Accuracy of Apps Compared to Fitness Bands

Do you need a Fitbit or can you simply trust a pedometer app on your smartphone to tell you how much you are moving throughout the day? University of Pennsylvania researchers put them to the test side-by-side counting steps walking on a treadmill at three miles per hour.

Smartphone Pedometer and Activity Apps

The latest generation of smartphones has sophisticated accelerometer chips that work in the background. They are sensing movement all day and that data can be accessed by apps and converted into all-day step counts, exercise tracking and sleep tracking. The researchers used a Samsung Galaxy S4 and an iPhone 5s running the Moves app on each, and the Fitbit app and Withings Health Mate app on the iPhone. For the test, the smartphones were carried in pants pockets while the subjects walked on a treadmill at three miles per hour for 500 and 1500 steps, each twice.

Wearable Fitness Bands and Hip Pedometers Tested

On their wrists, the test subject wore the three most popular fitness bands: Fitbit Flex, Jawbone UP24, and Nike FuelBand. On their waistbands, they wore the Yamax Digi-Walker SW-200, which has been used in many pedometer studies and is viewed as highly accurate for research. They also wore the Fitbit One and Fitbit Zip on their waistbands.

Treadmill Test: Walking 500 and 1,500 Steps

Loaded up and wearing all of these devices at the same time, each of 14 subjects then walked on a treadmill at three miles per hour while researchers visually counted their steps. They walked a set of 500 steps and a set of 1,500 steps, and the count on the apps, fitness bands and pedometers were recorded.

Results

• The waistband pedometers had the best accuracy, varying only one percent
• The Nike FuelBand was the least accurate, with highly variable results and usually undercounting steps by as much as 22 percent
• The Fitbit Flex and Jawbone UP24 fitness bands on the wrist had a wider range of variation than the hip pedometers but on average came close to the observed step count. The Flex was more likely to slightly undercount steps.
• The apps varied by +/- six percent and were mostly consistent between the 500 and 1500 step tests.

The bottom line is that the apps were pretty good at counting steps when you carry your smartphone in your pocket while walking for a steady bout. But the study did not test how well the devices and apps compared for all-day activity, such as a goal of 10,000 steps in varied conditions.

I have worn and used all these same fitness bands, pedometers, and apps for all-day data. My personal experience is that the step count on the Nike FuelBand SE is indeed 10-20 percent less than that of any Fitbit. That said, I continue to wear it as I find it very motivational for reaching my Nike Fuel goal each day.

The problem with using only a smartphone app is that you may not carry your phone with you continuously during the day. The step count is likely to be less than a fitness band or hip pedometer you wear consistently. If you don't care about missing out on some of these "lifestyle steps," that is acceptable.

Cost Factor

If you already have a smartphone, apps are free or nearly free. All it takes is consistently checking it and using any built-in motivational features. Furthermore, some GPS trackers, used for hiking, also include a fitness-tracking feature.

Hip-worn pedometers are far less expensive, and the Fitbit Zip is not only under $60 but you can also view your steps throughout the day both on the device and on the Fitbit app. The Yamax Digi-Walker SW-200 at under $20 is a stand-alone device and doesn't connect to a computer or app.

Wearable fitness bands are very popular but cost $90-$150. Even with the high cost and limitations of accuracy, many people find them motivating. They link to apps and online dashboards to track a variety of exercise and activity information, sleep, inactivity, and diet. A smart scale may also combine these tracking functions.

Motivation From Fitness Bands and Apps

The study authors previously wrote about the challenges that wearable devices face in actually motivating and changing our health behavior:

• The user must want to wear it and be able to afford it.
• The user must remember to wear it and recharge it.
• The wearable device must be accurate in tracking its targeted behavior.
• The information display must be easy to understand, motivate action and sustain the motivation.
• The authors note that individual encouragement (such as badges for milestones), social competition and collaboration (competing with friends or being on a team with those with the same device/app) and effective feedback loops are best for driving behavior change.