Importance: Unhealthy alcohol use is common and increasing in adults and is the most common cause of premature mortality in the United States.
Objective: To systematically review the benefits and harms of screening and nonpharmacologic interventions to reduce unhealthy alcohol use to inform the U.S. Preventive Services Task Force.
Data Sources: MEDLINE, PubMED, PsycINFO, and Cochrane Central Register of Controlled Trials through October 12, 2017; references of relevant publications; government Web sites; and ongoing surveillance through August 1, 2018.
Study Selection: English-language trials of benefits and harms of screening in health care settings or other comparable populations and nonpharmacologic interventions to reduce unhealthy alcohol use in screen-detected persons who report unhealthy alcohol use, and test accuracy studies of selected screening tools to detect unhealthy alcohol use.
Data Extraction and Synthesis: Two investigators independently reviewed abstracts and full-text articles, then extracted data from fair- and good-quality trials, based on predetermined criteria. Random-effects meta-analysis was used to estimate benefits of the interventions.
Main Outcomes and Measures: The primary drinking outcomes were drinks per week, exceeding recommended alcohol use limits, heavy use episodes, and, for pregnant women, abstinence. Other outcomes included mortality; quality of life and consequences of alcohol use; injuries, accidents, and acute health care utilization; family, social, and academic functioning; and legal outcomes.
Results: We included 113 studies (n=314,466) across all Key Questions. We did not find any studies that examined the benefits or harms of screening programs to reduce unhealthy alcohol use. For adolescents, data supported the use of the National Institute on Alcohol Abuse and Alcoholism Youth Screen and other similar one- or two-item screeners to detect alcohol use disorder. For adults, brief (1- to 3-item) screeners commonly reported sensitivity and specificity between 0.70 and 0.85, typically having better sensitivity than the full Alcohol Use Disorders Identification Test (AUDIT) for identifying the full spectrum of unhealthy use. However, the AUDIT tended to have higher specificity, particularly at the standard cutoff of 8 or higher. Evidence on the effects of interventions to reduce unhealthy alcohol use in adolescents was limited to two trials; both found mixed results for reduced alcohol use and did not report health or related outcomes. In adults, interventions reduced the number of drinks per week (weighted mean difference, −1.82 [95% confidence interval {CI}, −2.42 to −1.22]), the proportion exceeding recommended drinking limits (odds ratio [OR], 0.60 [95% CI, 0.53 to 0.67]), and the proportion reporting a heavy use episode (OR, 0.62 [95% CI, 0.55 to 0.71]), and increased the proportion of pregnant women reporting abstinence (OR, 1.92 [95% CI, 1.19 to 3.09]) after 6 to 12 months. Analyses limited to trials conducted in primary care settings and the United States suggested that effects in these most applicable trials were comparable or larger than the overall effect (e.g., for trials in primary care settings, the weighted mean difference was −2.82 [95% CI, −3.87 to −1.76]). Benefits remained through 24 months or beyond in four of seven trials with longer-term outcomes. Heterogeneity was high and effect size was associated with a number of study characteristics such as setting, target age of the population, publication year, study size, and average baseline-use levels, but not clearly associated with any intervention characteristics. Data on effectiveness in important subgroups were very limited, but analyses by sex, the most commonly reported subgroup analysis, did not indicate differences in effectiveness of the interventions. Health outcomes were sparsely reported and, with some exceptions, generally did not demonstrate group differences in effect. We found no evidence that these interventions could be harmful.
Conclusion: Among adults, screening instruments are available that can effectively identify persons with unhealthy alcohol use and that are feasible for use in primary care settings, and interventions in those who screen positive are associated with reductions in unhealthy alcohol use. There was no evidence that these interventions have unintended harmful effects. More evidence is needed to determine whether screening for unhealthy alcohol use is beneficial for adolescents.