Meta-Analysis: Behavioral Counseling Interventions for Nondependent Unhealthy Alcohol Use Decrease Drinking in Adult Primary-Care Patients
To help the US Preventive Services Task Force update its guidelines, researchers searched for English-language controlled trials published between 1985 and 2012 that evaluated behavioral counseling interventions for unhealthy alcohol use identified by screening in primary care settings. Twenty-three trials met eligibility criteria. Interventions were usually multicontact and included brief advice/feedback, motivational interviews, and cognitive strategies delivered in very brief, brief, or extended formats.
- Compared with controls, adults receiving behavioral counseling interventions:
- decreased their weekly alcohol consumption more (mean difference, 3.6 fewer drinks per week), were more likely to drink lower risk amounts (mean risk difference, 11%), and were less likely to report heavy drinking* episodes (mean risk difference, 12%) at 12 months. Brief (<15 minutes each) multicontact interventions had the best supporting evidence.
- had fewer hospital inpatient days (low strength of evidence) but no difference in emergency department (ED) visits, legal problems, mortality, or quality of life.
- Compared with controls, young adults/college students receiving behavioral counseling interventions decreased their consumption and had fewer motor vehicle crashes, ED visits, and academic consequences.
- Evidence was not sufficient to make conclusions about other groups (e.g., pregnant women, adolescents, older adults).
*Defined across studies as ≥5 drinks per occasion for men and ≥4 drinks for women.
Comments:This well-done systematic review supports behavioral counseling interventions for primary care patients who screen positive for nondependent unhealthy alcohol use. The main observed benefit was for intermediate outcomes such as alcohol consumption. However, response to initial or repeated interventions over time could lead to reductions in “hard” outcomes such as mortality, alcohol-related trauma, and liver disease. The results do not apply to patients with alcohol dependence, since most of the included trials excluded such patients. Kevin L. Kraemer, MD, MSc