Acceptability of Active and Passive Data Collection Methods for Mobile Health Research: Cross-Sectional Survey of an Online Adult Sample in the United States.

Abstract

Background: Digital health technologies, including wearable devices and app-based cognitive and health assessments, are pervasive and crucial to better understanding important public health problems (eg, Alzheimer's disease and related dementias). Central to understanding mechanisms driving individuals' willingness to share various data streams are concerns regarding data privacy, security, and control over generated data.

Objective: This survey was designed to learn more about attitudes and opinions related to digital health technologies and the sharing of associated data.

Methods: A total of 1509 adults were recruited from Prolific to complete an online survey via Qualtrics. Of these, 1489 participants provided valid data for analyses. Participants completed a structured survey consisting of multiple modules after informed consent was provided. These included: (1) demographic characteristics; (2) prior research experience; (3) mobility factors (eg, use of mobility aids, driving frequency); (4) technology ownership (eg, smartphones, tablets, home Wi-Fi); (5) social media use (eg, frequency of engagement with platforms such as Facebook, Instagram, and TikTok); (6) willingness to contribute different types of data across categories, including activities, sensors, and metadata; (7) opinions about data control and privacy options (eg, data deletion, stream-specific control); and (8) willingness to interact with assistive technologies such as robots, for Instrumental Activities of Daily Living.

Results: The final cohort (N=1489) had a mean age of 35.5 years (SD 12.0), was 44% female (n=652), and predominantly identified as White (76%, n=1134), with high rates of smartphone ownership (99%, n=1479) and home Wi-Fi access (98%, n=1464). Participants were most willing to share data streams with clear health implications and least willing to share data streams with greater privacy or reidentification potential (eg, GPS location, in-vehicle dashcam footage). On average, people were willing to complete ambulatory cognitive assessments for 56.7 (SD 36.2) days, air quality monitoring for 58.1 (SD 37.7) days, and GPS location monitoring for 37 (SD 39.0) days. People expected control over their data, including the ability to delete all or specific streams of the data contributed for research. Most participants prioritized control over their data, with 71% (n=1061) favoring the ability to delete all data contributed for research purposes. Stream-specific data deletion (65%, n=960) and time-specific deletion (44%, n=653) were also valued; interest in sharing data with insurance providers (30%, n=453) or caregivers (26%, n=384) was notably lower.

Conclusions: Findings have implications for the design of digital health technologies and education-related to the use and implications of collected data.