BACKGROUND

Developing integrative screening strategies to improve early identification of alcohol use disorder (AUD) is critical. This study examines cardiovascular parameters and alcohol-related phenotypes associated with two groups of alcohol drinkers: high-risk and low-risk.

METHODS

Data from 520 high-risk and 586 low-risk non-smoking alcohol drinkers were analyzed. Generalized linear models analyzed the relationship between AUD-related outcomes (Alcohol Dependence Scale, number of DSM Alcohol Dependence criteria, Clinical Institute Withdrawal Assessment for Alcohol, Revised, and Penn Alcohol Craving Scale) and cardiovascular measures (mean arterial pressure, heart rate, QTcF, QRSD, and PR intervals, and QRS, P-wave, and T-wave axes). Multiple logistic regression examined associations of sociodemographic and cardiovascular variables with the odds of being a high-risk drinker. Statistically significant cardiovascular variables were retained as explanatory variables in Tweedie regression models for alcohol-related phenotypes. Interaction effects of risk group by cardiovascular measure were included in each model testing the association between cardiovascular parameters and alcohol-related phenotypes.

RESULTS

Higher mean arterial pressure (MAP) and heart rate (HR) were associated with increased odds of high-risk drinking, while greater P-wave axis was associated with increased odds of low-risk drinking. Compared with low-risk drinkers, alcohol outcomes for those who engaged in high-risk drinking were not only significantly greater but also significantly less dependent on MAP and HR variations. The P-wave axis was significantly associated with low-risk drinking; however, it showed no significant association with any other alcohol outcomes.

CONCLUSIONS

Findings from this study suggest that MAP, HR, and an abnormal P-wave axis can be useful signals for detecting increasing and potentially harmful alcohol drinking among patients who do not yet meet the threshold for high-risk drinking. Early, objective, and targeted identification can improve the current undertreatment of this population at risk by decreasing the interval between onset of AUD and initial clinical care and treatment.