Med Sci Sports Exerc. 2007 May;39(5):8229.
Longitudinal Modeling of the Relationship between Age and Maximal Heart Rate.
Gellish RL, Goslin BR, Olson RE, McDonald A, Russi GD, Moudgil VK.
School of Health Sciences, Oakland University, Rochester, MI.
PURPOSE:: Maximal heart rate (HRmax)prediction equations based on a person's
age are frequently used in prescribing exercise intensity and other clinical
applications. Results from various crosssectional studies have shown a linear
decrease in HRmax during exercise with increasing age. However, it is less well
established that longitudinal tracking of the same individuals' HRmax as they
age exhibits an identical linear relationship. This study examined the
longitudinal relationship between age and HRmax during exercise. METHODS:: A
retrospective analysis of maximal graded exercise test (GXT) results for members
participating in a universitybased healthassessment/fitness center between
1978 and 2003 was undertaken in 2006. Records were examined from individuals (N
= 132) of both sexes who represented a broad range of age and fitness levels and
who had multiple GXT (total N = 908) conducted over 25 years. HRmaxprediction
equations based on participants' age and HRmax elicited during the tests were
developed using a linear mixedmodels statistical analysis approach. RESULTS::
Clinical measurements obtained during the administration of the GXT included in
this longitudinal study resulted in the generation of a univariate prediction
model: HRmax = 207  0.7 x age. Model parameters were highly statistically
significant (P < 0.001). CONCLUSIONS:: The relationship between age and HRmax
during exercise developed in this longitudinal study has resulted in a
prediction equation appreciably different from the conventional HRmax formula
(220  age) often used in exercise prescription, and it confirms findings from
recent crosssectional investigations of HRmax.
revised method of estimating maximum heart rate
revised method of estimating maximum heart rate
In another forum, Alissa mentioned limitations to the traditional method of estimating maximum heart rate (220age). Below I've pasted the PubMed abstract of a May 2007 journal article that presents a method that more accurately reflects people's true max heart rate. In the full text of the article, the authors also provide a couple of more elaborate formulas that provide estimates of max HR that are slight better still.
ET
Excel code for revised formulas for maximum heart rate
In case anyone is interested, below is some Excel code that will compute your maximum heart rate according to the traditional formula and according to the revised methods in the May journal article by Gellish.
I had to use some formatting workarounds to get it to be legible in the post because I couldn'tor didn't know how topaste in the code directly from Excel. So you won't be able to just copy and paste; you'll have to manually type them into your spreadsheet program.
Row............Column A.....................................................Column B
1.................your date of birth.......................................... fill in as MM/DD/YYYY
2.................Today's date................................................=TODAY()
3................ Your exact age............................................ =(B2B1)/365.25
4................ Age squared................................................=B3^2
5
6..................Estimates of Max Heart Rate:
7
8................ traditional method..........................................=220B3
9................"preferred" per Gellish 2007 article....................=206.90.67*B3
10................using 2nd order ("square") term......................=191.50.007*B4
11................most complex (1st + 2nd order terms)..............=163+1.16*B30.018*B4
I had to use some formatting workarounds to get it to be legible in the post because I couldn'tor didn't know how topaste in the code directly from Excel. So you won't be able to just copy and paste; you'll have to manually type them into your spreadsheet program.
Row............Column A.....................................................Column B
1.................your date of birth.......................................... fill in as MM/DD/YYYY
2.................Today's date................................................=TODAY()
3................ Your exact age............................................ =(B2B1)/365.25
4................ Age squared................................................=B3^2
5
6..................Estimates of Max Heart Rate:
7
8................ traditional method..........................................=220B3
9................"preferred" per Gellish 2007 article....................=206.90.67*B3
10................using 2nd order ("square") term......................=191.50.007*B4
11................most complex (1st + 2nd order terms)..............=163+1.16*B30.018*B4
ET
The sample size did not seem small at all. There were 132 subjects with test procedures repeatedly for a total sample size of 908 testing procedures.
Here's a quote from the Discussion section with a caveat addressing your point about estimation vs direct measurement:
Here's a quote from the Discussion section with a caveat addressing your point about estimation vs direct measurement:
In any case, formulas do seem to enjoy some popularity, and this seems to be an improvement over the one that has represented the conventional wisdom.The final linear prediction equation developed in the current study, HRmax = 207  0.7 × age, has confidence interval bounds corresponding to ± 1 SD of the population's mean HRmax (for people roughly 3075 yr of age) that reflect a range of ± 58 bpm. Alternatively, the nonlinear equation detected in this research, HRmax = 192  0.007 × age2, though perhaps less desirable from a usability point of view, has yielded even tighter corresponding confidence interval bounds of ± 25 bpm. These ranges are narrower than the variations inherent in some other HRmaxprediction equations, including (probably) the formula of 220  age. Despite this assurance, prescriptions of exercise intensity should be based on direct measurements of HRmax if possible, because an equation may not predict the true HRmax in some individuals or for specific populations and modes of exercise (36). Moreover, because percentages of HRmax vary considerably in relation to individual anaerobic threshold levels, reliance on standard heart rateintensity guidelines can result in differing levels of metabolic stress across individuals (17,27). Accordingly, subjective ratings of perceived exertion during physical activity should always be used as a complementary way of monitoring exercise intensity along with checking heart rate.
ET
I think they're missing the point. Which is that there are differences in people's HR trend over the years, due to keeping fit as well as maybe other factors outside our control. If we don't keep fit, no doubt the MHR will drop with the 220age rule. If you DO keep fit, HR = 205  ½ age may be a better fit. Looking for some magical average formula is an absurdity. It's surprising that researchers who bothered to look at the trend for single people over time (i.e "longitudinal" ?) didn't even think of this correlation. What was the work for, obsession with averages rather than finding out why HRs are different? Sometimes I think that the only result of research is that the researchers demonstrate what nuts they are.
I've 1735 MHR and I'm 66; and 205  33 = 172.
I've 1735 MHR and I'm 66; and 205  33 = 172.
78y, 188cm, 87kg, MHR 155. Last 2k (24 May 19) 8.46.6@22
jamesg wrote:Looking for some magical average formula is an absurdity. It's surprising that researchers who bothered to look at the trend for single people over time (i.e "longitudinal" ?) didn't even think of this correlation. What was the work for, obsession with averages rather than finding out why HRs are different? Sometimes I think that the only result of research is that the researchers demonstrate what nuts they are.
I agree totally. Your formula 205  1/2 age is a much better formula for (very) fit people (like us ).
Tom
That's just the point: MHR formulae are useless, because they try to shove us into some average when no one is. No doubt the real numbers could say a lot to anyone who might care to listen. The question would be, what's your formula?
First go, we could use the 220Age. So at 20 I'd guess I had 200 MHR. I'm 66 and I've lost 25; and 25/(6620) = 0.54
The only reason for 220age is to avoid unfit people killing themselves on day 1. If then they use a flat % of HR rather than Karvonen, there'll be no risk at all: say 50y: 22050=170; 70% = 120. Accuracy has got nothing to do with it, the less accurate the better, so long as it's low.
First go, we could use the 220Age. So at 20 I'd guess I had 200 MHR. I'm 66 and I've lost 25; and 25/(6620) = 0.54
The only reason for 220age is to avoid unfit people killing themselves on day 1. If then they use a flat % of HR rather than Karvonen, there'll be no risk at all: say 50y: 22050=170; 70% = 120. Accuracy has got nothing to do with it, the less accurate the better, so long as it's low.
78y, 188cm, 87kg, MHR 155. Last 2k (24 May 19) 8.46.6@22
Exactly. Combine that with BMI (another useless average based formula) and you're all set.jamesg wrote:The only reason for 220age is to avoid unfit people killing themselves on day 1. If then they use a flat % of HR rather than Karvonen, there'll be no risk at all: say 50y: 22050=170; 70% = 120. Accuracy has got nothing to do with it, the less accurate the better, so long as it's low.
This forum is titled "Links and Articles". So far much of the commentary has involved a lot of personal opinion with little data to back it up. Of course one's individual data ultimately trumps what might have been observed in hundreds of other people, but only for that individual.
If you think all formulas are "useless", maybe you're right, but back it up with some data. Don't ask us to simply take your word for it.
By doing a study involving hundreds of people, you can calculate 95% confidence intervals, so that you have reasonable assurance that the results apply to 95 out of every 100 people of similar characteristics. The article abstract says that the subjects had varying levels of fitness, so it may be reasonable to argue that you are among the 5 people out of 100 nonjocks to which that particular study might not apply. (BTW, the article also cites previous studies finding that max HR is independent of level of fitness.) But how do you know whether your personal "mileage" applies to most other people?
On a related point, calling researchers "nuts" or other such names is not helpful and certainly not informative.
One of the comments criticized the study on the assumption that the sample size was small. Though reading the article's abstract shows that that was not a valid criticism, at least it addressed a scientific / statistical issue.
Ironically, statements that such studies are hogwash are often based on their own anecdotal experience involving the smallest sample size imaginable, N=1.
So a question that the moderator may be the best one to ultimately address:
Shouldn't the "links and articles" forum be where people can find out about research on these questions and perhaps engage in a scientific discussion. (For example, something like, "One serious methodological flaw with Study A is the sample size was only 10. By contrast, Study B had a sample size of 200 and found...").
Anecdotal N=1 experiences, IMHO, can be shared in other forums.
If you think all formulas are "useless", maybe you're right, but back it up with some data. Don't ask us to simply take your word for it.
By doing a study involving hundreds of people, you can calculate 95% confidence intervals, so that you have reasonable assurance that the results apply to 95 out of every 100 people of similar characteristics. The article abstract says that the subjects had varying levels of fitness, so it may be reasonable to argue that you are among the 5 people out of 100 nonjocks to which that particular study might not apply. (BTW, the article also cites previous studies finding that max HR is independent of level of fitness.) But how do you know whether your personal "mileage" applies to most other people?
On a related point, calling researchers "nuts" or other such names is not helpful and certainly not informative.
One of the comments criticized the study on the assumption that the sample size was small. Though reading the article's abstract shows that that was not a valid criticism, at least it addressed a scientific / statistical issue.
Ironically, statements that such studies are hogwash are often based on their own anecdotal experience involving the smallest sample size imaginable, N=1.
So a question that the moderator may be the best one to ultimately address:
Shouldn't the "links and articles" forum be where people can find out about research on these questions and perhaps engage in a scientific discussion. (For example, something like, "One serious methodological flaw with Study A is the sample size was only 10. By contrast, Study B had a sample size of 200 and found...").
Anecdotal N=1 experiences, IMHO, can be shared in other forums.
ET

 2k Poster
 Posts: 258
 Joined: September 10th, 2006, 12:13 pm
 Location: Durham, UK
Ah, but was the measurement accurate, or was it a spurious reading due to artefacts? Without recording individual beattobeat times you can never really be sure.Citroen wrote:It's still quite a long way off. 205  (age/2) gives 183. 220age gives 176.tomhz wrote:I agree totally. Your formula 205  1/2 age is a much better formula for (very) fit people (like us ).
Tom
My measured maximum while rowing a 3K race was 191.
Cheers
Dave
Suunto does beat to beat.Snail Space wrote:Ah, but was the measurement accurate, or was it a spurious reading due to artefacts? Without recording individual beattobeat times you can never really be sure.
http://faculty.css.edu/tboone2/asep/Robergs2.pdf discusses why 220age (and a few other formulae) aren't worth the paper they're written on.