In Summary

A new study involving Swinburne University of Technology has found that a web-based self-assessment and feedback program for university students who drink heavily had little impact on their alcohol consumption.

The research focused on 3,422 students in New Zealand who had been identified in a screening test as exhibiting patterns of hazardous or harmful drinking.

The students were than randomized, with half involved in the web intervention and half in a control group.

Five months later the students were asked to complete a web questionnaire about the frequency and volume of their drinking.

The intervention produced a small (7%) drop in the volume of alcohol consumed per typical drinking occasion but had no effect on the frequency of drinking, the total volume consumed or the frequency of alcohol-related academic problems.

Swinburne biostatistician Dr Steve Bowe developed the analysis plan and undertook the statistical analysis for the study.

“Although it is disappointing to find that the intervention was ineffective, we were able to show that a rigorous intervention study can be implemented on a national scale at relatively low cost,” he said.

“In the past 10-15 years there have over 50 trials of web-based interventions for alcohol problems in young people. Some smaller studies, including some we’ve conducted in Australia and in the indigenous university student population of New Zealand, have shown promising results,” Dr Bowe said.

The research indicates that, on their own, web-based alcohol screening and brief intervention cannot be relied upon to deal with unhealthy drinking by students. Instead, it says they should be used in conjunction with other measures such as restricting the availability and promotion of alcohol.

The study involved collaboration with the University of Newcastle, the University of Otago, Boston University, and the London School of Hygiene and Tropical Medicine.

It was published in the prestigious Journal of the American Medical Association, whoseeditors praised the design and scale of the study.