Association between alcoholic interventions and abstinence rates for alcohol use disorders: A meta-analysis

OBJECTIVE: The aim of present study is to quantitatively evaluate the association between different interventions and abstinence rates based on network meta-analysis. METHOD: Following PRISMA guidelines, randomized clinical trials that compared different alcoholic interventions for alcohol use disorders associated with abstinence rates in treatment sessions or/and follow-up sessions were recruited. Main data synthesis was performed by Bayesian random-effects network meta-analyses, and the relative ranking of interventions was estimated by cumulative probability P values (SUCRA). Funnel plot symmetry was used to detect publication bias. Moreover, pair-wised comparison was also conducted to determine the statistical difference and forest plots were generated to calculate the differences between the groups. The Grades of Recommendations Assessment, Development and Evaluation (GRADE) criteria were utilized for the recommendations of evidence from pairwise direct comparisons. RESULTS: A total of 137 RCTs containing 27,282 participants and 8 variations of psychotherapy, pharmacotherapy, contingency management, and brief intervention used as treatment interventions were included. In summary, contingency management plus psychotherapy was demonstrated to be effective and possessed the best rank of achieving the highest abstinence rate in treatment sessions (SUCRA, 0.61). Pharmacotherapy plus psychotherapy also revealed its efficacy and was associated with the highest abstinence rate in follow-up sessions (SUCRA, 0.40). More importantly, psychotherapy alone was demonstrated not to be associated with higher abstinence rates in both treatment (OR, 1.052; 95% CI, 0.907–1.220) and follow-up sessions (OR, 0.967; 95% CI, 0.552–1.693), yet pharmacotherapy seemed to be the only intervention associated with higher abstinence rates compared to controls in both sessions (treatment session: OR, 1.074; 95% CI, 1.002–1.152) (follow-up session: OR, 1.442; 95% CI, 1.094–1.900). CONCLUSIONS: Contingency management plus psychotherapy and pharmacotherapy plus psychotherapy were demonstrated to be associated with the highest abstinence rates in treatment sessions and follow-up sessions, respectively. However, contingency management and pharmacotherapy seemed to be the substantial crucial factors allowing for the maintenance of the highest abstinence rates in respective sessions, although we need more evidence for further validation.