In a recent study published in JAMA, researchers assessed the efficacy and compared the relative effectiveness of various pharmacological treatments for alcohol use disorder (AUD).
Background
Unhealthy alcohol consumption is the third leading cause of preventable deaths in the United States (U.S.), with 145,000 fatalities yearly. Over 28 million Americans aged 12 and up fit the DSM-5 criteria for AUD in 2020, with potentially increased rates due to the coronavirus disease-2019 (COVID-19) pandemic.
Further research is needed. Despite the high prevalence and mortality associated with AUD, only a small fraction of affected individuals receive pharmacotherapy, underscoring a significant treatment gap that may have widened during the COVID-19 pandemic.
About the study
The research protocol was registered with the International Prospective Register of Systematic Reviews (PROSPERO) and followed a strategy to review the efficacy of pharmacotherapies for alcohol use disorder, as detailed in a comprehensive technical report.
Searches were conducted across several databases, including PubMed, the Cochrane Library, and others, by a librarian, with a peer review by a second librarian using the Peer Review of Electronic Search Strategies (PRESS) checklist. Inclusion criteria focused on adults with AUD in studies assessing Food and Drug Administration (FDA) approved medications or certain off-label medications for a minimum duration of 12 weeks.
Eligible studies for this systematic review and meta-analysis were required to report on alcohol consumption, health outcomes, or adverse events. For efficacy, only double-blind, randomized clinical trials were considered, whereas, for adverse effects, a broader range of study designs was permitted due to the limitations of randomized controlled trials (RCTs) in detecting rare harms. Two reviewers independently screened studies, with any conflicts resolved through discussion or by a third reviewer.
Data extraction was thorough, and the risk of bias was rigorously assessed using predefined criteria, considering the consistency, directness, and precision of the studies, with the strength of evidence being graded from high to insufficient. For substances with at least low evidence of benefit, results were synthesized, primarily focusing on alcohol consumption as the primary outcome.
Meta-analyses were performed using random-effects models, and various statistical tools were employed to calculate differences and ratios, as well as to evaluate heterogeneity. Subgroup analyses explored potential variables affecting outcomes. When meta-analysis was not feasible, a qualitative synthesis was provided instead.
Study results
A thorough database search yielded 2,860 citations, and through a precise initial review process focusing on titles and abstracts, the majority of these, 2,543 citations, were discarded. Further examination of 317 full-text articles led to the exclusion of 267 articles, culminating in a focused selection of 156 articles that reported on findings from 118 RCTs. Out of this group, 81 RCTs had been included in a comprehensive 2014 systematic review, and the other 37 RCTs were new contributions to the field.
These recently added RCTs had a wide range of participant numbers, from as few as 12 to as many as 921, and the treatment durations varied from 12 weeks to a full year. Most studies concentrated on individuals diagnosed with alcohol dependence. The recruitment methods employed were diverse, mirroring the extensive interest and the various investigative paths in this domain.
Most trials merged psychosocial interventions, noting that medications paired with psychological support seemed to offer added advantages, which was reflected in the effect sizes, thereby suggesting an amplified efficacy when coupling medical and psychosocial treatments in tackling alcohol dependence.
Research on acamprosate and naltrexone was mainly carried out in Europe and the U.S., often alongside psychosocial support. Medications like varenicline, ondansetron, and prazosin showed limited evidence of effectiveness. Acamprosate and naltrexone, approved by the FDA for alcohol use disorder, significantly improved drinking outcomes, but acamprosate did not notably reduce heavy drinking relapse. Oral naltrexone proved beneficial, contrasting with its injectable version.
Further analyses revealed that other non-FDA-indicated medications like topiramate and baclofen demonstrated significant benefits in reducing drinking rates. The strength of evidence for these findings was moderate to low, whereas gabapentin did not show a significant association with lower drinking rates. Direct comparisons between acamprosate and naltrexone showed no significant differences in improving alcohol use outcomes. Health outcomes related to medication treatment were not adequately reported in the RCTs, making it difficult to assess any substantial improvements in areas like quality of life or mortality.
Adverse effects across the medications were not consistently captured due to non-standardized collection methods, which often were not reported in the trials. Nonetheless, it was observed that dizziness was the most commonly mild side effect across the medications. Acamprosate and naltrexone were more likely to cause gastrointestinal distress compared to placebo.
Other medications, like baclofen and topiramate, were linked to higher instances of drowsiness, numbness, and cognitive dysfunction. Direct comparisons between acamprosate and oral naltrexone highlighted a lower incidence of nausea with acamprosate.
