This debate really misses the point. The key laboratory predictor of performance is the peak work rate achieved during the VO2max test. The values needed to be an exceptional cyclist or runner are well established. Since a VO2max of 100ml/kg/min is about 15% above the very highest values measured in very elite athletes, this athlete needs to achieve a peak workload that is clearly much higher than anyone else has ever achieved. So the question is: What was the peak work rate that the athlete achieved? Is it also exceptional?
The low RER at exercise termination indicates a calibration error most likely.
I wonder at what point we are going to just question the accuracy of the equipment rather than point fingers at the athlete like he cheated a test. As much as the technology he is using is solving the portability and affordability issue of testing, I question its accuracy, especially for this type of athlete.
Yeah, that's a good point, especially when someone starts sucking in, what, 7 or 8 litres of oxygen per minute! Even under the best of circumstances (and with the best of equipment), these measurements are finicky.
Why do these studies on exercise and longevity factor out the effects of exercise on blood pressure and so forth? As you say, it makes sense to factor out the effects of diet— maybe runners are also healthier eaters on average. But theoretically, if we knew all of the favorable effects of exercise on health, we could factor them all out and conclude that exercise does nothing to improve longetivy and is in fact harmful because it causes injuries.
I remember you pointing out this problem over ten years ago with one of the earlier U shaped curve studies on running and mortality rates and it seemed like such an obvious flaw in statistical analysis that people would stop doing it. Yet it seems to continue. What is the rationale behind it?
Yeah, I think this is a “damned if you do, damned if you don’t” situation. Not correcting for those factors will skew the results; but fully correcting for them will also skew the results. Something like BMI is probably both a confounding variable AND a causal variable in the relationship between exercise and longevity.
I don’t know of a perfect solution (other than doing randomized trials, which isn’t realistic when you want to see the effects over many decades), so it’s more a question of understanding the limitations of the data and not leaping to conclusions. Personally, I like it when studies clearly show both the adjusted and unadjusted data, so that the reader can see what effect the adjustment is having and draw conclusions accordingly.
Thanks for the response. The scientists writing these papers should be pointing out the interpretative problems in the paper. I haven’t looked at this latest one— I am pretty sure the ones from about ten years ago did not and I have seen people online treat the U- shaped mortality curve as established fact.
My new Coros watch gets it about est. VO2max. I don't pay attention to a lot of the data but that made me happy after seeing a super high number on my old Garmin Forerunner 55.
I missed the study on exercise and longevity. Thanks for pointing me to your column. The study is a perfect example of how non-expert media personalities get results (I hope unintentionally) mixed up. It must be extremely exhausting to try to explain study results with so much noise from lay people.
Thanks, Helen -- much appreciated! I think we all (including me) tend to view results through the lens of what we hope to see, so studies like this can turn into a Rorschach test. In that sense, it's useful to have a robust public debate about what the results mean, without assuming that any one perspective is indisputably correct. But yes, it's definitely frustrating when distorted narratives catch fire and spread rapidly!
Yeah... to be fair, there's definitely a possibility that the Instagram post was basically trolling, because they knew it would get the exercise physiology crowd worked up! At this level of sport, getting a mental edge is a big deal, so who knows what kind of 4D chess they're playing. :)
Yes, you could imagine “lower than usual,” but this is pushing it pretty far. One explanation that someone suggested to me is condensation on the airflow sensor if it was a prolonged workout/testing session.
This debate really misses the point. The key laboratory predictor of performance is the peak work rate achieved during the VO2max test. The values needed to be an exceptional cyclist or runner are well established. Since a VO2max of 100ml/kg/min is about 15% above the very highest values measured in very elite athletes, this athlete needs to achieve a peak workload that is clearly much higher than anyone else has ever achieved. So the question is: What was the peak work rate that the athlete achieved? Is it also exceptional?
The low RER at exercise termination indicates a calibration error most likely.
I wonder at what point we are going to just question the accuracy of the equipment rather than point fingers at the athlete like he cheated a test. As much as the technology he is using is solving the portability and affordability issue of testing, I question its accuracy, especially for this type of athlete.
Yeah, that's a good point, especially when someone starts sucking in, what, 7 or 8 litres of oxygen per minute! Even under the best of circumstances (and with the best of equipment), these measurements are finicky.
Why do these studies on exercise and longevity factor out the effects of exercise on blood pressure and so forth? As you say, it makes sense to factor out the effects of diet— maybe runners are also healthier eaters on average. But theoretically, if we knew all of the favorable effects of exercise on health, we could factor them all out and conclude that exercise does nothing to improve longetivy and is in fact harmful because it causes injuries.
I remember you pointing out this problem over ten years ago with one of the earlier U shaped curve studies on running and mortality rates and it seemed like such an obvious flaw in statistical analysis that people would stop doing it. Yet it seems to continue. What is the rationale behind it?
Yeah, I think this is a “damned if you do, damned if you don’t” situation. Not correcting for those factors will skew the results; but fully correcting for them will also skew the results. Something like BMI is probably both a confounding variable AND a causal variable in the relationship between exercise and longevity.
I don’t know of a perfect solution (other than doing randomized trials, which isn’t realistic when you want to see the effects over many decades), so it’s more a question of understanding the limitations of the data and not leaping to conclusions. Personally, I like it when studies clearly show both the adjusted and unadjusted data, so that the reader can see what effect the adjustment is having and draw conclusions accordingly.
Thanks for the response. The scientists writing these papers should be pointing out the interpretative problems in the paper. I haven’t looked at this latest one— I am pretty sure the ones from about ten years ago did not and I have seen people online treat the U- shaped mortality curve as established fact.
My new Coros watch gets it about est. VO2max. I don't pay attention to a lot of the data but that made me happy after seeing a super high number on my old Garmin Forerunner 55.
I missed the study on exercise and longevity. Thanks for pointing me to your column. The study is a perfect example of how non-expert media personalities get results (I hope unintentionally) mixed up. It must be extremely exhausting to try to explain study results with so much noise from lay people.
Thanks, Helen -- much appreciated! I think we all (including me) tend to view results through the lens of what we hope to see, so studies like this can turn into a Rorschach test. In that sense, it's useful to have a robust public debate about what the results mean, without assuming that any one perspective is indisputably correct. But yes, it's definitely frustrating when distorted narratives catch fire and spread rapidly!
Yeah... to be fair, there's definitely a possibility that the Instagram post was basically trolling, because they knew it would get the exercise physiology crowd worked up! At this level of sport, getting a mental edge is a big deal, so who knows what kind of 4D chess they're playing. :)
Yes, you could imagine “lower than usual,” but this is pushing it pretty far. One explanation that someone suggested to me is condensation on the airflow sensor if it was a prolonged workout/testing session.