Drinking problems remained a risk factor for reoffending even among men with stronger social skills

Alcohol-related problems and difficulties with social and cognitive skills were each associated with reoffending among men who had perpetrated intimate partner violence, but the two did not appear to combine in a way that further increased risk, according to a study published in Applied Cognitive Psychology. Some skills that would normally be expected to protect against reoffending were unexpectedly linked to higher odds of a new violent offense.

Problematic alcohol use is common among men who perpetrate intimate partner violence, and alcohol can impair judgment, weaken behavioral control and make people more prone to impulsive reactions. However, alcohol is not the sole cause of violence. Other factors, including hostility, impulsivity, poor problem-solving, difficulty understanding another person’s perspective and trouble controlling behavior, are also linked to violence against partners and to reoffending.

Researchers have suggested that alcohol might be especially dangerous for people who already have these difficulties. For example, alcohol could make it harder for someone with poor impulse control to pause before acting or could intensify hostile interpretations of a partner’s behavior, while stronger skills might help buffer against alcohol’s effects. Few studies had tested whether these factors actually interact to predict reoffending, and most had looked at only a handful of skills.

The study was led by Crystal J. Giesbrecht and Kaila C. Bruer of the University of Regina in Canada. The team analyzed existing assessment and follow-up data from 7,781 men in Alberta who had perpetrated intimate partner violence and had started community supervision between 2009 and 2012. The men ranged in age from 18 to 84, with an average age of about 35, and about 27% were Indigenous.

The data came from the Service Planning Instrument, a risk assessment that correctional staff complete when a person begins serving a sentence under supervision in the community. Staff score it using interviews, criminal records, previous assessments and information from people such as family members, partners and employers. The researchers then examined whether the men had any recidivism within three years, which included new charges as well as violations such as breaching release conditions or failing to appear in court, and whether they had a new violent offense, such as assault or uttering threats.

Alcohol use was measured in three ways: how often the men drank, whether drinking had disrupted their work, family life, friendships or health, and whether alcohol use contributed to their criminal behavior. Staff also rated eight social and cognitive skills: impulsivity, hostile attributions (the tendency to read neutral actions as hostile), consequential thinking (thinking ahead about the results of one’s actions), perspective-taking, problem-solving, behavioral control, interpersonal skills, and goal-setting and planning. The researchers classified each skill as either a risk area or not.

Alcohol use was widespread: 93% of the men used alcohol, and about a third drank at least three days a week. For about half the men, alcohol contributed to their criminal behavior, and for 27% it had disrupted their functioning. Within three years, 38% of the men had some form of recidivism and 20% had a new violent offense.

After accounting for the other factors in the analysis, men whose drinking had disrupted their functioning had about 27% higher odds of any recidivism and 28% higher odds of violent recidivism. Alcohol-related criminal behavior was linked to similar increases for both outcomes. Frequent drinking was associated with higher odds of any recidivism compared with not drinking, but not with violent recidivism.

Among the skill measures, men rated as having problems with impulsivity were more likely to have both types of recidivism, and those with weaker goal-setting and planning were more likely to have any recidivism. Hostile attributions were linked to higher odds of violent reoffending. Problem-solving, behavioral control and interpersonal skills were not reliably linked to either outcome once the other factors were taken into account.

Two results ran in the opposite direction from what the researchers expected. Men rated as having no problems with perspective-taking had higher odds of any and violent recidivism, and those with no problems in consequential thinking had higher odds of violent recidivism. The authors suggested that such skills, without adequate self-control or prosocial goals, might help some people manipulate others or navigate the justice system in self-serving ways, though the pattern could also reflect factors the data did not capture.

Alcohol problems and skill deficits tended to go together: men whose drinking disrupted their lives or contributed to their offending were more likely to be rated as having deficits in every skill area. But the researchers found little evidence that the two interacted to predict reoffending. In other words, alcohol problems and skill deficits each mattered, but having both did not produce an extra increase in risk beyond what each contributed on its own. The models explained only about 3% of the variation in any recidivism and about 2% for violent recidivism, which the authors note is typical of research predicting a behavior as complex as reoffending.

“This indicates that problematic alcohol use may represent a stand-alone criminogenic need that remains salient even among individuals with relatively strong cognitive or social skills,” the authors wrote. Criminogenic needs are risk factors for reoffending that can change over time, which makes them targets for treatment.

The researchers recommend screening for both alcohol problems and social and cognitive skills when people enter supervision, and building alcohol treatment into intimate partner violence programs. Such programs often exclude people who are actively using substances, and access to them in Canada is limited and inconsistent.

There are a few things to keep in mind. Because the dataset contained no victim information, the researchers could not tell whether new offenses were committed against an intimate partner, so the findings concern general and violent reoffending rather than repeat partner violence specifically. Canada’s Criminal Code also has no specific offense for intimate partner violence, which makes such cases hard to identify in official records.

The study relied on existing assessment data, so the researchers could not add measures or check for errors, and recidivism was based on official records, which miss offenses that never come to the attention of authorities. Because the data are observational, the findings show associations rather than evidence that alcohol or skill deficits cause reoffending. The authors also suggest that future research account for differences among men who perpetrate partner violence, since alcohol may be more closely tied to violence for some types of perpetrators than for others.

The study, “The Connection Between Alcohol Use and Social and Cognitive Skills for Men Who Have Perpetrated Intimate Partner Violence,” was authored by Crystal J. Giesbrecht and Kaila C. Bruer.