Heart Steps: Adaptive mHealth Intervention for Physical-Activity Maintenance

This project file refers to Aim 2 and Aim 3 of the National Heart, Lung, and Blood Institute (NHLBI) grant. Aim 1 is registered in clinical trials.gov under a separate project file number, NCT03225521.

Many of the risk factors for heart disease are behavioral, such as physical inactivity, smoking, and diets high in saturated and trans-fats. Prevention programs are effective at helping patients make the initial lifestyle changes needed to reduce their risks, but patients often struggle to maintain those changes.

In this study the investigators propose to develop a novel mHealth application for supporting maintenance of physical activity. Kaiser Permanente Washington Health Research Institute (KPWHRI) will conduct a prospective study to test the application among Kaiser Permanente Washington (KPW) men and women with blood pressure that falls in the stage 1 hypertension range. By taking advantage of the frequent interactions that individuals have with their mobile phones throughout the day, the investigators will design and evaluate an adaptive, personalized application that (1) keeps patients reminded of their health goals and motivations; (2) provides actionable ideas for how patients can be active right now, given their current context; and (3) helps patients plan and reflect on their physical activity to enable creation of robust and sustainable physical-activity habits. In addition, the application will adapt its functioning for each patient over time in order to minimize user burden while optimizing its ability to encourage physical activity and maintain engagement with the intervention.

Through three Specific Aims-intervention development, intervention optimization, and pilot evaluation-this project will contribute: (1) understanding of how burden and intervention engagement change over time and how they are influenced by intervention delivery and patients' behavior and context; (2) generalization of learning algorithms that are used to personalize web content based on users' characteristics and past behavior for use in mHealth interventions for behavioral maintenance; and (3) new insights about how reflective and automated self-regulatory processes interact over time to support behavioral maintenance and how to optimally recruit these processes through an mHealth intervention.

Aim 2: Optimize intervention and investigate influences on participant response.