However, one of the most commonly tracked metrics — calories burned — may not be completely accurate, especially for one group of people.
uncertain
one → track → people
A study from Florida International University’s Medical Photonics Laboratory tracked 58 men and women of various ages and body weights who simultaneously wore smartwatches from Apple, Garmin, Samsung and Fitbit.
asserted
who → track → Apple
Participants alternated between two minutes of fast pedaling and five minutes of rest on a recumbent bicycle for 10 minutes, while researchers monitored their energy expenditure with a metabolic analyzer face mask.
asserted
researchers → alternate → mask
Researchers then compared how many calories recorded by the watches with measurements from the masks, which are considered the most reliable way to measure calorie expenditure.
asserted
which → compare → expenditure
All four watches recorded substantial errors in calorie tracking, ranging from 15% to 25% compared with the lab measurements.
asserted
watches → record → measurements
Three of the brands overestimated how many calories the wearers burned, with Apple having the smallest average error, while Garmin and Samsung produced larger overestimates.
asserted
Garmin → overestimate → overestimates
While Fitbit’s measurement errors weren’t consistently positive or negative compared with the lab readings, a different problem arose.
asserted
problem → compare → readings
The wearable produced several implausible estimates, including multiple measurements that exceeded 450% of the metabolic system’s readings.
asserted
that → produce → readings
Despite the errors across all four smartwatches, the researchers weren’t surprised by the findings, as previous studies have found similar inaccuracies.
asserted
studies → find → inaccuracies
What stood out most was the strong relationship between calorie-tracking error and body fat percentage.
asserted
stood → stand → error
As the body fat percentage of each wearer increased, the watches’ measurements of calories burned became more inaccurate, ranging from 50% to 100% or more.
For example, if the metabolic analyzer measured an energy expenditure of 100 calories, a watch with a 50% overestimate would display 150 calories burned.
asserted
watch → increase → calories
A 100% overestimate would double the calories.
asserted
overestimate → double → calories
Researchers aren’t entirely sure why body fat played such a role in calorie measurement, but they have a theory based on how the watches and their calorie-burning estimation algorithms were developed.
asserted
watches → play → theory
These types of watches use accelerometers to measure movement and light-based sensors to estimate heart rate.
asserted
types → use → rate
The data from these sensors is combined with age, sex, height and weight and fed into an algorithm that estimates how much energy a person expended.
asserted
person → combine → energy
The watches’ caloric algorithms were made using data from leaner people, so increased soft-tissue motion at the wrist, combined with limited ranges for calorie data, may be more likely to produce errors.
uncertain
motion → make → errors
The researchers also emphasized that while 50 calories may not seem like much, these differences can add up over time, resulting in thousands of calorie being miscounted.
uncertain
thousands → emphasize → calorie
Instead of tossing your smartwatch, exercise scientist and study author Jason Kostrna suggests treating the calorie tracker as an estimate rather than an exact measurement.
“If your goal is to lose weight, I would be conservative about using that number to determine how much to eat or to evaluate the success of a workout,” Kostrna wrote in an article for The Conversation.
uncertain
Kostrna → toss → Conversation
A companion study from Kostrna also found that heart rate measurements from watches were accurate and that wearables may provide useful information for other metrics, like distance and pace while walking, running or biking.
uncertain
wearables → find → distance
The Aporia
