How Much Exercise Is Too Much?

How much should you exercise per day and per week? It may be prudent to limit chronic, vigorous exercise to no more than an hour a day and no more than five hours a week, taking at least one or two days off. For runners, the recommended upper limit for longevity benefits is 30 miles a week.

Check out these videos on athletic recovery:

Thanks for watching. I hope you’ll join in the evidence-based nutrition revolution!
-Michael Greger, MD FACLM

[BruceDouglass]

From the graph of the meta-analysis (shown at 1:23 in the video), the minimum of Mortality occurs at about 2000 MET-minutes/week. At 6 miles an hour (generally considered a jogging pace), that translates to 20 miles/week. This would be 6.7 miles, 3x per week, which at that pace would take 67 minutes for each run. Interesting.

[Eightfathorses]

I'm still doing my ultra marathons because they make me feel alive!

[greensmoothieparty]

I believe u-shaped curves representing health outcomes from an overwhelmingly suboptimal dietary lifestyle cohort is not very helpful to those of us who are more diligently striving to achieve nutritional excellence. Go team nutritarian!

[larrysellers]

They ran themselves into the ground. Time was they were referred to as ‘fitness fanatics’. Now they are presented as the norm and something we should all emulate. Take it easy out there. Your body will thank you for it.

[volcanis]

I understand that there are some reproduced and large-cohort studies showing that VO2 max is strongly correlated with all cause mortality, and there appears to be no upper limit of the benefits to this effect. How do we square these results off? Is the weight of the evidence actually in favour of being as fit as we can manage, rather than moderately active?

[guillaumeleguludec]

The problem with this kind of study is that is seldom gives insight about causality, it's usually mere correlation. Like higher mortality in people who jog more could be caused by a greater need to exercise for psychological reasons, which might be linked to more risky behavior, or something of that kind.

[StephenMarkTurner]

Some people exercise to "burn off" a poor diet, or eat too much refined carbohydrate or steak and eggs or whatever to support their programs. Still, probably good to heed this information.

[jeremyleake]

Dr Greger is always worth listening to. Though in this case I think he hasn’t quite been as thorough as the chart he shows with upward sloping mortality hazard ratio also shows that the confidence intervals widen with mets. That means the likelihood of not living as long with say 7000mets of exercise per week is not statistically different from that at 2000mets per week (the apparent optimum). It suggests either number of participants is too low or else there are confounding factors meaning a greater spread of life expectancy for the higher exercise group - which you can see from the lower confidence interval sloping downwards. Whereas it is most definitely statistically significant for improving life expectancy from moving from 0 mets to 2000mets as the confidence interval there is very small. A rough calculator to translate mets per week into hours per week of exercise for say cycling is hours = (mets x body weight in kg) / (power in watts on a bike x 200). If your moderate exercise for cycling is 200 watts and you weigh 75kg then 2000 mets is around 3.75 hours at that power per week. It scales, so 4000 mets is around 7.5 hours per week and so on. So even up to around 15 hours per week exercise (8000 mets) this chart does not show you are statistically significantly more likely to reduce your life expectancy if that exercise was all moderate in intensity. And in fact it suggests it may not matter if a small part of that was also high intensity.

[pavolhorvath]

I would guess, that shorter life in extreme sportsmen has to do something with too much calories consumed to support the extreme calorie demand. If you run marathons or ultras, you eat a lot of concentrated calories in various junk sources, just to fulfil your energy needs.

[WiseMindNutrition]

As with most things, moderation seems to be key!

[kateelderson]

Great admirer of yours Dr Greger. Reading your book How Not to Age and appreciate your humour, terrific application and knowledge. Hope USA will honour you, though Big Fast Food won't find you so hot. Respect from UK.

[fabianrares]

I'll stick to my 5-8h of weekly exercise at different intensity. Gym, bike, run and swim occasionally

[myqp]

Elaine and Jack LaLanne . Movement lots of easy movement all day long cannot go wrong.

[tallwaters]

Interesting, kind of confirms other things I've seen recently showing that it's less about the intensity of exercise and just the fact that you do it.

[Ron-knur]

I have been hiking for 49 years and some of my day hikes are 8 hours duration. I only need two days rest before I head out again. I am 71 yrs old with no chronic conditions. This study does not support the increase in mortality with exercise duration "Long-Term Leisure-Time Physical Activity Intensity and All-Cause and Cause-Specific Mortality: A Prospective Cohort of US Adults" Lower mortality was seen at vigorous physical activity of 150 to 300 min/week or moderate physical activity of 300 to 600 min per wk but no increase in mortality above those times.

[QuantumOverlord]

I do around 20 runs a week, but only 1/20 of those runs is strenuous, the others are very gentle (zone 2). There is so little research on this subject, and what there is makes little distinction between stuff like intensity, frequency, duration and so on. Marathons are particularly dangerous because you run down your glycogen and electolyte stores and are therefore forced to supplement on the go; but what about people that run alot but never these huge distances all in one go? There are so many other confounders too, more running = more risk of injry which equals extended periods of inactivity. How do we know its not those inactivity periods that cause issues?

[lorriangus]

That's fascinating, thanks for the information.

[JoshBender]

Very interesting! It makes sense in light of evolution. Running short distances is a survival advantage. Running long distances not so much.

[jstekic]

Did anyone do study on ultra runners? How training for 100 km, 100+miles, doing those races, 24, 36, ....even more than 100 hours of racing with no or minimum sleep affect the longevity?

[OmegaPointZen]

"Exercise moderately." Leonardo da Vinci.