Abstract
BACKGROUND: In a multisite clinical trial, varenicline was effective in reducing drinking in both smoking and non-smoking patients with alcohol dependence. Because alcohol use disorder (AUD) is heterogeneous, research aimed at characterizing features associated with treatment response could advance personalized pharmacotherapy. The current study examined the utility of a multidimensional assessment of AUD severity to identify responders to varenicline treatment. METHODS: The study utilized data from a 13-week, Phase 2, randomized, double-blind, placebo-controlled, multisite trial of varenicline in 200 alcohol-dependent patients. Four hypothesized measures of AUD severity (i.e., DSM-IV criterion count, withdrawal, craving, and alcohol-related consequences) were combined into a single severity factor. A series of multilevel models that included the severity factor were conducted to examine its effects on treatment outcomes. RESULTS: All hypothesized indices of AUD severity loaded significantly onto a singular severity factor. Among low-severity groups, varenicline treatment significantly reduced drinking (i.e., percent heavy drinking days, drinks per day, and drinks per drinking day) and, in the lowest severity group, it improved the mental health component of quality of life more than placebo. The most severe group showed no differences between the varenicline and placebo groups on drinking or quality of life measures. CONCLUSIONS: Treatment response to varenicline may be greater among individuals with less-severe AUD, as evidenced by both reductions in drinking outcomes and improvements in psychosocial functioning.