Development of a Smartphone App for Regional Care Coordination Among High-Risk, Low-Income Patients.

Despite widespread adoption of health information technology, U.S. providers face persistent barriers to coordination of care. We sought to develop and implement a patient-centered smartphone app that facilitates care coordination when patients receive care at any hospital in a region. Partnering with patients and primary care teams at a federally qualified health center (FQHC), we developed an app that (1) used real-time location data to identify encounters at 41 regional hospitals; (2) sent notifications to users' phones, asking them to confirm hospital arrival/discharge, and; (3) sent automated messages to primary care teams about confirmed hospital encounters. App design included multiple, successive rounds of active patient participation. In a small beta test of the initial version of the app, high-risk, low-income FQHC patients ran the app on their phone for 3 months. A formative mixed methods evaluation examined the app's technical performance and user experience. Twelve patients enrolled in the beta test and provided follow-up data; 11 (92%) were racial/ethnic minorities. Participants obtained emergency or inpatient care at four regional hospitals. The app had 75% sensitivity to detect events when notifications should have fired, and 90% positive predictive value (PPV) of events when notifications fired. Barriers to implementation related to the app's user interface and the performance of its location tracking algorithm. We partnered with patients from a traditionally underserved population to develop a new smartphone-based approach to regional care coordination. The app had moderate sensitivity and high PPV for identifying regional hospital visits.