Longitudinal stability of reward and relief drinking phenotypes in community and treatment‐seeking individuals who engage in heavy drinking

Abstract

AbstractBackgroundPrecision medicine approaches aim to improve treatment outcomes by identifying which treatments work best for specific individual phenotypes. In the treatment of alcohol use disorder (AUD), precision medicine approaches have been proposed based on phenotypes characterized by individuals who drink primarily to enhance rewarding experiences (i.e., reward drinking) or those who drink primarily to relieve negative states (i.e., relief drinking). This study examined these phenotypes across treatment‐ and nontreatment‐seeking individuals and the stability of the phenotypes over time.MethodsWe used latent profile and latent transition analyses to identify and assess longitudinal stability (over 3 or 4 months) of reward and relief drinking subgroups within a nontreatment‐seeking community sample that engaged in hazardous drinking (n = 189) and two treatment‐seeking samples of individuals with AUD enrolled in two large clinical trials (n = 1726, n = 1383). We examined prospective associations with alcohol consumption and consequences at long‐term follow‐up (15 or 18 months).ResultsResults supported four subgroups: low reward/low relief, low reward/high relief, high reward/low relief, and high reward/high relief. The community sample contained more individuals classified within the high reward/low relief subgroup than treatment‐seeking samples. Subgroups were generally more stable over time in the community sample than in the treatment‐seeking samples. Alcohol consumption and consequences decreased over time for the treatment‐seeking samples, with consequences and drinking frequency decreasing for the community sample. Participants classified within the high reward/high relief and low reward/high relief groups reported the most consequences and consumption at long‐term follow‐up.ConclusionReward and relief drinking phenotypes can be identified within community and treatment‐seeking samples of individuals who drink heavily. The phenotypic subgroups appear to be stable over time in the absence of treatment, change somewhat during treatment, and provide utility in predicting alcohol consumption and consequences.