From: Systematic review of the validity and reliability of consumer-wearable activity trackers
 | Sample characteristics | Tracker wearing protocol | Measurements | Interdevice reliability results | ||||
---|---|---|---|---|---|---|---|---|
Author (year) | n | % female | Activity | Lab/field | Type | Placement | Measures | |
Adam Noah et. al (2013) [29] | 16 | 38 | Treadmill walking (3.5 mph), walking with incline (3.5 mph at 5Â %), jogging (5.5 mph), and stair stepping (30.5 centimeter step at 96 beats/min) | Lab | Ultra (Fitbit) | Waist (1 on each side) | Steps/min, kilocalories/min | ICC comparing 2 different devices worn at once: range 0.76-0.99 (steps), range 0.91-0.97 (kilocalories) |
23 | 43 | Classic (Fitbit) | Waist (1 on each side) | Steps/min, kilocalories/min | Comparing 2 different devices worn at once: ICC = average 0.88, range 0.86-0.91 (steps); average 0.87, range 0.74-0.92 (kilocalories) | |||
Diaz (2015) [15] | 23 | 57 | 6Â min each of treadmill walking (1.9 mph, 3.0 mph, 4.0 mph) and jogging (5.2 mph) | Lab | One (Fitbit) | 2 on right hip, 1 on left hip | Steps/min, kilocalories/min | Pearson CC left and right hips: 0.99 (steps), 0.97 (kilocalories); Pearson CC two right hip devices: 0.99 (steps), 0.96 (kilocalories) |
Flex (Fitbit) | 1 on each wrist | Steps/min, kilocalories/min | Pearson CC left and right wrists: 0.90 (steps), 0.95 (kilocalories) | |||||
Dontje (2015) [32] | 1 | 0 | 8 consecutive days excluding sleep and water-based activities | Field | Ultra (Fitbit) | 5 over left pants pocket, 5 over right pants pocket | Steps/min, steps/hour, steps/day | 10 devices collected movement (yes vs no) across minutes (98Â %); two-way median ICC of absolute agreement 0.90 (steps/min), 1.00 (steps/hour), 1.00 (steps/day); concordance CC 0.90 (steps/min), 1.00 (steps/hour), 0.99 (steps/day); from Bland-Altman plots 95Â % of the measures were within the boundaries of 28 steps above and below the mean difference; maximum difference for all devices was 3.3Â % |
Mammen (2012) [36] | 1 | 0 | 6 trials were performed while the researcher wore the devices and walked 20 steps | Lab | Ultra (Fitbit) | 3 trials on right hip, 3 trials on left hip | Steps/trial | All trackers were within +/−5 % of each other |
Meltzer (2015) [37] | 9 | Not reported | 1 night's sleep | Lab | Ultra (Fitbit) | 2 on nondominant wrist | TST, sleep efficiency | Among n = 7: no differences between trackers for TST (468.7 vs. 471.1 min normal mode; 300.4 vs. 289.9 min sensitive mode) or sleep efficiency (92.9 % vs. 93.3 % normal mode; 59.4 % vs. 57.4 % sensitive mode) |
Montgomery- Downs (2012) [38] | 3 | Not reported | 1 night's sleep | Lab | Classic (Fitbit) | 2 on nondominant wrist | Sleep vs. wake | 3 participant's recorded 96.5Â %, 99.1Â %, and 97.6Â % agreement at 1-minute epochs |
Takacs (2014) [43] | 30 | 50 | 5Â min each of treadmill walking (0.90, 1.12, 1.33, 1.54, 1.78 meters/second) | Lab | One (Fitbit) | 1 on the waist at each hip, 1 in front pocket of the dominant leg | Steps/trial, distance/trial | Across 5 treadmill speeds ICC: range 0.95-1.00 (steps), range 0.90-0.99 (distance) |