A Distributed Registry: Smartphone-Connected Database Driven by Patient-Reported Outcomes

Long-term follow-up in registries has been reported to range from 10% to 70%, depending on resources available at participating institutions. Often even a patient's vital status is difficult to ascertain. The purpose of this study is to test the feasibility of a patient being connected directly to a quality assurance database via their personal smartphone to evaluate the ability to passively determine vital status. This distributed registry will have patient-reported outcomes (PRO) data entered into the database directly from a patient's mobile phone for 12 months. Our aims are to evaluate long-term vital status from (1) email, (2) text message, and (4) passive data. Patients with cardiovascular disease participated in the distributed registry study by downloading our research application. A medical survey, surgical survey and the Short Form-12 quality of life survey was completed in the application during on-boarding and every 3 months for a 12-month period. Patients received notifications once a week to perform an at-home 6-minute walk test to evaluate their functional capacity. In addition, daily total steps walked, distance walked and max steps without stopping were collected passively. Vital status was assessed hierarchically via email, text messaging, and passive data acquisition. One hundred one patients were enrolled. Mean follow-up time was 9 months. Email responses were received from 51.9%, text message responses were received from 68% and passive data were received from 82.5% up to the week of analysis (Fig). Of the 17% from whom we did not receive passive data, one patient passed away, one patient had a stroke and stopped using their phone, and 15 patients likely signed out of their application or deleted the application. The one patient who expired in our study was accurately identified with no passive data starting soon after his hospital admission. PROs in the form of passive activity and at-home walk test results were received from all patients. Passive data evaluation took the least effort and provided the greatest vital status response of 82.5%. This finding is comparable to the best reported long-term follow-up percentages. Text messages provided the next highest level of response, and email the least. All deaths were successfully identified. Long-term vital status assessment in registries can be significantly enhanced by connecting patients directly to a database via their smartphone. Additional benefits of remote active (6-minute walk test/surveys) and passive PRO collection are also gained.