A web-based peer-modeling intervention aimed at lifestyle changes in patients with coronary heart disease and chronic back pain: sequential controlled trial.

Background: Traditional secondary prevention programs often fail to produce sustainable behavioral changes in everyday life. Peer-modeling interventions and integration of peer experiences in health education are a promising way to improve long-term effects in behavior modification. However, effects of peer support modeling on behavioral change have not been evaluated yet. Therefore, we implemented and evaluated a website featuring patient narratives about successful lifestyle changes.

Objective: Our aim is to examine the effects of using Web-based patient narratives about successful lifestyle change on improvements in physical activity and eating behavior for patients with coronary heart disease and chronic back pain 3 months after participation in a rehabilitation program.

Methods: The lebensstil-aendern ("lifestyle-change") website is a nonrestricted, no-cost, German language website that provides more than 1000 video, audio, and text clips from interviews with people with coronary heart disease and chronic back pain. To test efficacy, we conducted a sequential controlled trial and recruited patients with coronary heart disease and chronic back pain from 7 inpatient rehabilitation centers in Germany. The intervention group attended a presentation on the website; the control group did not. Physical activity and eating behavior were assessed by questionnaire during the rehabilitation program and 12 weeks later. Analyses were conducted based on an intention-to-treat and an as-treated protocol.

Results: A total of 699 patients were enrolled and 571 cases were included in the analyses (control: n=313, intervention: n=258; female: 51.1%, 292/571; age: mean 53.2, SD 8.6 years; chronic back pain: 62.5%, 357/571). Website usage in the intervention group was 46.1% (119/258). In total, 141 trial participants used the website. Independent t tests based on the intention-to-treat protocol only demonstrated nonsignificant trends in behavioral change related to physical activity and eating behavior. Multivariate regression analyses confirmed belonging to the intervention group was an independent predictor of self-reported improvements in physical activity regularity (β=.09, P=.03) and using less fat for cooking (β=.09, P=.04). In independent t tests based on the as-treated protocol, website use was associated with higher self-reported improvements in integrating physical activity into daily routine (d=0.22, P=.02), in physical activity regularity (d=0.23, P=.02), and in using less fat for cooking (d=0.21, P=.03). Multivariate regression analyses revealed that using the website at least 3 times was the only factor associated with improved lifestyle behaviors.

Conclusions: Usage of the lebensstil-aendern website corresponds to more positive lifestyle changes. However, as-treated analyses do not allow for differentiating between causal effects and selection bias. Despite these limitations, the trial indicates that more than occasional website usage is necessary to reach dose-response efficacy. Therefore, future studies should concentrate on strategies to improve adherence to Web-based interventions and to encourage more frequent usage of these programs.