The Effectiveness of a Brief Intervention for Illicit Drugs in Primary Health Care Settings

Brief intervention (BI) can decrease alcohol consumption among some patients with nondependent unhealthy use. But does BI have efficacy for illicit drugs? The World Health Organization published the first randomized trial of drug BI among 731 primary-care patients identified with the >70-item Alcohol, Smoking and Substance Involvement Screening Test (ASSIST). Patients with current cannabis, cocaine, other stimulant, or opioid use were recruited from sexually-transmitted disease and walk-in clinics, a dental clinic, and community primary care sites in Australia, Brazil, the US, and India. Few patients had other drug use (e.g., hallucinogens), but most reported recent tobacco and alcohol use. Patients scoring in the moderate-risk range for drug use were randomly assigned to BI (up to 60 minutes; median, 8–20 minutes) or no intervention until after the study. Patients scoring in either the low- or high-risk range for drug use were excluded.

• Brief intervention was associated with a 3-point greater decrease (compared with no BI) in a substance use score with a maximum of 336 points.
• Cannabis and stimulant scores also decreased more for BI subjects (by about 2–3 points on scales with a maximum of 39 points); opioid scores did not.
• Substance use was not significantly impacted by BI in the US.

Comments by James Harrison, MHS, CADC
The overall results of this study show that participants in the BI group did change their substance-taking behaviors compared with the control group. However, the effect of BI may have been underestimated since the control participants also changed their behaviors. Also, negative results in the US suggest that cultural factors might be at play. Nonetheless, screening with the ASSIST tool coupled with BI could be valuable for reducing illicit drug use in primary health-care settings.

Comments by Richard Saitz, MD, MPH
These results suggest that BI for drugs may have some benefit in general health settings, although the clinical significance is unclear. The use of the same interviewers for the BI and follow-up assessments at most sites could introduce a bias favoring BI, and since the source of most patients was not what is usually considered primary care (meaning longitudinal, continuous comprehensive care), whether or not it will have important clinical benefits remains unknown. Nor has the clinical meaning of small differences in substance use scores been established. Screening and BI for illicit drug use is clearly more complex than it is for alcohol.