If you think Fitbit is going to save us from obesity, think again.

COLBY COSH ON THE WEIRD SCIENCE THAT SEEMINGLY CONDEMNS FITNESS TRACKERS — BUT DOESN’T

It’s a weird one, all right. In 2010, a group of researchers, mostly from the University of Pittsburgh, made plans for a study designed to show whether high-tech “wearables” that monitor physical activity can help people lose weight. The scheme was not complicated: 471 overweight or obese participants were recruited and divided randomly into two groups. Both groups got six months of weight-loss advice, plans and help, of an oldfashioned kind, and were turned loose to pursue ordinary lives.

But not before one group was given and shown how to use an automatic fitness- monitoring device that tracked members’ energy consumption when they exercised. The other group wasn’t. Question: would the subjects in the device group do better than the controls, or merely come out even?

As you might have heard when the results were published in the Journal of the American Medical Association last week, the answer was “neither.” The individuals in the no-tech control group lost more weight, measured either in pounds or as a percentage of prior body mass, than the cyber- equipped group. This is consternating. A gadget designed to assist with weight loss appears to have had the effect of preventing it.

Millions of wearable fitnesstracking devices of various kinds must have been sold since the Pittsburgh scholars began this experiment. In 2016, your phone can probably count the number of steps you take in a day: my latest iPod does, as I found out a few months after I bought it. Could this silent information-gathering be a possible source of harm? Am I getting even fatter every time I peek to see how many miles I covered on my latest shopping excursion?

Well, anything that seems so unlikely probably is. The most important thing to know about the study is that both groups of participants, on average, lost somewhat impressive amounts of weight. The subjects were people aged 18- 35 who had initial body mass indices between 25 and 40. Through the six months of initial instruction and encouragement that both groups got, the average net weight loss was eight kilograms.

Another 18 months later, the controls were still 5.9 kg lighter than at the start of the study. They had lost ground, but only a bit. The wearables group, by contrast, still averaged 3.5 kg lighter than baseline. The researchers measured plenty of secondary health endpoints, such as body-fat percentage and cardiorespiratory efficiency, and the picture was about the same across the board: both groups improved, with the no-device control group doing a tad better.

What this tells us is not that fitness trackers are evil, but that depending carelessly on a fitness tracker may interfere with other weight- loss methods. The truth is that these experimental subjects received a level of mothering that would probably cost an ordinary person considerable cash. In the six-month run-in period, everybody attended weekly “group sessions” to talk about lifestyle and habits: if anyone missed a session, he was chased down and encouraged to appear at an extra one.

Attendees were “given feedback” on weight change at every meeting, assigned meal plans, made to keep a food diary and handed customized targets for weekly exercise time. The process is not described in much detail, but one envisions Weight Watchers plus Nurse Ratched. After the groups were split up, the subjects, in both groups, continued to get phone calls and text messages from the researchers.

Quickie news stories about the study might have led you to imagine that the people in the control group were mostly left alone during the experimental period. In fact, the controls logged their daily physical activity on a website. The difference between the groups was that the controls had to do the work of recording their daily activity and punching in the information manually. The fitness- tracker group had a high-tech armband to handle the data-gathering.

If you thought electronic fitness trackers were likely to save the Western world from widespread obesity, the results of the Pittsburgh study may be disappointing. If you’re just a regular person, and not a policymaker or a scientist, the study does not tell you whether a fitness-tracking wearable will help you — compared to anything else you might do on your own, or for free. There is a hint here that the important thing to do with a datarecording health device is to engage with the data: to really look at and think about the numbers, rather than treating a fitness device as a magical fat-melting amulet. Put that way, it is the opposite of a surprise.

DEPENDING CARELESSLY ON ONE OF THESE DEVICES MAY INTERFERE WITH OTHER WEIGHTLOSS METHODS.