Comments on Stuart et al.’s “Genetic Associations With Intimate Partner Violence in a Sample of Hazardous Drinking Men in Batterer Intervention Programs”

Typical Batterers

The authors make a compelling and well-documented argument that the bio in biopsychosocial has been largely ignored in intervention with IPV perpetrators, suggesting that such inattention may ultimately harm victims of abuse. They do not present any practical applications in their discussion of the results, save for a throwaway comment on the possibility of meds and genetically informed behavioral therapy. Given the relative absence of research in genetic, neurological, and metabolic associations with IPV, we are a long way from being able to discuss possible applications of such research to people who batter their intimates.

To what extent do the participants in this article’s study resemble the characteristics of IPV perpetrators in the general population and in the population of men in BIP? My speculation is that the correlation between a cumulative genetic score and IPV for a sample of IPV perpetrators drawn from the general population would be substantially less than in the current study. Generalizing to men in the population is not a concern of this research because these men are not the targets of gene-based or any other IPV interventions.

Men who are the target of such interventions, whether voluntary (i.e., partner-referred) or mandated, are more of a concern. As noted by the authors, the characteristics of the men in this sample are unusual due to the research context, and this limits the study’s generalizability. All of the men in the current sample were involved in an intimate relationship and all were using alcohol or drugs at a potentially harmful level. In both cases, this is approximately twice the rate we would expect in a typical BIP, assuming that by “intimate relationship” we mean cohabiting or married. A common observation in most BIPs is that a substantial proportion of participants are not partnered at the time they enter the program. It is also commonly reported that about half of the men in BIPs have a substance use disorder. In the Multisite Study of 840 men in BIPs, less than half were married, less than half were cohabiting, and 56% had MAST scores of 5 or more (Gondolf, 1999). Likewise, in our Chicago study of 899 men in BIPs, 36% were married or cohabitating and 32% had a CAGE score greater than 1 (Bennett, Stoops, Call, & Flett, 2007). While these samples are not designed to be representative of men in BIPs, they are far closer to the typical BIP than the men in this sample. What might this mean? Based on the literature reviewed by Stuart et al. and ignoring my lack of background in behavioral genetics, I speculate that relationally disengaged men such as those in typical BIPs would have levels of genetic variation greater on average than the men in this study. This would be even more likely if we looked at generally violent men or men with criminal backgrounds. The average man in our Chicago study of IPV offenders, for example, had more than three arrests for non-IPV crimes.

Impact on the IPV Field

The greatest impact of this research, should its applications come to fruition as the authors imagine, would be on the victims of IPV. That impact is in the distant future, I fear, if it happens at all. The impact I am concerned about here is much more immediate, focused on three overlapping clusters of potential research consumers: those people who work with victims or perpetrators of abuse, those people who call for application to BIPs evidence drawn from controlled studies of interventions developed for other populations, and those people who dismiss feminist-influenced perspectives on IPV as a politically driven and willful aversion to science. I confess that I enter this discussion with some trepidation, in part because the authors make no mention of such reactions in their article. However, when I read this article, these were the consumer groups who came immediately to mind.

Genetics are another candidate in an emergent bio-psycho explanation that adds to (and challenges) the historically dominant social explanation of IPV. The ubiquitous power and control explanations and applications have been labeled inadequate by those who argue for more research-driven interventions. Proponents of this dominant social explanation, derived from feminism, are even sometimes referred to in print as data impervious. Critics draw a cartoon version of these allegedly data-impervious types as rigidly adhering to the idea that feminist constructs best explain IPV, that women are non-violent, and that those women who are violent do so in self-defense. Personally, I have not heard or read anything from these known feminists in the last few decades that would suggest that they think patriarchy alone explains IPV or that women are non-violent. Their critics, however, must have heard plenty of this, because otherwise they wouldn’t say it. The fact that some of these cartoon feminists often read the very journal in which this genetic research on batterers is appearing makes it all the more enticing. Stuart and his co-authors present no such science versus politics condescension, however. Their treatment of the sometimes at-odds perspectives on IPV is respectful and they acknowledge in their very first sentence that the causes of IPV are many.

When I look at Stuart et al.’s article through my advocate/practitioner eyes, my concerns immediately go to how the research will be viewed by funders, policy makers, researchers, and other authorities far removed from the daily work in IPV. While researchers are not responsible for how funders apply research to their funding decisions, neither can researchers ignore the effect of these decisions. The authors clearly state their aversion to such rash applications, but such applications are well beyond their control. In my mind’s eye, I vividly recall presentations and meetings with United Way and other funding panels deciding how to allocate dollars for local IPV services. In these meetings, IPV agency representatives were often asked if they did neuropsychological testing of batterers at their agency, as it seemed obvious to most panel members that there was clearly something wrong with the brains of these IPV perpetrators. This was about the time that early correlational studies were demonstrating elevated levels of head injury in people who batter (e.g., Rosenbaum et al., 1994), studies which were apparently getting some popular press. Despite my eye-rolling at the time, it now seems that those United Way panels were just hoping that the BIP was doing all it could do to look for, locate, and flip that switch, by which I mean intervene with great authority and in a very short time frame to generate an observable and meaningful reduction in IPV. That is what we all want, is it not?

Fast forward a few decades and we find BIPs and victim service agencies still struggling for understanding, even though there has been considerable evolution toward professionalization (for better and for worse), recognition of the role of trauma in the lives of victims and batterers alike, expansion of services for same-sex victims and batterers (at least in urban areas and college towns), and development of victim-informed interventions for women arrested for IPV. For advocates, counselors, and supporters, it is widely believed that the women and children they see on a daily basis are poorer, suffer greater cumulative trauma, and are more likely to be mentally ill and addicted than those seen in shelters 30 years ago.

Some advocates will be encouraged to hear that genetics may someday add to our ability to reduce aggression, particularly in some of the more virulent IPV perpetrators, such as those described as generally violent or antisocial. On the other hand, they will also understand that greater emphasis on genes and biology will continue our trend toward the medicalization of all human problems, with an accompanying shift in dollars away from community services and into the coffers of the private sector. Some will regard this article as another step in that direction. For some advocates, it will be difficult to appreciate how something as remote as genes can influence the domestic assault whose results they see in their shelters and walk-in programs, particularly compared with the observable anti-woman displays on television and other media. Other advocates will welcome this genetic research in the abstract and turn their attention back to the crisis de jour.

An enhanced biological understanding of IPV may also impact professionals, academics, and government people who enthusiastically endorse adoption of evidence-informed practices drawn from research in the behavioral sciences to IPV in general and to BIPs in particular. For the past two decades batterer intervention programs have had quite a run in the withering spotlight of data-based practice, and the news has not been good. With batterer program completion rates and re-offense rates both hovering around 50%, it has been increasingly difficult to make a case for continuing the standard model of court-ordered pro-feminist group-based cognitive-behavioral and psycho-educational intervention. But like advocates, this evidence-informed group comes in colors, ranging from “We need some data to support what we are doing” to flat-out methodolatory—worship of multiple randomized clinical trials as the only legitimate knowledge for practice. Most of us fall somewhere on this spectrum. Proponents of standard model BIPs are aware of the tepid results of the few BIP clinical trials, but maintain that batterer programs work for most of the people in the program, or at least those with some stake in conformity, particularly older men with jobs and relationships they want to maintain. For some die-hards, the clinical trials have been seriously flawed and the jury is still out. The development of genetic interventions in the future may ratchet up the success rate if it can improve the functioning of the minority of IPV offenders who commit the majority of the IPV re-offenses.

As for my fourth group, I hesitate to stir the pot with my seat-of-the-pants description and I admit to drawing a bit of a cartoon myself. This group seems to regard much of current IPV theory and practice as the droppings of feminist sympathizers and gender symmetry deniers. These consumers of genetic research on IPV perpetrators would likely herald the work of Stuart and his colleagues as yet another nail in the coffin of feminism as a viable solo theory of IPV. This relatively small group of academics is peopled with well-respected scientists who range from all-out anti-feminists to people who understand that patriarchy is not the stand-alone explanation that some are alleged to believe. I hedge my description here because I count myself as one who does not acknowledge feminism/patriarchy as a stand-alone explanation of what we observe, nor as the sole theoretical source of interventions for men who are IPV perpetrators. But as I said earlier, I have never actually heard or seen any of these known feminists say that the artifacts of patriarchy are the only explanation for what we observe. These scholars see little value in standard model batterer programs (code name “Duluth,” regardless of actual program practices) because they are infused with pro-feminist ideas. Genetic research will be fast-tracked onto their web sites and policy briefs as soon as it is published, but without Stuart-esque warnings about early stages of research, limitations on causality, and pre-mature applicability. A hard-wired switch for attachment-related IPV—were one ever established—would be a soothing balm for these colleagues.

Improving Batterer Intervention Programs

The authors propose that genes do not act directly on IPV behavior, but may act instead on sets of interrelated factors which could increase the risk of adult IPV, such as witnessing violence as a child, attachment, and various other personality characteristics, which in turn increase the risk for more immediate risk factors such as substance abuse or depression. It may be that genes affect underlying pathways shared by IPV and substance abuse, for example, leading to disorders with biological underpinnings and contributing to factors antecedent to both problems such as impulsivity, poor decision-making, and the inability to assess consequences (Fishbein, 2000). A pyramid of behavioral sequels could be set in motion by an individual’s genetic makeup, perhaps even including social class, a variable ignored in most IPV research and practice circles alike. However, given this seemingly distant effect of genes, as well as the apparently small effect size, it remains to be seen how applications of such research would eventually play out in interventions with IPV perpetrators. It is not clear to me how interventions based on genetically influenced risk factors would improve interventions that target the risk factors themselves, as we do now.

Given the weak performance of BIPs in a handful of clinical trials and two meta-analyses, it is not surprising that practices that have been found effective in other areas are sometimes offered as evidence-based alternatives to standard model BIPs, despite the fact that they have not been tested with the court-ordered, partner disengaged, generally violent and substance-abusing offenders that often populate perpetrator programs not associated with university research programs and subject to exclusion criteria. Couples counseling, dialectical behavior therapy, addiction treatment, and dynamic treatment based on attachment theory are a few of the alternative approaches that may be applied to work with IPV perpetrators. Despite their status as effective treatments for other issues, these alternatives have not yet been exposed to rigorous clinical trials with batterers. Couples counseling in particular usually excludes participants with the characteristics of court-ordered batterers.

One of the primary criticisms of the standard approach to BIPs is that of the seemingly one-size-fits-all approach of these programs. The authors suggest that genetic research “. . . may provide us with . . . the ability to match individuals to the treatment that may be most effective for their individual needs.” The IPV genetic advances of the future would add to the idea that IPV perpetrators are a heterogeneous group and can be classified into various typologies along different dimensions, not just the antisocial-type alluded to above. The senior author of the article has made a number of contributions in this area. In principle, differentiating batterers would not be an academic exercise, but would have applications to practice. I am not aware that such applications exist which have been evaluated. In the Multisite Study of batterers and their partners, about three quarters of the re-offenses were committed by about one quarter of the offenders (Gondolf, 2002), suggesting that there is a sub-group of men who are not being helped by the standard model programs. While the standard BIP approach may be applicable to many IPV perpetrators, what we can do with the group of offenders who are unreachable with the standard model remains an open question, an opening in which behavioral genetics may add another important new dimension. Other approaches suggested for batterers, such as therapy tailored to attachment issues and trauma-informed care, may yet demonstrate superiority to the standard model (although they have not yet done so) and provide effective treatment for this most difficult group. This, to me, is the promise of genetic research. I do not think the products of genetic research will be very useful with so-called family-only or normal batterers, but may have applicability to IPV offenders who have not been successfully treated by current practices.

The Stuart et al. article advances an important focus of research on IPV and BIPs, but possible applications to current and future practice with IPV perpetrators are unknown. The immediate impact of this article, and others like it to come, is on the researchers, practitioners, and policy people whose standpoint on IPV and BIPs will be challenged or supported by the results.