Social support and contingency management among alcohol-impaired drivers: A secondary analysis of a randomized clinical trial.

OBJECTIVE

This study examines whether perceived social support moderated contingency management (CM) outcomes among alcohol-impaired drivers.

METHOD: One hundred forty-five non-mandated adults arrested for driving while intoxicated in Bexar County, Texas, were randomized to 8-week CM (n = 74) or yoked Control (n = 71). Participants (mean age = 37.16 years, SD = 10.61) were predominantly male (77.2%) and Hispanic or Latino (71.0%). Weekly contingency compliance required clinic attendance and transdermal alcohol concentrations below 0.02 g/dL throughout the preceding week. Group-based multi-trajectory modeling jointly modeled three Multidimensional Scale of Perceived Social Support subscales assessed at baseline and Weeks 4 and 8. Mixed-effects logistic regression tested treatment-by-trajectory-by-week interactions, adjusting for sex, age, and baseline drinking.

RESULTS

Three groups emerged: low-level (25.5%), moderate-stable (61.4%), and high-stable (13.1%). The CM × high-stable support × week interaction yielded OR = 1.53 (p = .09). Observed compliance was higher in CM than Control at Weeks 2-5 and 8 in the low-level group and all eight weeks in the moderate-stable group. Low-level compliance did not change significantly in either condition. Moderate-stable compliance declined in both conditions, significantly only in Control (p = .04), with largely parallel trends. High-stable compliance increased in CM (p = .003), more than in Control (p = .03); Week 8 estimated probabilities in the high-stable group were 0.87 versus 0.33 (p = .01).

CONCLUSIONS

Perceived social support may shape responses to CM among alcohol-impaired drivers. Findings highlight the potential value of CM approaches that align financial incentives with everyday social reinforcement.