Are fitness trackers effective?

Fitness trackers allow individuals, care providers and companies to measure activities like exercise and sleep, and increasingly physiological measurements like heart rate and blood glucose levels. For the first time, we are able to quantify our physical and physiological activities so minutely.

As improvements in measurements have led to scientific breakthroughs across fields, fitness trackers have the potential to create a significant impact on understanding and affecting people’s health related behavior.

{mosads}Although companies like Fitbit claim that their fitness trackers have positive outcomes for their customers, the scientific research on the effectiveness of the most wearable devices remains quite divided.

A recent study by researchers at the University of Pittsburgh published in JAMA showed that patients, who tracked their activities levels using fitness trackers lost less weight than others who were not monitoring their activities using such devices.  On the other hand, a study published in AJPM showed that older women exercise more when they can track their activity using a Fitbit.

This difference in outcomes can come from two important underlying factors: How is this information presented to individuals and how does it change the internal and external motivators that affect their behavior?

Currently, most fitness trackers track and report the number of steps taken, level of strenuous exercise and amount of sleep. Just presenting this information to users is not very useful because most users do not know how to internalize this information.Is walking 10,000 steps daily good? Probably, but what happens when I walk only 8,000 or 15,000 steps per day?

Once the novelty factor wears off, and users don’t see a direct link between data generated from these devices and their health outcomes in the short-term, they stop using these devices. In fact, this problem is even more problematic for mHealth apps, which users stop using in a few weeks. For continuous engagement, it is extremely important to provide individuals personalized and actionable insights based on the data collected from their fitness tracker.  

Another issue is that fitness trackers currently track very coarse measures of activities, e.g. steps. This does not provide a sufficient resolution into the intensity of the activity and ignores activities like weightlifting. Perhaps as sensors evolve, this problem can be addressed.

It is also unclear how these devices affect individual’s motivations. There are very few studies that evaluate the underlying mechanism through which mHealth interventions affect consumer behavior. In a recent study at Carnegie Mellon University, we found that a diet related mHealth app increases engagement by making it easier to record meals, while app assisted feedback from a dietitian promotes healthy eating by changing people’s intention to eat healthy. Clearly, this shows that different types of mHealth interventions affect people in very different ways.

The link between technology and behavior if often tricky and requires multiple levels of analysis. Unless we truly understand the interactions and exploit them to change people’s behavior, the promise of wearable technology will not be met.

More than 20 percent of the U.S. adults use a fitness tracker, according to a recent Forrester report [6]. However, just giving someone a fitness tracker will not change his/her health-related behavior, just like giving someone a tape measure will not make him/her an architect.  

Fitness trackers are only one part in the big puzzle of improving people’s health-related choices, albeit an important one. Insights from technology usage, behavioral economics and design need to be combined to create truly interactive apps, whose effectiveness can be measured using fitness trackers. Currently, the mHealth industry is in the Wild Wild West and a tremendous amount of experimentations need to be done to realize its true potential.

It is too early to write off fitness trackers. To truly change people’s behavior, we need a combination of engaging interfaces, personalized feedback and a sense of fun. Maybe a combination of Pokémon GO and Fitbit on steroids would do the trick.

Vibhanshu Abhishek is an assistant professor of information systems at Carnegie Mellon University’s Heinz College.

The views expressed by contributors are their own and not the views of The Hill.