Cellphones more reliable when counting the steps we take

MANY of us have invested in the wristbands or other wearable devices that tell us how many steps we’ve taken, calories we’ve burnt and other information. But an interesting new study that compared activity monitors found that while some are accurate when measuring step counts, others are way off.

And few are more accurate than the convenient and inexpensive apps you can find on your cellphone.

What’s more, the researchers say, while many of us hope these activity trackers may motivate us to become more active and healthier, none of them has proven able to persuade reluctant exercisers to start and stick with a workout routine.

The promise of these devices obviously relies heavily on their accuracy and ease of use. If the trackers tell us that we have moved more or less than we actually do, our responses may not be appropriate or ideal.

If, for instance, the monitor says that we have burned more calories than we actually have that day, we may overeat and gain weight. If, alternatively, the monitor says we have taken fewer steps in a day than we actually have, we may become discouraged, blame the device, throw it in a drawer and stop walking for exercise altogether.

Similarly, if a fitness monitor is difficult to program, requires frequent charging, feels uncomfortable or is pricey, many people who might benefit from more exercise will avoid buying or wearing the thing. As the authors of the study point out, only about 1%- 2% of Americans own an activity monitor, and many stop using the device within a few months of buying it.

Most Americans own a smartphone, however, and recently, a number of apps have become available that promise to measure someone’s steps, calories and so on, much the way a wearable fitness tracker does, but at a lower cost and, presumably, with greater convenience.

THE accuracy of these phone apps, however, has not been established, especially in comparison to the accuracy of the dedicated fitness trackers.

So for the new study, which was published this week in JAMA, researchers at the Perelman School of Medicine and the Center for Health Incentives and Behavioral Economics at the University of Pennsylvania in Philadelphia bought three of the most popular wearable fitness trackers — the Fitbit Flex, the Nike Fuelband and the Jawbone UP24 — and a clunkier pedometer and several types of accelerometers, often used by scientists to track physical activity.

They also looked at four fitness-tracking applications for iPhones and Androidbased phones, including an iPhone application from Fitbit and the Moves application for both types.

In all, they gathered 10 devices or apps that claim to track steps and otherwise monitor health and fitness.

Next, they rounded up 14 game adult volunteers and loaded them up with the devices. Each volunteer was fitted with a pedometer, two accelerometers, several wristband monitors and, in each pocket, a cellphone, one of which ran three iPhonebased fitness-tracking apps and the other featuring an Android phone running one app.

The volunteers then began walking on treadmills set to a gentle 5km/h pace. A researcher stood nearby and manually counted every step each volunteer took until that volunteer had finished first 500 and then, separately, 1,500 more steps.

Each volunteer completed two sessions of this step counting while wearing all of the devices. Then the researchers compared the step counts the devices had recorded against those measured by the researcher (who was unlikely to have miscounted by more than two).

The totals diverged considerably. The pedometer and the accelerometers were generally quite accurate, but one of the wristbands, the Fuelband, underreported the number of steps the volunteers had taken by more than 20%.

SOME of the other monitors were more accurate but, by and large, no more so than the smartphone apps. The upshot, said Mitesh S Patel, a professor at the University of Pennsylvania who oversaw the study, is that smartphones could offer “an easy, less expensive, but still accurate” means for people to track their activity.

But the broader issue, as Dr Patel and his colleagues pointed out in a related commentary published recently in JAMA, is that no fitness tracker of any kind has yet proved able to motivate people disinclined to exercise to start moving. Pedometers, after all, are accurate and cheap. But their widespread availability has not led to any overall increase in how much people walk.

Dr Patel said he and his colleagues were testing different types of incentives and rewards that could be incorporated into nextgeneration fitness apps and trackers.

But for now, while it is comforting to know that our smartphones can correctly count our activity, they won’t work unless we take that first step on our own and move. New York Times