Baumel A, Yom-Tov E. (2018). Predicting user adherence to behavioral eHealth interventions in the real world: Examining which aspects of intervention design matter most. Translational Behavioral Medicine. doi: 10.1093/tbm/ibx037
Researchers examined relationships between consumer intervention use and intervention quality ratings for a sample of 30 publicly available web-based behavioral health interventions. Intervention quality was evaluated by 2 trained raters on 6 domains: usability, visual design, user engagement, content, therapeutic persuasiveness (i.e. promotion of behavior change), and therapeutic alliance (i.e. intervention efforts to build rapport with users). Intervention use data was extracted from anonymized web activity data from consenting users of a web browser toolbar distributed by Microsoft. Researchers extracted activity on intervention-related URLs and recorded number of days visited and cumulative duration of use for visitors over a 1 year period. Additionally, researchers identified percentage of users with 3 usage profiles combining days used and duration of use: 1) at least 3 days and at least 1.5 hours, 2) at least 5 days and at least 3 hours, and 3) at least 7 days and at least 5 hours. Therapeutic persuasiveness was significantly positively correlated with all 5 usage measures. User engagement was significantly positively related to number of days used and percentage of users using an intervention on at least 5 days and at least 3 hours total. Usability was significantly negatively related to total time used. Researchers note that high usability can also indicate a program has few features to navigate, which may explain the negative relationship between usability and intervention use. Results highlight the importance of intervention therapeutic persuasiveness, but future research should further examine how different domains of intervention quality can influence metrics of intervention success.