|
Abstract:
|
Mobile devices can be used to provide treatment or support whenever needed and adapted to the context of the user. The treatment policies, also known as Just-In-Time Adaptive Interventions (JITAIs), are composed of decision rules that, at each decision time, map user's context (e.g. location, weather, current time, social activity, stress, and urges to smoke) to select an intervention component to be delivered via a mobile device. However, data science in the mobile health field has to catch up with the fast-paced development of mobile technology. In this poster, we will introduce an offline data analysis method based on experimental or observational data for estimating the average of a long-term positive health outcomes that would accrue should a given JITAI be followed. We provide inferential methods for constructing confidence intervals and comparing different JITAIs. The proposed method is illustrated by simulation and a real data application on HeartSteps Study, a mobile intervention study for physical activity.
|
