Effectiveness and cost‐effectiveness of face‐to‐face and electronic brief interventions versus screening alone to reduce alcohol consumption among high‐risk adolescents presenting to emergency departments: three‐arm pragmatic randomized trial (SIPS Junior high risk trial)

BACKGROUND AND AIMS: Alcohol use increases throughout adolescence. Emergency department (ED) attendance is an opportunity for alcohol screening and brief intervention (ASBI), which is effective for adults. This trial evaluated the effectiveness and cost‐effectiveness of ASBI compared with screening alone (SA) in high‐risk adolescents. DESIGN, SETTING AND PARTICIPANTS: Multi‐centre, three‐group, single‐blind, individually randomized trial with follow‐ups after 6 and 12 months in 10 ED settings in England. From October 2014 to May 2015 we screened 3327 adolescents aged 14 to 18 years, of whom 756 (22.7%) scored at least 3 on the Alcohol Use Disorders Identification Test: consumption (AUDIT‐C) and consented to participate in this trial. Mean age was 16.1 years; 50.2% were female and 84.9% were white. INTERVENTIONS: Interventions were personalized feedback and brief advice (PFBA), personalized feedback plus electronic brief intervention (eBI) and SA. MEASURES: The primary outcome was the weekly alcohol consumed in standard UK units (8 g ethanol) at 12 months post‐randomization, derived from extended AUDIT‐C. Economic outcomes included quality of life and service use, from perspectives of both the National Health Service and personal social services (NHS&PSS) and society. FINDINGS: At 12 months, mean weekly consumption was 2.99 [95% confidence interval (CI) = 2.38–3.70] standard units for the SA group, 3.56 (95% CI = 2.90, 4.32) for PFBA and 3.18 (95% CI = 2.50, 3.97) for eBI, showing no significant differences. The PFBA group consumed mean 0.57 (−0.36, 1.70) units more than SA; and eBIs consumed 0.19 (−0.71, 1.30) more. Bayes factors suggested lack of effectiveness explained non‐significance. From the NHS&PSS perspective, economic analysis showed that PFBA and eBI were not cost‐effective compared with SA: PFBA yielded incremental cost‐effectiveness ratio of £6213 (−£736 843, £812 884), with the intervention having 54% probability of being cost‐effective compared with SA at the £20 000 WTP threshold. CONCLUSIONS: In emergency departments in England, neither personalized feedback and brief advice nor personalized feedback plus electronic brief intervention showed evidence of being effective or cost‐effective when compared with screening alone in reducing alcohol consumption among adolescents.