Alcohol assessment and feedback by e-mail for university students: main findings from the AMADEUS-1 randomised controlled trial

Background

Brief interventions can be efficacious in changing alcohol consumption and related problems and increasingly take advantage of the internet to reach high risk populations such as students.

Aims

To evaluate the effectiveness of a brief online intervention, part of the national strategic response in Sweden, controlling for the possible effects of the research process.

Methods

A three arm parallel groups design permitted exploration of the magnitude of the feedback and assessment component effects via randomisation to fully automated: 1) routine practice assessment and feedback; 2) assessment only without feedback; or 3) no contact and thus neither assessment nor feedback. The study was undertaken simultaneously in two universities randomizing the e-mail addresses of all 14,910 students (4,969, 4969 and 4972 respectively to Groups 1-3) who were entirely blinded to trial participation. Outcomes were evaluated after 3 months via an invitation to participate in a brief cross-sectional lifestyle survey.

Results

Overall, 52% (n=7,809) of all students completed follow-up, with small differences in attrition between the three groups (2,546, 2,594 and 2,669 respectively in Groups 1-3). For each of the two primary outcomes, there was one statistically significant difference between groups, with Group 1 having 3.7% fewer risky drinkers at follow-up than Group 3 (P=0.006) and Group 2 scoring 0.16 points lower than Group 3 on the AUDIT-C (P=0.039).

Conclusions

This study provides some evidence of population-level benefit attained through intervening with individual students.

Trial registration

ISRCTN28328154