Accuracy of Polar "calories burned" study: in plain english

Many thanks go to Sadie Kate for forwarding me a copy of a study done at University of Tennessee to investigate the accuracy of the Polar OwnCal. OwnCal is Polar's method for estimating how many calories you burned during an exercise and to my knowledge they use this same method on all of their heart rate monitors.

A brief summary of the article can be found here. If you have access to online journals or libraries, here is the citation reference for the article:

CROUTER, S. E., C. ALBRIGHT, and D. R. BASSETT, JR. Accuracy of Polar S410 Heart Rate Monitor to Estimate Energy Cost of Exercise. Med. Sci. Sports Exerc., Vol. 36, No. 8, pp. 1433-1439, 2004.

What they did:
First they wanted to see how accurate Polar's predicted max VO2 and max heartrate predictions are. When you're setting up your Polar HRM it will give you suggested VO2max and HRmax numbers that are based on typical values for men and women, based on your age, height, and weight. The study measured the participants max VO2 and max heartrate by putting the guinea pigs onto treadmills and drawing blood to measure lactic acid concentrations.

Once they had measured the "true" VO2 and max HR, they set each guinea pig up with two HRMs. One HRM was configured with Polar's predicted HRM/VO2max, and the other HRM was configured with the actual HRM/VO2max that they had determined in the first part. They had the guinea pigs do different activities (treadmill, stationary bike, rowing machine) at different intensity levels, and evaluated all of the data.

What they found:
For men, the predicted VO2max was pretty close to the actual measured values. The predicted calories burned during exercise was also fairly accurate. If a man uses the Polar predicted VO2max to figure calories burned it will be about 2% off. If a man sets up the Polar with an actual measured VO2max value, it will be about 4% off.

For women, the Polar HRM predicted VO2max are not as accurate. VO2max was overestimated by 10.9mL/kg x min on average. If a woman uses the Polar predicted VO2max to estimate calories burned, it will be off by 33%. If a woman sets up the Polar with an actual measured VO2max value, it will be about 12% too high.

The ability to input your own custom VO2max and HR seems to be available only in the S-series Polar watches.

-------

My bottom line: Until I can afford a fancy visit to a sports medicine clinic to get my personal VO2max numbers measured, I think I'll just divide OwnCal calories burned number by 1.33. If I get that fancy visit and know my actual VO2max, I will divide divide OwnCal calories burned number by 1.12.

I know that using a HR monitor is still more accurate than the vague formulas out there where you plug in your weight, speed on bike, etc because it takes into account actual energy spent. You could be going faster and burning less calories due to tailwinds or going slower but burning more calories due to headwinds. A HRM is still a better way to gauge exercise regardless of activity type, too.

Are we talking real actual guinea pigs? or people? How do they get those HRM straps around the tiny guinea pig chests? (OK sorry - couldn't get the visual out of my head - back to reading your excellent synopsis!

Thanks for posting this Dianyla, and for the original article SK.

My Polar 720i shows that I burned more than 3200 calories on my 20 mile hill ride yesterday, and about the same amount on the 26 flat/windy ride on Saturday - granted - I'm pretty heavy, and I'm working hard, but, wow, that seems like a lot of energy!

So dividing that by 1.33 gives 2400 calories burned, not 3200.

Good thing I haven't been using that number to determine what to eat!

Quote:

---

Originally Posted by bikerz

Are we talking real actual guinea pigs? or people? How do they get those HRM straps around the tiny guinea pig chests?

---

Bwahahaah!!!! :p

Actually, they were all slim and fit college students, the average BMI for the women was 21! :eek: I'd love to see the same study repeated but using young, middle-aged, and older folk as well as different BMI ranges.

Though - you mention HRM on the animals, when I was googling around on Polar HRM's I found one that's designed for a horse! How cool is that! :cool:

Think about having to shave their little tiny chests and applying the bah-rump goo to make the HRM work! :eek:

Back on topic. Thank you, thank you, Dianyla, for the translation. That would have been very painful for me to do and nowhere near as clear. I hadn't read the article closely enough to realize that they had found a difference in outcomes between measured and predicted VO2max.

And how do they hold the handles on the rowing machine with their tiny little paws?

Sorry - we were all posting at once!

Does my question about the cals burned for heavier people make since - do you think the over-estimate is linear?

You're most welcome. I'm just glad to see the data, I was looking for this a few months ago when I first got my Polar and wanted to know how accurate this is. All I ran across was a bunch of men saying the OwnCal data was accurate.

Yeah, it's accurate... for THEM!

What really scares me is that the article made complete and total sense to me. I've had to do a lot of scientific reading of this nature lately what with going back to school this year. A year ago I would have had a very hard time reading that article, my eyes would have been crossing. Help! I'm turning to the dark side of academia...

Quote:

---

Originally Posted by bikerz

Good thing I haven't been using that number to determine what to eat!

---

The bad thing is... I have been using that as a dietary needs estimate. :eek:

No wonder I haven't been losing as much weight. :mad:

ETA: I think these numbers would be accurate regardless of body weight. It's all about how fast your heart is thumping away, and it will beat faster if it has to supply blood to a larger volume body. But who knows...

My friend, Tim the Mad Cyclist, is the one who gave me the paper. His blog is: http://madcyclist.blogspot.com/. He is a grad student in exercise physiology. He'd probably love to field the question.

Quote:

---

Originally Posted by Dianya

What really scares me is that the article made complete and total sense to me. I've had to do a lot of scientific reading of this nature lately what with going back to school this year. A year ago I would have had a very hard time reading that article, my eyes would have been crossing. Help! I'm turning to the dark side of academia...

---

My first job out of college involved a lot of reading and summarizing scientific (mostly toxicology) articles - I got so I could write a pretty coherent abstract of these really complicated articles I hardly understood (I was an anthro major, for crying out loud!), but REALLY understanding them - uh uh! So really - thanks a ton for clarifying all of that for us!

Quote:

---

Originally Posted by bikerz

And how do they hold the handles on the rowing machine with their tiny little paws?

---

And their stumpy little legs would never reach the pedals of an exercise bike. Maybe they just go around in those hamster wheels?

So you're telling me I didn't really burn 10,611 calories on Saturday, that it was really only 9337? :p (I have been tested. )
Oh no, I guess I need to lay of the Klondikes.

V.

Quote:

---

Originally Posted by Veronica

(I have been tested. )

---

Oohh, really? Wherehowmuch? :cool:

I joined a weight loss thread on BJ this year. Have been using Fitday program since Feb 2006. I started at 173 Feb 1, 168 Mar 1,165.5 Apr 1 and 168 May 1.That's a gain :( and I'd like to talk to anyone who has been using Fitday type programs to monitor calories in/out.
Here's the deal: I pushed hard all month on activities and staying under calories(truly I weigh/measure all my stuff even to counting out # nuts to eat and am about 800-1200 cal under calculations per day) but weight has just moved up/down (~165 low to 169 hi). I drink lots of liquids ...splenda sweetened stuff often but also lots of water. Is it possible to eat too few calories for amount of exercise...does that slow one down(metabolism)? I am very tired(or just bummed?) today. After yesterday's activities(severe gardening =4 hrs and 21 mile ride @15= mph I had some cramping in my quads for first time last night...nasty nasty hurt. I drank 56 oz liquids during and immediately after my ride. What's the cramping about? thoughts?
Also noticed the comments about calculation of calories burned. I had a resting metabolic rate study done in 1988 and it showed I needed 20% fewer calories than the average woman (who is she,anyway???). If that is factored into my basal rate does it also mean I require 20% fewer for all other exercise. I want to understand why I can't get/stay below 167. I can try to fudge back the numbers with Fitday (calories expended/biking), if that's the problem. I just don't know what to do ,eat more, eat less, exercise more,exercise less???:eek:
Thanks to anyone with patience to read this and provide any insight

UC Davis. It was about 300 dollars for body fat, lactate threshhold and VO2 max. It was a very interesting group of tests. It was almost three years ago and I want to go back and do it again.

OwnCal predicts my VO2 max to be 51. Testing showed it to be 41.

V.

Quote:

---

Originally Posted by easterbird

I had a resting metabolic rate study done in 1988 and it showed I needed 20% fewer calories than the average woman (who is she,anyway???). If that is factored into my basal rate does it also mean I require 20% fewer for all other exercise. I want to understand why I can't get/stay below 167. I can try to fudge back the numbers with Fitday (calories expended/biking), if that's the problem. I just don't know what to do ,eat more, eat less, exercise more,exercise less???:eek:
Thanks to anyone with patience to read this and provide any insight

---

Did you really mean 1988?? I think any metabolic study done almost 20 years ago would need to be revisited at the very least. Even if you mean 1998, that's almost 10 years ago and I would expect the data to be a bit suspect.

I don't know if any of this will apply to you, but here's some input - I've tried using calorie counting/exercise programs (FitDay and Balancelog) and they don't really work for me - I think that the calculations for how many calories the "average" person burns while cycling don't apply to a pretty well trained cyclist. Simply put, I think I burn fewer calories for the same ride than an "average" person of my weight because my body is efficient at cycling. Now, if I tried running or swimming, I'd be back to "average". But I found those programs to give me a HUGE allowance of calories for rides - more than I really wanted. If you are using a HRM, as noted in the above posts, there may be other shortcomings.

Also, a consistent problem with any program requiring a food log (and I'm not knocking these because they WORK) is diligence and honesty in the portion sizes. Are you actually weighing and measuring, etc.?

You seem to believe your metabolism requires you to eat less - it is possible you are not eating enough and this is making your weightloss difficult.

I'm a weight watchers fan. Been a lifetime member for almost 20 years and it really works for me when I am diligent, as I have been for the last 6 months. It's fairly painless, especially given how much I ride.

Re: the cramping - what are you drinking? Splenda sweetened "stuff" may be a nice at your desk beverage, but for a 21 mile ride at 15 mh, you may want to consider an actual sports drink with electrolytes. That should help the cramping.

Good luck to you. Maybe a visit to a nutritionist would be helpful?