A meta-analysis of web-delivered tailored health behavior change interventions.

Abstract

Web-based tailored intervention programs show considerable promise in effecting health-promoting behaviors and improving health outcomes across a variety of medical conditions and patient populations. This meta-analysis compares the effects of tailored versus nontailored web-based interventions on health behaviors and explores the influence of key moderators on treatment outcomes. Forty experimental and quasi-experimental studies (N =20,180) met criteria for inclusion and were analyzed using meta-analytic procedures. The findings indicated that web-based tailored interventions effected significantly greater improvement in health outcomes as compared with control conditions both at posttesting, d =.139 (95% CI = .111, .166, p <.001, k =40) and at follow-up, d =.158 (95% CI = .124, .192, p <.001, k =21). The authors found no evidence of publication bias. These results provided further support for the differential benefits of tailored web-based interventions over nontailored approaches. Analysis of participant/descriptive, intervention, and methodological moderators shed some light on factors that may be important to the success of tailored interventions. Implications of these findings and directions for future research are discussed.