Online Program Home
My Program

Abstract Details

Activity Number: 662 - Methods for Meta-Analysis, and Longitudinal and Clustered Data
Type: Contributed
Date/Time: Thursday, August 1, 2019 : 10:30 AM to 12:20 PM
Sponsor: Section on Statistics in Epidemiology
Abstract #306613 Presentation
Title: Validation of Sleep Measures Derived from Phone-Based Activity Data Compared to Self-Report
Author(s): Briana Cameron* and Devika Dhamija and Matthew McIntyre and Robert Gentleman and 23andMe Research Team
Companies: 23andMe and 23andMe and 23andMe and 23andMe and 23andMe
Keywords: mobile; sleep; activity; longitudinal; HealthKit; steps
Abstract:

The increasing use of smartphones and wearable activity trackers presents a unique opportunity to collect activity and sleep data at a large scale. This data is valuable in providing a near-continuous measure of activity and sleep across large cohorts over long periods of time. At 23andMe, we are able to collect phone-based activity data from participants who have consented to research through Apple’s HealthKit integration. With over 10,000 customers providing up to 1.5 years of activity data, the utility and validity of phone-based measures can be demonstrated through a comparison with 23andMe’s database of self-reported survey data. In one study, customers were asked to answer a series of questions regarding their previous night of sleep; this data was used to look at the concordance of phone-derived measures of sleep with self-report sleep habits. The results of this large-scale validation study will be presented, as well as a discussion of the opportunities in using phone-based measures of sleep. Finally, a discussion of the challenges in using mobile-based data collection will be presented with a particular focus on how phone-use behaviors impact the collection of mobile data.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2019 program