More evidence against wearable activity trackers

Posted on September 26, 2016

2


This topic has previously been covered in Some insights on fitness and health trackers- Helping? Hurting???

Looking at a new study-

Effect of Wearable Technology Combined With a Lifestyle Intervention on Long-term Weight Loss

This is the best wearable tech study I’ve seen, overwhelmingly because of the two year follow up. It’s not what you lose, it’s what you maintain.

Results-

“For the enhanced intervention group, mean baseline weight was 96.3 kg and 24-month weight 89.3 kg. For the standard intervention group, mean baseline weight was 95.2 kg and 24-month weight was 92.8 kg. Weight change at 24 months differed significantly by intervention group (estimated mean weight loss, 3.5 kg in the enhanced intervention group and 5.9 kg in the standard intervention group; difference, 2.4 kg. Both groups had significant improvements in body composition, fitness, physical activity, and diet, with no significant difference between groups.”

You may have noticed it says the wearable / enhanced group started at 96.3 kg, after two years was at 89.3 kg, yet lost 3.5 kg, which was less than the standard intervention. This was an error in the abstract, which was also in the PDF:

wearable-tech-abstract-mistake-1wearable-tech-abstract-mistake-2

The 24 month numbers got switched around in the white chart above. I notified the corresponding author about this. They said thanks, that the publishing journal was aware, but there wasn’t time to correct it before going to print. Seems like an odd answer as to why the online version can’t be changed, but perhaps there are some caveats to quickly making changes like this.

Anyways, within a week it was corrected, and it’s worth bringing this up because,

  1. Upon initial release, I would not be surprised if some -those with a wearable bias- read the first couple sentences of the abstract’s results and said “See, wearables help,” then moved on to their heathen tweetin’.
  2.  Shows researchers make mistakes, even rudimentary ones…even in the Journal of the American Medical Association. The president of the United States just wrote a paper for this journal. It’s about as prestigious as it gets. Yet this falls through the cracks. We’ll come back to this.

The tech and subjects used

18-35 year olds were the subjects. If any tech lovers were thinking “I bet the subjects were too old to appreciate the elegance of wearables,” that’s ruled out. I was thinking along these lines, as having a good deal of clients above 50 has caused me to regularly encounter the “What do I need that for?” mantra. The people-get-set-in-their-ways thing is real.

-> Though if older people, who tend to be heavier than younger, who usually need to pay more attention to their weight, have issues using the tech, how helpful is it really? Truth about wearables- the people most likely to use / enjoy them are the people least likely to need intervention.

The study started in 2010 and recruited people until 2012, so we’re not talking smart watches for the wearable.

-> Apple lovers rejoice! “Ah, the tech undoubtedly did not have our simplicity.” Samsung lovers rejoice! “There was no chance of setting someone’s wrist on fire.” Google lovers rejoice! “We don’t know how to make hardware anyways.”

It was an upper arm tracker. While a wrist device could be different, for instance more aesthetically pleasing so you have better adherence to wearing it, the authors make the point accuracy of wrist devices has been brought into question, so there are issues with using those as well.

-> FitBit lovers bow your heads in disgrace! “We’re being sued for inaccuracy right now aren’t we.”

However, the wearable was only worn 170 out of 540 days. About 30% of the time. It’s tough to know what to do with this information. In one regard, you can say

“People aren’t wearing it enough.”

In another regard,

“Many don’t really like wearables, evidenced by how many stop using them.”

In another,

“30% is still a sizable chunk of the study, and that had no positive impact on weight loss. In fact, it had a negative impact. It’s hard to imagine upping the percentage would all of a sudden have a positive impact when so far 30% was only negative.”

I lean towards the latter explanations.

The wearable group trailed in nearly every category

While not statistically significant, the standard intervention group also did better in BMI, waist circumference, umbilical circumference, lost more fat mass, lost more body fat %, had better cardiorespiratory fitness.

The only thing the wearable group was noticeably better in was “Bone Mass.” This measure is confusing as the difference in bone density was virtually identical, yet there was a big difference in “Bone Mass.”

bone-mass-wearable-study

Notice how the unit is kilograms, yet the starting number is in the three thousands. That’d be a lot of bone mass. Something seems off here.

– 

Why did the wearable group do worse?

Many of us might expect to see either the wearable group do better, or no difference, but what we saw here was a negative influence of wearing something. The rationale of telling someone “Hey, it might not help, but it’s not hurting anything,” doesn’t work. On average (in this study), it WILL hurt your weight loss efforts.

The authors plausible explanations,

“This may be a result of the technology not being as effective for changing diet or physical activity behaviors compared with what was achieved with the standard intervention; however, the study found no significant difference in these measures between the standard intervention and enhanced intervention groups. Thus, the reason for this difference in weight loss between the standard intervention and enhanced intervention groups warrants further investigation.”

So, no explanations are given. Some ideas:

There are examples of humans compensating when it comes to their energy balance. Tell people a granola bar is low fat, and they may very well eat more granola bars. (Sub)consciously going “It’s less fat, so I can eat more of it.” They may eat so much more of it they end up eating more calories than if they were told the bar was not low fat! “The Dieter’s Paradox.”

Standing desks. Use them at work, you’re more likely to sit more at home. How much will depend on the person- so far it looks like you still get an overall caloric expenditure benefit. (Search “standing desk sitting at home compensation” if you wish.)

Exercise in general. Watch people get out of spin class and see how many immediately hit up the smoothie bar. “I just worked out really hard. I deserve a treat.” Offsetting the calories they burned. This is why some people DO start exercising, yet gain weight.

The people in this wearable tech study wore the device on average 241 minutes per day, each day worn. Four hours worth. What if these people somehow compensated for wearing the device? The study tracked activity (and diet), but it was self reported. Even if people were 100% accurate –utter fantasy– that still wouldn’t track non-exercise activity. Like how much you stand, how much you do this-

Leg Twitching GIF

Or how about how tracking can lead to exercise being less enjoyable? (Also discussed in this post.) While the wearable group may have said they did X amount of activity, maybe they didn’t do it as intensely because they weren’t enjoying it as much.

While the authors briefly mention self reporting dietary intake is a limitation, they don’t mention it’s wildly inconsistent- not thought to be, but is. That it’s a gigantic limitation. Humans are known to underreport intake by 12-38%, so who the hell knows how these people actually ate. Did the wearable group go “I tracked activity today, so I can eat more,” yet didn’t properly report the extra eating?

It’s surprising the authors didn’t mention any of this, or make much of an attempt to offer an explanation. Maybe the fact the paper started with such a basic error, despite having nine authors and you’d assume other peer reviewers, made me too cynical, but the paper struck me multiple times as lazily and or rushingly written. If that’s how the study was written, how was it conducted?

That, coupled with less than ideal tactics -some of which are expected due to the costs associated with better methods (though should we even bother with self reported intakes anymore???)- and I don’t think we can put the nail in the coffin for wearables yet.

But,

  1. this is another opportunity to show the tech mantra of “Just give people this device and they’ll change their habits” is comical
  2. the tech world is genuinely lost when it comes to changing people’s fitness / activity / eating habits

I don’t think this is hyperbole. Wearables were supposed to be the next big thing. THE fix for many. 579,000 results!wearable tech search results.png

(I didn’t click every link, but I’m betting “cause people to lose less weight” isn’t on any of those lists.)

Just like calorie counting apps, exercise apps, bariatric surgery, heart rate monitors, pedometers, home exercise equipment so we don’t have to travel to the gym, home workout videos so we can fit exercise into our five hours of television per day busy lives, endless drugs, innumerable supplements, the shake weight.

For fashion? To each their own. For those who find a benefit in getting messages on their wrists -ironically, I’ve found this helpful while training clients in the gym- alright. For fitness? I’m not recommending my clients go out and buy a wearable. I’m not at the point where I’d tell those who have one to stop using it, but we’re getting there.

Enter your email address to follow this blog and receive notifications of new posts by email.