Translating effective web-based self-help for problem drinking into the real world.

Background: Drinking Less (DL) is a 24/7 free-access anonymous interactive web-based self-help intervention without therapeutic guidance for adult problem drinkers in the community. In a randomized controlled trial (referred to here as DL-RCT), DL has been shown effective in reducing risky alcohol consumption.

Objective: To assess whether the findings of DL-RCT are generalizable to a naturalistic setting (DL-RW) in terms of ability to reach the target group and alcohol treatment response.

Methods: Pretest-posttest study with 6-month follow-up. An online survey was conducted of 378 of the 1,625 people who used DL-RW from May to November 2007. Primary outcome measures were (1) problem drinking, defined as alcohol consumption in the previous 4 weeks averaging >21 or >14 standard units (male/female) per week or >or=6 or >or=4 units (m/f) on 1 or more days per week; and (2) mean weekly alcohol consumption. DL-RW and DL-RCT data were compared and pooled. Intention-to-treat (ITT) analysis was performed to analyze and compare changes in drinking from baseline to follow-up.

Results: In the DL-RW group, 18.8% (n = 71) were drinking successfully within the limits of the Dutch guideline for low-risk drinking (p < 0.001) 6 months after baseline (ITT). The DL-RW group also decreased its mean weekly alcohol intake by 7.4 units, t(377) = 6.67, p < 0.001, d = 0.29. Drinking reduction in DL-RW was of a similar magnitude to that in the DL-RCT condition in terms of drinking within the guideline [chi(2)(1) = 1.83, CI: 0.82-3.00, p = 0.18, RD = 0.05, OR = 1.55] and mean weekly consumption (a negligible difference of d = 0.03 in favor of DL-RW group).

Conclusion: The results from DL-RCT and DL-RW were similar, and they demonstrate that web-based self-help without therapeutic guidance is feasible, well accepted, and effective for curbing adult problem drinking in the community.