Fitbit HR Inaccurate

thefuzz1290

May 19, 2016 (San Francisco, CA) – A comprehensive new study conducted by researchers at California State Polytechnic University, Pomona (“Cal Poly Pomona”) reveals that the PurePulse™ heart rate monitors in the Fitbit Surge™ and Charge HR™ bear an “extremely weak correlation” with actual users’ heart rates as measured by a true echocardiogram (ECG) and are “highly inaccurate during elevated physical activity.”

http://www.lieffcabraser.com/2016/05/comprehensive-testing-confirms-fitbits-purepulse-heart-rate-monitors-highly-inaccurate/

Christine_72

Yeah, I've never been interested in getting a HR monitor device. I started off with the zip and now have settled on the Alta.

Sarc_Warrior

I had the HR and have since switched to the jawbone UP24

Christine_72

indy_cruizer wrote: »

I had the HR and have since switched to the jawbone UP24

Was there much of a difference between the two @indy_cruizer ?

fitmom4lifemfp

It's been pretty well publicized that wrist-based HRM's are not as accurated as the ones with chest straps. Kind of a no-brainer.

DorkothyParker

I have the HR Charge. I put my info in on that class-action suit. I'm uncertain if it matters if the number is showing low. but it would be really unfortunate if I think I am working 150 BPM and I am only at 130BPM.

janejellyroll

I'm less interested in my actual heart rate than having a consistent measurement over time. I have no idea of what my Fitbit tells me is 140 BPM is actually 140 BPM, but as long as it's consistent with previous days, I can still use the information to help me understand my workouts.

It works for me, although I understand why other people might not care for it.

shinycrazy

I liked the idea of a heart rate monitor for my workouts and I knew a wrist based monitor wouldn't be as accurate for a workout. I ended up getting a Garmin Vivofit 2 with the hear rate monitor. It started out great and tracked my heart rate so I could get in to the right zones for cardio. Well, it's stopped working now and I don't recommend it. My 15 year old plus polar heart rate monitor works great though, just doesn't have any tracking since it's much older tech, just give you an average at the end.

NorthCascades

Yep. My girlfriend returned hers because it was about as accurate as rolling dice when she was exercising. It seemed better when she was sitting down relaxing, but you don't need a heart rate monitor for that.

fitmom4lifemfp wrote: »

It's been pretty well publicized that wrist-based HRM's are not as accurated as the ones with chest straps. Kind of a no-brainer.

While this is true, it's not what we're discussing. Plenty of wrist-based optical sensors work accurately and reliably, Fitbit's just isn't one of them.

Need2Exerc1se

I pretty much just use my Fitbit as a very expensive pedometer. It is accurate for counting steps. The resting HR is consistent with professional measurements I've had. I don't know or care about HR during other activity.

andrektan

I recently upgraded to a Blaze from a Flex (RIP, good friend) and my main annoyance is that the calorie burns seem about 50-100% higher than they should be..

CatchMom11

NorthCascades wrote: »

Yep. My girlfriend returned hers because it was about as accurate as rolling dice when she was exercising. It seemed better when she was sitting down relaxing, but you don't need a heart rate monitor for that.

fitmom4lifemfp wrote: »

It's been pretty well publicized that wrist-based HRM's are not as accurated as the ones with chest straps. Kind of a no-brainer.

While this is true, it's not what we're discussing. Plenty of wrist-based optical sensors work accurately and reliably, Fitbit's just isn't one of them.

It really doesn't matter what wrist-based device you're talking about - none are as accurate as a chest strap that is actually detecting your heart rate straight from the source.

Also, how is FitMom4Lifemfp's response "not what we're discussing"? The topic is about the inaccuracy of the Fitbit HR. Seems pretty on point to me.

thefuzz1290

You're right, they're not as accurate as chest monitors, but no one is going to wear a chest strap at all times. My Garmin is pretty close. I've used chest straps in the past and I know what my body feels like at 180-190bpm and when I start feeling that way I'm in the 180s.

NorthCascades

MandaB9780 wrote: »

It really doesn't matter what wrist-based device you're talking about - none are as accurate as a chest strap that is actually detecting your heart rate straight from the source.

Mio's wrist-HRMs are accurate. It's very slightly delayed compared to a chest strap, but the exercise community has embraced them because they're accurate and useful. Fitbit's HRM isn't just "slightly delayed," it's off by an average of 20 beats per minute (!!), it's wildly inaccurate. Yes, it really does matter which wrist-device you use, because some are much more accurate than others.

Here is a comparison of a Garmin wrist-HRM against a chest strap. The wrist one isn't flawless, but it does a good job.



Here are two images comparing the Fitbit Charge HR to a chest strap for the same exercise. You can see vague similarities; the Fitbit does not do a good job.

MandaB9780 wrote: »

Also, how is FitMom4Lifemfp's response "not what we're discussing"? The topic is about the inaccuracy of the Fitbit HR. Seems pretty on point to me.

The point isn't that chest straps are better than wrist straps (which isn't really true anyway for most people), and the Fitbit doesn't perform as well as other wrist straps. We're discussing the Fitbit's HR system specifically, not what body part is best.

DorkothyParker

Is the offage reading higher or lower during periods of intense exercise?

Christine_72

DorkothyParker wrote: »

Is the offage reading higher or lower during periods of intense exercise?

That's what I want to know

thefuzz1290

DorkothyParker wrote: »

Is the offage reading higher or lower during periods of intense exercise?

DorkothyParker wrote: »

Is the offage reading higher or lower during periods of intense exercise?

I believe it was mid points of exercise. If you were trying to stay around 160bpm, it could read 140-180...if the 20 point swing was both ways.

NorthCascades

DorkothyParker wrote: »

Is the offage reading higher or lower during periods of intense exercise?

It isn't consistent. It can be wrong in either direction, with an average error of 20 bpm (ignoring times when it could not read at all).

srecupid

Need2Exerc1se wrote: »

I pretty much just use my Fitbit as a very expensive pedometer. It is accurate for counting steps. The resting HR is consistent with professional measurements I've had. I don't know or care about HR during other activity.

True but the selling point of the HR is the HR. 20 BPM in either direction is a big difference. Personally I just use a basic vivofit 2 for counting my steps and put on a chest strap monitor for cardio sessions. If a wrist based wearable witb hear rate monitor doesn't have a good heart rate monitor then you might as well save the $50 or so and get the model without it

Need2Exerc1se

srecupid wrote: »

Need2Exerc1se wrote: »

I pretty much just use my Fitbit as a very expensive pedometer. It is accurate for counting steps. The resting HR is consistent with professional measurements I've had. I don't know or care about HR during other activity.

True but the selling point of the HR is the HR. 20 BPM in either direction is a big difference. Personally I just use a basic vivofit 2 for counting my steps and put on a chest strap monitor for cardio sessions. If a wrist based wearable witb hear rate monitor doesn't have a good heart rate monitor then you might as well save the $50 or so and get the model without it

That's a good point. Luckily mine was a gift.

Actually not so lucky since it was a gift from a close relative who expects me to wear it all the time and quite frankly I don't care what my HR is at every given moment or how many steps I take per day.

NorthCascades

Honestly I don't care how many steps I take in a day, either. I get more steps walking to the coffee shop than riding my bike 50 miles, so I don't consider it a good indication of how much exercise I've done.

ubermofish

My Charge HR was doing a pretty bad job until I stopped wearing it on my actual wrist, you need to move it up your arm a bit and tighten it so it stays. Now its pretty much +- 5bpm

Need2Exerc1se

ubermofish wrote: »

My Charge HR was doing a pretty bad job until I stopped wearing it on my actual wrist, you need to move it up your arm a bit and tighten it so it stays. Now its pretty much +- 5bpm

The instructions say to do this during exercise but I have a hard time getting mine to stay up there. It naturally wants to slip down to my wrist because the wrist is smaller. I usually get tired of pushing it back up and just let it go where it wants to go.

ubermofish

Need2Exerc1se wrote: »

ubermofish wrote: »

My Charge HR was doing a pretty bad job until I stopped wearing it on my actual wrist, you need to move it up your arm a bit and tighten it so it stays. Now its pretty much +- 5bpm

The instructions say to do this during exercise but I have a hard time getting mine to stay up there. It naturally wants to slip down to my wrist because the wrist is smaller. I usually get tired of pushing it back up and just let it go where it wants to go.

You gotta make it pretty tight

Need2Exerc1se

ubermofish wrote: »

Need2Exerc1se wrote: »

ubermofish wrote: »

My Charge HR was doing a pretty bad job until I stopped wearing it on my actual wrist, you need to move it up your arm a bit and tighten it so it stays. Now its pretty much +- 5bpm

The instructions say to do this during exercise but I have a hard time getting mine to stay up there. It naturally wants to slip down to my wrist because the wrist is smaller. I usually get tired of pushing it back up and just let it go where it wants to go.

You gotta make it pretty tight

Making it tight just makes it want to slip to the smaller more, especially when I get sweaty. Unless you mean so tight that it actually creates a bulge on the lower side like a tourniquet, which sounds incredibly uncomfortable and maybe a little dangerous.

MeanderingMammal

MandaB9780 wrote: »

It really doesn't matter what wrist-based device you're talking about - none are as accurate as a chest strap that is actually detecting your heart rate straight from the source.

It's not quite that simple.

Both electrical and optical measurement has some error, and both have a greater level of error at the higher range. At normal range the difference in relative accuracy is pretty negligible, and at the top end that goes from negligible to not very significant.

Factors that affect accuracy include device fit and placement, sampling frequency and with opticals they become vulnerable to skin tone and hair growth as well.

I think the materiality is that for the vast majority of people optical measurement is good enough. That said, the FitBit range do seem to come off worst in most tests I've seen results for.

If, on the other hand, you're interested in rate variation, recovery time, cumulative fatigue and sports performance, then undoubtedly an electrical measurement is superior. For that stuff you're spending 4-5 times the amount of money on a device though.

EvgeniZyntx

MeanderingMammal wrote: »

MandaB9780 wrote: »

It really doesn't matter what wrist-based device you're talking about - none are as accurate as a chest strap that is actually detecting your heart rate straight from the source.

It's not quite that simple.

Both electrical and optical measurement has some error, and both have a greater level of error at the higher range. At normal range the difference in relative accuracy is pretty negligible, and at the top end that goes from negligible to not very significant.

Factors that affect accuracy include device fit and placement, sampling frequency and with opticals they become vulnerable to skin tone and hair growth as well.

I think the materiality is that for the vast majority of people optical measurement is good enough. That said, the FitBit range do seem to come off worst in most tests I've seen results for.

If, on the other hand, you're interested in rate variation, recovery time, cumulative fatigue and sports performance, then undoubtedly an electrical measurement is superior. For that stuff you're spending 4-5 times the amount of money on a device though.

Good enough is correct.
This study was commissioned by lawyers with the interest of demonstrating what exactly? That these devices are inaccurate. Lawyers will then get paid when they sue FitBit. Conflict of interest? A bit.

Yet, if people bothered to actually read the study (and understand the math) they'd see a correlation of R=0.70-=0.85 and an error of about 10%. Damn better than the TDEE estimator equations.

In this study, these devices had an error of less than 8 beats at rest and about twice that during exercise. If you are doing HR training, they are not that useful (a 15+ beat error is a HR zone) but for general use they are good enough. It is unclear that these users had properly positioned the device tightly on the arm.

Other optical devices like the Mio or Scosche are intended to fit tightly on the mid arm and do not seem to have the same issues. My own testing (Scosche) and dcrainmakers (neither of which are valid as research but then neither is an unpublished lawyer paid article) show much smaller variance with a strap HRM.

abatonfan

But when does it cross into personal responsibility? If one has a medical condition that requires them to very closely monitor their pulse (the first thing that's coming into my mind is someone on digoxin who would need to hold their dose if their resting HR is below 60BPM), then it would be rather foolish for them to rely on a more inaccurate piece of technology. If I used a continuous glucose monitor to track my blood sugar (imagine it like how a fitbit measures HR), it would still be my personal responsibility to do a true blood glucose test (pricking my finger and checking it with my meter versus reading what the CGM thinks I am) if I am going to be making a diabetes decision that is highly dependent on my blood sugar (such as determining how much insulin to take to correct for a high blood sugar reading).

I was looking at some stats from my last workout at my university's gym. The elliptical machine I used said my average HR was 144 (3.1 mile distance, about 300kcals burned, 35 min duration, though my hands weren't on the HR bar the entire duration of the workout), while my charge HR says that my average HR was 135 (260kcals burned).

lemurcat12

I've noticed in the last year that the machines in my gym have a disclaimer about the heart rate that it's not necessarily accurate.

On the other hand, I've been wearing my Apple Watch (which I never trusted for HR), my Garmin (which I do -- chest strap), and occasionally checking on the machine, and for me the numbers have been identical. (I don't use machines that often, so this is just a spot check, but I was impressed with how much better the Apple Watch did than I expected.)

tomteboda

Useful information here. I am thinking about getting a hrm because my resting pulse has gotten super low (high 30s, low 40s consistently) , and my rate during activity has also dropped significantly, just to get more data before seeing a cardiologist.

jbrooks222

I have a FitBit Blaze. When I am exercising I put the band on a looser setting, slide it up my arm a few inches, and use a terry cloth tennis wristband to hold it in place. I also try not to swing my arms too much.
Overall I have found that if I baby it, the Blaze seems pretty accurate. If I don't, it loses contact with my arm and gets crazy heart rate readings that are way too high.
I still like my Blaze but expect a lot of people (quite reasonably) wouldn't have that much patience.