Table 12 Sensor usage and compliance rates. Compliance is computed as the presence of data exceeding half of the measurement period per day among the participants that opted in for each sensor. Parts of this table have been reproduced from31. \({}^{\dagger }\)Realizd data only shows the times when phone interaction occurs, thus it is not possible to differentiate between periods with no interaction and the application not working.
From: TILES-2018, a longitudinal physiologic and behavioral data set of hospital workers
Sensor | Participant opt-in | Total hours | Compliance rate | Definition of compliance |
|---|---|---|---|---|
Fitbit | 208/212 (98%) | 236,725 | 73% | Average fraction of days per participant with >12 hours of data |
OMsignal | 208/212 (98%) | 44,240 | 60% | Average fraction of work days per participant with >6 hours of data |
Jelly Pro | 184/212 (87%) | 37,065 | 62% | Average fraction of work days per participant with >6 hours of data |
Owl-in-Ones | — | 37,065 | 98% | Uptime of the sensor network (244 Owl-in-Ones) |
Realizd | 172/212 (81%) | — | — | Compliance cannot be estimated due to sampling scheme\({}^{\dagger }\) |
