Standardized Effect Sizes for Preventive Mobile Health Interventions in Micro-randomized Trials.

Area of Science:

• Digital health interventions
• Behavioral science
• Clinical trial methodology

Background:

• Mobile Health (mHealth) interventions leverage mobile devices for behavioral support.
• mHealth interventions often comprise 'pull' and 'push' components.
• Push components, delivered via notifications, require specific evaluation methods.

Purpose of the Study:

• To review the micro-randomized trial (MRT) design for evaluating mHealth interventions.
• To propose a method for computing a standardized effect size for push intervention components.
• To enable comparison of different push components and inform sample size calculations for future MRTs.

Main Methods:

• Review of micro-randomized trial (MRT) design principles.
• Development of a time-dependent standardized effect size calculation for push notifications.
• Application of the methodology to the HeartSteps mHealth intervention data.

Main Results:

• A standardized effect size can quantify the impact of time-varying push notifications.
• This effect size facilitates comparison across different push intervention strategies.
• The approach aids in determining appropriate sample sizes for subsequent MRTs.

Conclusions:

• Micro-randomized trials provide valuable data for understanding mHealth push components.
• A standardized, time-dependent effect size is crucial for evaluating these components.
• This methodology enhances the design and analysis of future mHealth intervention studies.