Statistics from Altmetric.com
Medline, Cochrane Controlled Clinical Trials, PsycINFO, HealthSTAR, and CINAHL (all from 1994 to February 2003); Cochrane Database of Systematic Reviews and Database of Research Effectiveness (Issues 2 and 3, 2001; Issue 1, 2002); bibliographies of systematic reviews; the 1996 US Preventive Services Task Force recommendations; and experts.
Study selection and assessment:
English language randomised controlled trials (RCTs), non-randomised controlled trials, or systematic reviews of participants ⩾12 years of age with risky or harmful alcohol use who received a brief primary care behavioural counselling intervention primarily to reduce alcohol intake. Exclusion criteria included specialty addiction treatment settings, behavioural health department settings, and comorbid patient populations. Study quality was assessed (eg, randomisation, attrition rates, and replacement of missing values in outcome analyses).
alcohol consumption and health outcomes assessed at the endpoint nearest to 12 months of follow up.
12 RCTs of non-pregnant adults (about 33% women) met the selection criteria. All trials were done in multiple primary care practices (3–47 practices/study), and most included >300 participants. Length of follow up for drinking outcomes was ⩾12 months for all trials except 1 trial with 6 month follow up and 1 trial with ⩾9 month follow up. Interventions were examined by level of intensity: brief multicontact interventions (initial session ⩽15 min plus ⩾1 follow up contact); very brief interventions (1 session ⩽5 min); or brief interventions (1 session ⩽15 min). Use of 5 key intervention components (feedback, advice, goal setting, further assistance, and follow up) was also considered. The 12 trials included 15 intervention groups: 7 brief multicontact interventions, 2 very brief interventions, and 6 brief interventions. 12 interventions were delivered all, or in part, by the patient’s usual primary care physician.
Of the 7 trials of brief multicontact behavioural counselling interventions, 4 good quality trials reported net reductions of 13–34% in the mean number of drinks/wk in intervention groups, which resulted in 2.9–8.7 fewer mean drinks/wk compared with control groups. 1 fair quality trial reported reduced mean daily alcohol consumption. 5 good quality trials reported that 10–19% more participants in intervention than control groups were drinking at recommended or safe levels. 2 good quality trials reported reduced binge drinking among intervention participants.
Of the 8 trials of very brief or brief interventions, 3 reported improved alcohol outcomes for the intervention groups. All interventions that led to improvements in alcohol outcomes included ⩾2 of 3 key elements: feedback, advice, and goal setting. 6 studies reported morbidity related outcomes and found no differences or similar improvements in intervention and control groups.
In adults with risky or harmful alcohol use, brief multicontact behavioural counselling interventions in primary care settings reduce the number of drinks consumed per week and increase the proportion of participants drinking at moderate or safe levels.
Alcohol consumption is an interesting paradox for clinicians. Some evidence exists that light to moderate consumption in adults is associated with health benefits,1 whereas risky and harmful consumption can lead to serious health problems. Moving from moderate consumption to alcohol misuse (risky and harmful) is a process in which a person either increases his or her consumption or engages in binging behaviours.
This comprehensive, rigorous, and clearly defined review by Whitlock et al shows that time efficient screening and brief interventions using a single, 5–15 minute counselling session followed up by one or two 5 minute follow up calls or meetings are effective for reducing alcohol use to moderate levels. 4 of 7 studies also reported reductions in binge drinking. Effective counselling included elements of feedback, advice, and goal setting. However, it was difficult to separate the session content from the frequency and duration.
Given that harmful and risky drinking has a prevalence of up to 29%,2 it is surprising that routine screening and feasible brief interventions are not more common in primary care. These basic practices deserve the attention of primary care practitioners, and preventing adolescents and adults from developing more serious drinking patterns should be a priority for all primary care clinicians. Insufficient clinical time may no longer prevent the everyday use of these brief interventions. However, future efficacy studies involving longer follow up periods and long term morbidity outcomes are still warranted.
For correspondence: Dr E P Whitlock, Oregon Evidence-based Practice Center, Kaiser Permanente Center for Health Research, and Oregon Health & Science University, Portland, OR, USA.
Source of funding: Agency for Healthcare Research and Quality.