Adherence and Reliability of Mobile Health App-Based Surveys in the Electronic Framingham Heart Study

• Activity Number: 476 - Innovations in Analytic Approaches for Survey Data
• Type: Contributed
• Date/Time: Thursday, August 6, 2020 : 10:00 AM to 2:00 PM
• Sponsor: Survey Research Methods Section
• Author(s): Chathurangi H Pathiravasan* and Yuankai Zhang and Ludovic Trinquart and David McManus and Emelia J. Benjamin and Belinda Borrelli and Honghuang Lin and Amy Dunn and Eric Schramm and Chunyu Liu and Joanne M. Murabito
• Companies: Department of Biostatistics, School of Public Health, Boston University and Boston University School of Public Health and Boston University School of Public Health and University of Massachusetts and Department of Medicine, and Department of Epidemiology, Boston University and Department of Health Policy & Health Services Research, Boston University and Section of Computational Biomedicine, Department of Medicine, Boston University and School of Medicine, Framingham Heart Study and CareEvolution and Department of Biostatistics, Boston University School of Public Health, Boston and Section of General Internal Medicine, Department of Medicine, Boston University
• Keywords: eCohort; mobile health ; Framingham Heart Study ; physical activity index ; reliability ; adherence

Abstract

eCohort Studies offer an efficient approach to data collection but are challenged by low adherence.We designed an eCohort (eFHS) embedded in the Framingham Heart Study (FHS).We evaluated adherence and reliability of the eFHS app surveys deployed at baseline (the in-person FHS exam) and every 3 months.We defined adherence as the proportion of participants who complete at least one survey during each 3 month period. Among the 1948 eFHS participants (mean age 53, 43% men), we found high adherence to baseline surveys (89%) but a decrease over time (58% at 3 months, 52% at 6 months and 40% at 12 months). eFHS participants who return surveys are more likely to be female (OR 1.57, 95% CI 1.18-2.09) and less likely to smoke (OR 0.48, 95% CI 0.30-0.81). We compared baseline measures obtained in eFHS app vs FHS exam data using Concordance Correlation Coefficient (CCC) and Bland-Altman plots. We observed moderate agreement for physical activity index (PAI) (mean difference = 2.27, CCC = 0.56) and CES-D (mean difference =-0.91, CCC=0.66) and high agreement for drinks per week (mean difference = 0.54, CCC= 0.82).Our results suggest that app surveys tend to collect a higher PAI and lower CES-D.