- When it comes to patient-generated health data, a new study finds variable validity in activities measured by wearable activity-tracking devices. In general, consumer activity trackers produce more accurate readings for step-counting than measures such as calorie counts or sleep time.
Researchers from the UNC Gillings School of Global Public Health and RTI International conducted a systematic review of 22 published articles covering the ability of Fitbit and Jawbone, two popular activity trackers, to measure steps, distance, physical activity, calories and sleep. The study report was published online Dec. 18 by the International Journal of Behavioral Nutrition and Physical Activity.
"Wearable devices that track physical activity, sleep and other behaviors are growing significantly in popularity," said Robert Furberg, PhD, senior clinical informaticist at RTI International and co-author of the study, in a public statement. "We conducted this review to understand how accurate these devices are."
The review indicated higher validity of steps, few studies on distance and physical activity, and lower validity for energy expenditure and sleep.
Several studies indicated that the step-counting feature was accurate both in the lab and in the field. Only one study assessed distance tracking for the Fitbit, finding that the device tends to over-estimate at slower speeds and under-estimate at faster speeds.
Two field-based studies compared accelerometry-assessed physical activity to results from the trackers, with one study finding high correlation (in Fitbits) while another study noted a wide range in correlation (in both Fitbit and Jawbone brands).
Using several different comparison measures, other researchers found that both tracker brands over- and under-estimated calories used, and over-estimated total sleep time.
No single tracker had a complete assessment across the five measures of steps, distance, physical activity, energy expenditure and sleep.
Overall, the review indicated higher validity of step counting, inconclusive findings (based on few studies) for distance and physical activity, and lower validity for calories (energy expenditure) and sleep.
A Pew Research Center report, published in 2013, found that 69 percent of adults measured at least one health indicator for themselves, a family member or friend using a tracking device, paper tracking or other method. Within the same survey, 60 percent of adults reported tracking weight, diet or exercise.
The UNC/RTI study authors noted that they encountered some challenges comparing devices across previous studies due to varying methods and reported results. “The findings should be viewed in light of the variety of study protocols and methodology,” they wrote.
They also pointed out that Fitbit and Jawbone trackers represented the majority of the consumer market at the time of the literature review in fall 2014. IDC reported market share from first quarter 2015 sales as follows: Fitbit (34 percent), Xiaomi (25 percent), Garmin (6 percent), Samsung (5 percent) and Jawbone (4 percent). “There is a built-in time lag between manufacturing and sale of activity trackers to use in the research laboratory and field,” the authors wrote. “Thus, some activity trackers that are currently available to consumers were not represented in this review, but should be considered as future studies accumulate on new devices and brands.”
The researchers concluded, “As new activity trackers and features are introduced to the market, documentation of the measurement properties can guide use in research settings.”
Photo credit: Fitbit