Optimizing Engagement in an mHealth Intervention for Diabetes Support During Pregnancy: the Role of Baseline Patient Health and Behavioral Characteristics.

Background Mobile health (mHealth) technology that addresses diabetes mellitus in pregnancy has the potential to improve maternal and child health while diminishing socioeconomic and racial disparities. Little is known about health literacy, electronic health literacy, or patient characteristics that contribute to increased mHealth use. In this pilot study, we aimed to examine patient factors associated with user engagement with a novel app for diabetes support during pregnancy. Method Low-income pregnant individuals with gestational or type 2 diabetes mellitus were recruited for a 2-week usability assessment of a novel mHealth smartphone application, "SweetMama," designed to improve diabetes care. This analysis of user characteristics was a pilot assessment of data from the usability study phase. Participants completed assessments of patient health literacy, electronic health literacy, diabetes self-efficacy, and activation. User metrics (eg, sessions and total duration of use) were evaluated by participant characteristics. Descriptive analyses were conducted to assess outcomes with respect to patient's clinical history, health literacy, electronic health literacy, diabetes self-efficacy, and activation. Results Twenty-two diverse, low-income pregnant patients with either gestational or type 2 diabetes engaged with SweetMama for 14 days. Participants with gestational diabetes had greater mean minutes of use per session and mean total duration of use than those with type 2 diabetes. Participants with greater electronic health literacy, lower patient activation, and greater self-efficacy demonstrated greater total duration of use. No user metrics differed significantly by participant general health literacy. Conclusions Findings of this pilot study suggest engagement with a novel mHealth app for diabetes support during pregnancy varied by baseline behavioral and clinical characteristics. Results could indicate that digital health literacy better differentiates engagement patterns than general health literacy.