Selective Benefit from Brief Motivational Interviewing in DWI Recidivist Subgroups
Author(s): Moxley-Kelly, Ouimet, Marcantoni, Brown
Slidedeck Presentation Only (no paper submitted):
Abstract:
Drivers convicted of previous DWI offenses contribute disproportionately to alcohol and traffic-related morbidity; approximately 30% of first-time offenders are re-arrested within 5 years, and the risk of being involved in a fatal crash increases with each subsequent DWI infraction. Hence, intervention to curtail recidivism following a previous DWI conviction represents an essential facet of a selective prevention strategy. The DWI population, however, is highly heterogeneous and treatment refractory. Attempts to disentangle population heterogeneity have been limited to statistical or descriptive exercises, which have failed to identify the core mechanisms underpinning DWI for which interventions may be designed to disrupt. In previous correlational research (Brown et al., 2016), we identified two candidate phenotypal subgroups; 1) fearlessness, demarcated by features of a blunted stress response, callousness, sensation seeking, substance misuse, reward sensitivity, and DWI accompanied by other criminal behaviours (i.e., MIXED); and 2) severe alcohol misuse, disinhibition when sober, and sole engagement in DWI behaviour (i.e., pDWI). Due to their differential profiles and putative underlying mechanisms for their DWI behaviour, these subgroups could show different responsivity to distinct treatments, but this 'matching hypothesis' has not been explored previously. Using data from a previous randomized controlled trial of Brief Motivational Interviewing (BMI) (Brown et al. 2010) in DWI recidivists, we hypothesized that MIXED drivers would be more responsive to BMI than pDWI drivers. Utilizing secondary analysis, this project compared outcomes of MIXED and pDWI drivers from exposure to one 30-minute session of either BMI treatment or an information-advice control. Participants were assessed at intake, 6-months, and 12-months post-treatment for drinking (Timeline Followback) and at 5-years post-treatment for recidivism. Planned comparisons ANOVA tested changes in drinking over time, and survival analysis tested latency to recidivism. From 6- to 12-months, pDWI drivers receiving BMI significantly reduced percent risky drinking days compared to the pDWI drivers receiving the control treatment (p < .05). No evidence was found for different treatment responsivity in 5-year latency to recidivism. Irrespective of treatment, MIXED drivers were more likely to have subsequent DWI convictions than pDWI drivers (p < .05). Evidence for the interaction between treatment and driver phenotype in outcome, if replicated, could inform treatment matching. The long-term greater risk for recidivism in the MIXED typological profile suggests a need for more intensive remedial action for this group. These results suggest that the phenotypal schema investigated here are clinically significant, and can be used to adjust intervention selection and intensity for improved outcomes.