Personalizing self-management via behavioral predictive analytics with health education for improved self-efficacy

Abstract

SummaryThe objective of this research is to investigate the feasibility of applying behavioral predictive analytics to optimize diabetes self-management. This research also presents a use case on the application of the anaytics technology platform to deliver an online diabetes prevention program developed by the CDC. The goal of personalized self-management is to affect individuals on behavior change toward actionable health activities on glucose self-monitoring, diet management, and exercise. In conjunction with personalizing self-management, the content of the CDC diabetes prevention program was delivered online directly to a mobile device. The proposed behavioral predictive analytics relies on manifold clustering to identify subpopulations by behavior readiness characteristics exhibiting non-linear properties. Utilizing behavior readiness data of 148 subjects, subpopulations are created using manifold clustering to target personalized actionable health activities. This paper reports the preliminary result of personalizing self-management for 22 subjects under different scenarios and the outcome on improving diabetes self-efficacy of 34 subjects.