60% who currently drank hazardously or who met criteria for alcohol or drug abuse/dependence (Population)
- intervention was compared to either IPV perpetrator or substance use treatment as usual (TAU) or an IPV intervention of a lesser intensity or frequency (Intervention, Control); and
- outcomes included perpetrator and/or survivor reports of IPV perpetration, and/or substance use, and/or marital satisfaction/conflict (Outcome).
Titles and abstracts were reviewed. Full texts were then assessed against eligibility criteria.
Quality assessment
The Cochrane Effective Practice and Organisation of Care tool107 confirmed trials were conducted with low risk of bias.
Data extraction, synthesis and analysis
Data were extracted into tables for analysis. Intervention types were grouped and analysed separately to determine the superiority of: integrated IPV and substance use interventions (i.e. interventions addressing both together) (five trials), IPV interventions with adjunct substance use interventions (i.e. with a separate session or intervention addressing substance use) (two trials) and stand-alone IPV interventions (i.e. IPV interventions not addressing substance use) (two trials).
Comparisons were made between intervention and control group data, with at least two RCTs with combinable outcome data.106 A meta-analysis using a random-effects model was performed using Review Manager (RevMan) (version 5.1, The Cochrane Collaboration, 2019, Available at revman.cochrane.org). I2 statistic estimated statistical heterogeneity, where I2 > 50% may be indicative of substantial heterogeneity. Meta-analysis was conducted only on trials from the integrated IPV and substance use interventions group due to insufficient data for other intervention types.
Findings
Study selection
The review included 13 manuscripts from 9 trials (n = 1014 men) (Figure 5).108–119
FIGURE 5 The PRISMA diagram for the systematic review. (Figure not reproduced here; see the original article.)
Intervention characteristics
Six interventions were delivered to men individually in person110,111,113–115,117 and one by phone.112 One intervention was delivered in a group,108,109 and one provided both group and individual sessions.119
Intimate partner violence outcomes
Data from three trials showed a reduction in IPV perpetration at different time points for interventions compared with TAU.113,115,116,119
Data were examined for IPV outcomes for the four trials of integrated interventions with appropriate or combinable outcome data.108–112 Data from the Revised Conflict Tactics Scale (CTS-2)120 physical violence subscale were combined at 4–12 weeks’ time points using mean and SDs to produce a mean difference for each (Figure 6). The combined mean difference was 0.1 [confidence interval (CI) −0.37 to 0.57, p = 0.68; Figure 6]. One trial113 used the Index of Spousal Abuse (Figure 7).121 IPV perpetration at both 1- and 3-month follow-ups was significantly lower in intervention versus control group (p = 0.004, p = 0.005, respectively), but difference at 12 weeks was non-significant.
FIGURE 6 Intimate partner violence analysis: CTS-2. (Figure not reproduced here; see the original article.)
FIGURE 7 Intimate partner violence analysis: Index of Spousal Abuse. (Figure not reproduced here; see the original article.)
Discussion
This small number of individual trials demonstrated short-term reductions in IPV perpetration. However, meta-analysis of four trials showed no statistically significant differences between integrated interventions and substance use TAU groups.
Limitations
A small number of heterogeneous trials were identified and included in meta-analysis. It was not possible to examine the effectiveness of individual components of interventions.
Implications for future trials and intervention development
Outcome measures should assess all forms of abuse (including coercive control)29 and should also be administered to (ex)-partners to strengthen overall validity via corroborated evidence around abuse.122 The integration of support services to safeguard victims was paramount.122
While research suggests that cognitive–behavioural therapy (CBT) and motivational interviewing (MI) are suitable for this population, optimum intervention duration requires further research, more intensive MI interventions may better develop participant engagement and motivation for change.99,123–125 Treatment adherence and outcomes were significantly poorer for substance-using men who were referred to stand-alone IPV interventions.100,101,126 Therefore, developing a theoretically integrated substance use and IPV intervention tailored to the complex needs of substance-using men delivered within substance use treatment would be prudent and more efficient.101