Large single and multi-site studies of drug brief intervention (BI) have found no efficacy. This trial enrolled 780 emergency department patients with past 90-day drug use (either weekly or more, or less with a consequence, identified by screening) who were randomly assigned to a tablet computer or computer-guided therapist BI (and 4-page change plan summary), or to usual care (3-minute review of an HIV prevention brochure and resource lists). All were subsequently randomized to a brief informational session versus a 40-minute motivational counseling booster. Mean age was 31; 45% were male, 52% black, 74% unemployed, 91% used marijuana, 19% used another illegal drug; it is not clear how many met criteria for a DSM substance use disorder. Follow-up was 85% at 6 months and 87% at 12.
- Primary outcome analyses (adjusted) found a significant effect size of 0.24 (95% confidence interval -0.41 to -0.07) for the therapist group compared with usual care at 6 and 12 months (difference in mean self-reported past 90-day drug use days between groups divided by standard deviation). There was no significant effect on drug use days in the computer-only group, or for either group at 3 months.
- Of note, 84–87% of participants provided urine samples for urine marijuana testing that was largely concordant with self-reported marijuana use.
- In unadjusted analyses, the 12-month decrease in drug use was approximately 6% greater in the therapist with booster group (27-28% [an approximately 11-14 day decrease to 31-37 days/90]) compared with usual care (20.9% [a 12 day decrease to 44 days/90]).
Comments: In this well-implemented trial, there was a small difference (roughly 0-2/90) in drug use days in a sample of people primarily using marijuana favoring the therapist intervention, a small-to-moderate 0.2 effect size statistically. With that, one wonders if loss to follow-up or self-report among those without urine tests could have flipped the main results to null. There may be reasons explaining positive effects—such as a computer and in-person combined intervention, the other intervention components, or the severity of drug use. But regardless, in the context of other null studies, questions will remain regarding the reproducibility and value of such an intervention to achieve a several-day greater decrease in drug use days.
Richard Saitz, MD, MPH
Reference: Blow FC, Walton MA, Bohnert AS. A randomized controlled trial of brief interventions to reduce drug use among adults in a low-income urban emergency department: the HealthiER You study. Addiction. 2017;112(8):1395-1405.