nav search
Data Centre Software Security DevOps Business Personal Tech Science Emergent Tech Bootnotes
BOFH
Lectures

back to article
Got a Fitbit? Thought you were achieving your goals? Better read this

One of my friends, who does absolutely no exercise and is what you'd call skinny-fat, bought a Fitbit and was proudly demonstrating how his heart rate (sat with me in the office at that moment) was just over 50. I had to explain to him that considering his lifestyle, either the Fitbit was wrong or he had a medical problem causing brachycardia, because the only person that should have a heart rate of 50 in the middle of the day is an endurance athlete, which he is most certainly not. I measured his HR myself with my watch and a finger - it was over 70.

Fitbit's CEO James Park has said that: "People need to use common sense. It’s not a medical-grade device; it’s a consumer device. In that setting, it works incredibly well." This echoes what they said back in January when the suit was first filed: "it’s also important to note that Fitbit trackers are designed to provide meaningful data to our users to help them reach their health and fitness goals, and are not intended to be scientific or medical devices."

In other words, for Fitbit, it's fine for getting a baseline and seeing how much your heartbeat deviates from that during various activities, but you shouldn't use it to accurately measure your heartrate. Which is a bit bizarre, because they say it's better than, for example, the HRM on the machines at the gym, which measure continuously. A deviation of even 10% is a massive difference when you're trying to get fit. If Fitbits aren't able to measure to any degree of accuracy except by using a previous baseline, they're useless and potentially dangerous.

26
1
Anonymous Coward

"Which is a bit bizarre, because they say it's better than, for example, the HRM on the machines at the gym, which measure continuously."

They don't measure continuously, fitness machines only measure when you have a firm grip on the electrical contacts and that's not possible during vigorous movement.

The one in the gym says my heart (just after running at 16km/hr, I can't hold the contacts at full sprint) is around 100, measuring it myself after the sprint puts it at 140+. My Fitbit is not accurate I suspect, it puts me at 130, but its a damn site more useful than the fitness machine's one and probably closer than my attempt to measure my pulse while on a treadmill at brisk walk.

Nuisance lawsuit, I wonder whose funding it.

4
14

"...either the Fitbit was wrong or he had a medical problem causing brachycardia, because the only person that should have a heart rate of 50 in the middle of the day is an endurance athlete..."

Ah - my special moment, at last! My resting pulse is usually below 50, and rarely under 40. That's measured manually, against a watch. I do regular cardio training but I think it's mostly genetic. It has raised the odd eyebrow among medics and blood donation peeps but nobody seemed worried.

I agree with your point though. "you shouldn't use it to accurately measure your heartrate"

Like many measuring instruments, it's far more useful for making relative measurements, rather than absolute ones.

My suggestion: Don't waste dosh on wrist-worn gimmicky tat, get off your arse and do some exercise. Just don't be daft about it.

11
0
Silver badge

"They don't measure continuously, fitness machines only measure when you have a firm grip on the electrical contacts and that's not possible during vigorous movement"

It can be worse than that - There have been occasions in the gym where the HRM on my treadmill was still displaying after I let go of the handles - it turned out that the person on the next treadmill had a radio linked HRM and my treadmill was picking up their signal - it did make me look rather fitter than usual for a while!

3
0
Meh

That's not actually true

Some people can have a lower natural HR not caused by a medical condition. The resting HR has a normal distribution with a mean of around 60 for a normal person. So some people can have resting HR of 50 or even lower without either being extremely fit or a medical condition.

I have a resting HR of 48. I do run recreationally - park run 27 min - 20k total a week 90ish KG

I do own a Fitbit with HR and resting it is quite accurate exercising less so but I do find its stairs count to be particularly problematic. Often when I run on the flat down by the canal it logs 30odd flights of stairs.

Finally when I do run I can sustain a Fitbit HR above 160 for an hour on a 10k. Again had this checked out and yes my HR really does do that again not normal

1
0
Trollface

um...

> when you have a firm grip [...] during vigorous movement

You all have filthy minds, you dirty dirty people!

10
1
Silver badge
Meh

RE Nuisance lawsuit, I wonder whose funding it.

Garmin... ;)

I have a Fitbit blaze and I wouldn't trust the HR monitor as anything other than an indication that I have a heart rate. Fitbit themselves say that it loses accuracy as the workout becomes more vigorous.

I don't expect it to be medically accurate and for my purposes its good enough. It increases as I exercise and decreases when I rest. The range is what's important for me. If it tells me my resting heart rate is circa 62/63 pretty much all the time and then suddenly starts reporting it as 50 then that would ring a bell, likewise the upper end. If it suddenly peaks beyond the kind of number I'm used to seeing then it's a "hmmm" moment.

when we weight ourselves,we do it on the same machine at the same time in the same way because we understand the numbers in relation to the previous numbers.

None of this is rocket science.

3
0
Anonymous Coward

Sample of one reporting as ordered.

I've got a fitbit surge and was what you would describe as fat-fat.

After much effort which was greatly helped by the fitbit I'm now hopefully only somewhat fat (over 30kg lost so far).

During this health drive I suffered was is often known as a "significant health event" and spend some days in a hospital bed with my surge on and a proper medical machine on me going 'ping'. I can confirm for that end of the scale the fitbit was very accurate.

Also, for what it's worth, I have now and had then and had when fatter a very low resting heart rate. So low that during the incident in the ambulance the paramedic decided an intravenous of atropine was essential. I didn't get it but did get about 6 holes in my arm, blame the bumpy roads and my shy veins, the paramedic was lovely (really lovely, highlight of the day seriously, blue tints in her hair, bit punky, great smile, very assuring, sexy as hell).

Since then I have raised the subject of my resting heart rate (it's about 48 today) with my GP and with 3 different consultants all of whom have specialities in, or close to, the matter at hand and they've all basically responded with "your heart rate is stable, you've got no arrythmia (sic), don't worry about it."

For my level of needs; basic exercise, tennis, tracking and staying active, I find the fitbit to be excellent. Only problems with it are...

a) the surge is ugly, I only got the surge because my wife wanted the surge because she wanted one with a GPS in it, and then we decided to get matching ones.

b) the strap is NOT REPLACEABLE, because some tart at fitbit decided to put the damn bluetooth aerial in it. eejits.

7
0
Stop

Armchair doctor?

@ckdizz -

"the only person that should have a heart rate of 50 in the middle of the day is an endurance athlete, which he is most certainly not."

That simply isn't true, IMO Looking at heart rates in this overly narrow way is like using BMI to determine obesity. Testing someone's heart rate after telling them they may have a medical condition, is a great way of getting a higher reading.

Resting heart rates can differ greatly - <boast>while I'm in good shape I'm not an elite athlete, I was hooked up to a heart rate monitor at the hospital not long ago (so I assume was accurate) and my resting heart rate was 37.</boast>

There are many things that affect heart rates apart from cardiovascular training, age, sex, diet has a massive effect, stress levels, simple genetics, lifestyle.

I'm guessing your friend perhaps doesn't drink much tea, coffee or alcohol? Perhaps he doesn't eat much red meat or foods that are bad for cholesterol levels? Does he walk to work? - Many factors.

None of this is to say that Fitbits are accurate but that heart rates do vary greatly in perfectly healthy people, much like BMI.

2
0

@AC - "They don't measure continuously, fitness machines only measure when you have a firm grip on the electrical contacts and that's not possible during vigorous movement."

Those HRMs are terrible, fortunately most gym machines I've seen will accept data from chest strap HRMs (Polar, Garmin, etc) that are much more accurate, the chest strap just pairs with the machine. In my gym, all of the cardio equipment have the metal grips for monitoring but the gym staff hand out chest straps sensors when people are trying to do actual heart rate based training,

0
0
Silver badge
Trollface

"...the paramedic was lovely (really lovely, highlight of the day seriously, blue tints in her hair, bit punky, great smile, very assuring, sexy as hell)."

"...very low resting heart rate. So low that during the incident in the ambulance..."

Oh come on! What is wrong with you?!?

8
0

"the only person that should have a heart rate of 50 in the middle of the day is an endurance athlete"

Sorry but that's just not true. While it's more common in endurance athletes it's not that uncommon in people who maintain a moderate fitness routine. Although I've got a bit fatter now, when I was slimmer and running a few times a week I had a resting heart rate of around 48. Once I started marathon training it got as low as 44 - measured both manually and with a medical blood pressure testing cuff.

That said, if your friend abstained totally from exercise it's fairly unlikely the reading was correct, especially in the middle of the day when you're not fully at rest.

Although I found heart rate based training useful when I was inexperienced, I found it far more effective to train based on feeling once I understood how different levels of effort should feel like. After all, depending on many physical factors the amount of effort it takes to push a particular heart rate can vary quite a bit on a day to day basis.

0
0

I find the Fitbit resting heart rate out by about 10% to high my rhr is about 47 but fb is almost always > 52 but that is the alto as it reports the right heart rate, in moderate to high exercise when it's right it's right when it's wrong it's way off, but so are most optical monitors I use a chest strap for exercise the Fitbit is for day to day stuff walking etc, resting heart rate would be nice if it was correct, I checked when I first got it lots of people say the same it reads too high, Fb claim they have a special algorithm wtf average over 7 days on wake up is the accepted rhr special algorithm is stupid

0
0

Also while I remember resting heart rate, max heart rate not too much of a worry we are all different, but heart rate recovery seems to be important, the accuracy is not so much an issue here just as long as it Is consistent in fact for safety sake being low is better.

I am no doctor just interested in me and how I work, but if a doc wants to pipe in would be interested

Also agree with others don't need an ecg just take your pulse and check the average over time, technology is great but hey not always needed

0
0

Los HR

I have had a very low HR for all my life. When I was younger I ran 6 miles / day on a regular basis. (Can't do that any longer because of back and knee issues). I'm nearly 72. I walk on my treadmill 60 miles / month on a 10% elevation (as high as my machine will go).

My resting HR as measured by my Polar HR machine and my Dr's HR is normally about 40. Has been around 38 often and low as 36 occasionally.

My cardiologist thinks I'm doing well. His stress test will increase my HR to 130 but I'm pushing hard to get there.

I don't like the strap on my chest that the Polar system requires but I realize it provides the best opportunity for signal so the sensors provide an accurate report.

For me, accuracy is everything. Polar has it.

0
0
J 3

And it's not just heart rate either...

I have had a Charge for about 1.5 years now, and it is very prone to "false positive steps", lets call them. When you are actually walking, it is accurate. But ride the car/bus with it on and it will measure hundreds, if not thousands of steps. Tapping your chairs arm also counts steps -- sometimes I get the vibration signaling 10,000 when I'm sitting down, and had been for a while.

OK, I understand that it might be hard to differentiate vibrations due to walking from those of car riding (although regularity of steps would be expected in the former but not in the latter, I expect), but their app, which is far from stellar too by the way, should obviously have a function where you can log such things (as you can many others). Say, two hour car trip? Then zero steps from time X to time Y. Same thing if you were sitting down for a couple of hours, etc. Very simple, isn't it? To me, it goes to show that the guys who make the device do not actually wear it.

0
0
Silver badge
Childcatcher

Not Fitbit for purpose

The fact is that Fitbit claims to be offering the sort of information that is intrinsically clinical in nature, but then backtracks by claiming, in essence, that it's just a bit of fun that isn't really supposed to be meaningful in any clinical sense, which supposedly justifies the gross inaccuracy of that clinical data.

Well, I'm sorry Fitbit, but you can't have it both ways. Either your product is meaningful for the purpose of measuring heart rate, or you should not be allowed to ship a product that fraudulently claims to measure heart rate.

It never ceases to amaze me how brazenly American corporations flaunt their deceptions, as though they somehow had a constitutionally protected right to be con-artists.

1
0
Silver badge

heart rate measurement

Getting an accurate heart rate is easy IF you have a decent EKG signal - but that's quite difficult to do with a consumer grade device applied by an untrained user who is exercising.

If there are only off by 25 bpm then I'm very impressed - that's pretty good.

4
6
Silver badge
Boffin

@ Version 1.0 ... Re: heart rate measurement

Being off by 25 bpm can be a major problem.

When exercising the max heart rate is a formula... 220 minus your age.

So when you're 20... that's 200 bpm.

When you're 50, that's 170 bpm.

If you're trying to get back in to shape, you could be at 170 bpm while your monitor is showing 145.

That's an accident waiting to happen, especially if your doctor tells you to do some exercise but not to exceed X like 145 bpm and you're really doing 170 bpm.

Also there are two types of monitors.

Photoplethysmography which is measuring your HR via an LED scan, and an EKG which is taking an electronic signal over a sensor or between two sensors. Two different methods.

You can get 'medical grade' devices as a consumer if you know where to shop. They do make and sell portable (2 x AAA battery) finger 'Pulse Ox' monitors. (Photoplethysmography) However these are not conducive to use while trying to work out.

But I digress. The point is that having a monitor off by 25bpm can be dangerous and its an accident waiting to happen. This is something to be a major concern because there are watches and monitors (besides Garmin) that have been making heart rate monitors for years that are more accurate.

0
2

Re: @ Version 1.0 ... heart rate measurement

"When exercising the max heart rate is a formula... 220 minus your age."

I would go so far as saying that anyone who uses this particular formula doesn't care at all about their HRMax. (it only applies to 80% of the populace anyway - the rest can get stuffed)

The proper way of doing it requires a warmup, then to work as hard as you can while monitoring your HR using a device that isn't a piece of shit like the Fitbits.

0
0
Silver badge

Re: @ Version 1.0 ... heart rate measurement

I never said that it wasn't a problem - sure, at either end of the numbers it's an issue - going from 170 to 190 is a big deal and thinking you're resting at 60 but it's actually 40 could be an issue.

Quite simply though, getting a moderately accurate measurement of an actively exercising subject with a piece of kit that costs less than $40 to manufacture is impressive. Even medical grade Photoplethysmography is inherently vulnerable to motion artifact even if the subject is flat of their back asleep.

0
0

And my Apple Watch is always dead on.

2
8
Silver badge

It would be. That's because your body adjusts to match the Jobs field that the Apple Watch creates.

29
0
LDS
Silver badge

The only issue is when the AppleWatch decides you ought to be dead because it knows your bank account is no longer able to sustain Apple business properly.

14
1
Silver badge

"And my Apple Watch is always dead on."

I would genuinely like to know how you know that. Have you conducted rigorous testing at rest, during moderate activity and during strenuous exercise while wearing the Apple Watch and hooked up to a medical grade ECG?

4
0
Boffin

Why medical grade when you can easily count it with a finger and another timer? Want MORE accuracy? Then count out the full minute rather than average from 15 seconds. It's not that hard to do, and you certainly don't need "medical grade ECG" to do this simple HR check. If you do think you need the medical grade ECG, perhaps there's something wrong with your brain? Then you need to go and get a "medical grade EEG" and do let us know how that goes?

0
1
Silver badge
Boffin

@Sadmin Re: Using your finger...

Yes, using your finger can give you good enough numbers, if you know what you're doing and where you're taking your pulse. And its accurate enough.

But try doing that in the middle of a work out where you have to use your watch to try and get a six second count. Its not easy.

And its Medical grade EKG not EEG since we're looking at your heart and not your brain activity. ;-)

As to 'medical grade vs commercial grade' its a bit of a misnomer since you can purchase medical tech if you want to pay the price.

0
0
Silver badge

"Why medical grade when you can easily count it with a finger and another timer?"

Yes that's just fine when you're sat still or just walking along but I'd be beyond extremely impressed if you can accurately count your pulse with your finger while engaged in strenuous exercise that has got your pulse into the 130 or 140 range, even more so if you could do it while comparing it to the read out on your watch.

0
0

"Have you conducted rigorous testing at rest, during moderate activity and during strenuous exercise while wearing the Apple Watch and hooked up to a medical grade ECG?"

This is not a valid defence, because YOU'VE never used medical grade ECG test gear either.

0
0

"I'd be beyond extremely impressed if you can accurately count your pulse with your finger while engaged in strenuous exercise that has got your pulse into the 130 or 140 range,"

Exactly right. I max out 200bpm, and if you try to poke and prod me when I doing that, I'm going to get quite cross.

0
0
Silver badge

"This is not a valid defence, because YOU'VE never used medical grade ECG test gear either."

Eh? How do you know I haven't? I haven't (though I could probably get my hands on some), but what's that got to do with anything? I also don't own an Apple watch.

Also I have nothing to defend, I didn't make any point, I enquired as to how the person claiming his Apple watch is totally accurate knows that that really is the case.

If I'd said "my Xioami Mi Band 1s is more accurate than your Apple watch" you'd have a point, but I would never make that claim, especially as my device cost £20 and I don't expect supreme accuracy from it.

0
0
Anonymous Coward

How to get in on these scams

1. Get a junk MBA and some fancy PR blurb

2. Promise a brand new tech soon to be worth billions

3. Put up a website with pseudo-scientific/outlandish bullshit

4. Get some free interns with STEM degrees desperate for work, and prepared to lie to keep it

5. Ask the VC science-illiterate community for millions of dollars

6. Talk a load of shite promising a break thru soon

7. Ask for even more money whilst delivering nothing that works

8. PROFIT

11
0
Silver badge
Trollface

Re: How to get in on these scams

Why are you talking about the Fontus? it's not in the article!

2
0
Headmaster

Re: How to get in on these scams

Do not forget to give it a fancy new name like "photoplethysmography ". A word which I think, as a lover of the language of Shakespeare and Milton, ought to be thoroughly ashamed of itself. Choice of pedantic grammar Nazi icon? What else?

1
1
Silver badge

@Artic Fox Re: How to get in on these scams

WTF? Seriously.

-=-

Do not forget to give it a fancy new name like "photoplethysmography "

-=-

This term and devices have existed long before fitbit was a wet dream or even a concept.

These devices have been around for years. ( Decades even. Source goes back to the 1960's.)

Ever notice the little device that the ER team / Surgery team slip on your index finger and it has wires up until the monitor that shows EKG, and HR and O2 levels. (If they put a cuff on you, it can also take and display BP.)

1
0

The fantastic dcrainmaker comes to the same conclusion.

http://www.dcrainmaker.com/2015/01/fitbit-surge-depth-review.html#optical-heart-rate-sensor-accuracy

Well almost the same conclusion. He finds that it isn't particularly accurate but his solution is a bit different. Instead of recommending a class action lawsuit, he recommends being aware of the trade off you're making between a convenient on wrist optical sensor and a more accurate but awkward chest strap.

4
0

Exactly

If I'm going to do interval training using a heart rate monitor, I'm not going to be relying on my fitbit (charge HR), I'm going to be wearing a proper heart rate monitor (I need to replace my polar one but I'm not doing mtb racing at the moment so can't justify the cost)

0
0
Silver badge

Re: Exactly

He also tests several other LED HRM's in conjunction with a chest strap based one and reports which ones work best,

I had a Scosche Rythm + which is basically a LED sensor without the useless smart watch attached, worked well for me until someone at work nicked it. Works anywhere on your arm so like a chest strap except more convenient.

0
0
Gimp

Dr Disco

A case of consulting a professional DJ?

Google tells me 20 BPM is the difference between "Don't worry, Be Happy" (70 bpm) and NIN's "Closer" (90 bpm).

6
0
Anonymous Coward

Re: Dr Disco

Let's not do Moby's 1000 then.

3
0
Stop

From FitBits own site:

"With high-intensity interval training, P90X, boxing, or other activities where your wrist is moving vigorously and non-rhythmically, the movement may prevent the sensor from finding an accurate heart rate. Similarly, with exercises such as weight lifting or rowing, your wrist muscles may flex in such a way that the band tightens and loosens during exercise. Try relaxing your wrist and staying still briefly (about 10 seconds), after which you should see an accurate heart rate reading. Note that your tracker will still provide accurate calorie burn readings during these types of exercise by analyzing your heart rate trends over the course of the workout."

https://help.fitbit.com/articles/en_US/Help_article/Heart-rate-FAQs

2
0

Quote from the report PDF (page 4, section D.2) linked to:

The mean age, body weight, and height of the subject pool was 23.23 ± 3.46 years, 168.43 ± 9.76 kg, and 70.05 ± 14.33 cm, respectively

So the average participant is 168kg (that's 26½ stone!) and is 70cm tall (that's about 2'4") and thus has a BMI of 343...

I fear more for their longevity than the accuracy of their FitBit

7
0

"So the average participant is 168kg (that's 26½ stone!) and is 70cm tall (that's about 2'4") and thus has a BMI of 343."

'Murica!

2
0

Imperial

I think those figures are more likely to be pounds and inches (168, 70, respectively) rather than a misplaced 1 (68 kg, 170cm). I very much doubt you could recruit people of such size (in both horizontal and vertical directions) and claim them as representative of the general population...

0
0

Maybe

My wife was complaining her Charge HR was not very accurate, but I tried it compared to my Garmin with chest strap and it was within a 1bpm or so. This was just resting though. Fortunately my daughter has now lost the device so problem over.

3
0

I don't know, but a study commissioned by the class action law firm with such a small sample size and with only 1 test per subject kinda tells me that this is as biased as all those "Expert Witness" testimonies you see in other lawsuits.

But seriously though, is there any chance that this study will not get laughed out of the court?

Also, isn't it also obvious that you can't accept the true heart of the gadgets since there are so many thing that can affect the reading.

It's almost like suing to make a quick buck. /s

3
1
Silver badge

I assume that Fitbit is intelligent enough to have included disclaimers in their T&Cs...

0
0

Not surprised

I got given a free fitbit through work (they gave one to all employees - the shittest model, naturally).

What a heinous fucking waste of time that was. Used it for about a week before I decided it was pointless tat. I'm not surprised to learn it was inaccurate - I mean the pedometer alone could be easily fooled by a quick fap in the work toilets during lunch breaks.

1
0
Anonymous Coward

Re: Not surprised

It's better than that. I spent a Saturday wheeling my aged parent around Wisley RHS gardens.

Apparently I'd gone cycling.

1
0
Silver badge

Re: Not surprised

Admit it - you sneaked a 'backy'.

1
0

Page:

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

The Register - Independent news and views for the tech community. Part of Situation Publishing