Assessing the Effects of Running on Human Longevity

Armed with better tools, such as lightweight accelerometers, and given much more data to work with, epidemiologists are nowadays trying to quantify the degree to which specific forms of exercise are better or worse then others. Other teams are trying to put rigorous numbers to the dose-response curve for exercise: where is the optimal point when it comes to health and longevity? It is a given that regular moderate aerobic exercise is good for you, and the evidence for that is overwhelming. Sedentary people suffer shorter lives and a greater burden of age-related disease. But once we start to ask how much exercise is most advantageous, or whether one form of exercise is better than another, then the answers become much less certain. They depend far more upon interpretations of data and the limitations of specific data sets, the details of which can be quite complex - and will thus never appear in the popular press when specific research projects are discussed.

The only useful way to look at this sort of research is in aggregate, summed over many studies. At this point there are simply too few papers comparing different forms of exercise to say more than that it is an interesting topic: I would say that many more years of work are needed to assemble a good consensus for human data, and even then that consensus will be fuzzy around the edges, numbers subject to opinion on various scientific factions and their methodologies. Thus attempting to optimize lifestyle for health and longevity strikes me as a pleasant hobby to maintain, but no matter how much effort you put into it, you'll likely never find out whether you are in fact doing any better than the 80/20 level achieved with far less work.

Given that the future of our health will be increasingly determined by progress in medicine as we age, I would argue that we should direct our extra time and effort towards supporting research into therapies to treat the causes of aging, as opposed to chasing the mirage of a perfectly optimal lifestyle. Good enough is comparatively easy to achieve when it comes to personal health, while significantly more than that is a next to impossible goal. Meanwhile, absent new technologies based on the SENS rejuvenation research programs, even the healthiest of us in mid-life today have but a small chance of making it to 100, and we'll be very frail indeed should we manage to hold out to that point. Technology should be the focus.

An Hour of Running May Add 7 Hours to Your Life

Resarchers found that, compared to nonrunners, runners tended to live about three additional years, even if they run slowly or sporadically and smoke, drink or are overweight. No other form of exercise that researchers looked at showed comparable impacts on life span. The findings come as a follow-up to a study done three years ago, in which a group of distinguished exercise scientists scrutinized data from a large trove of medical and fitness tests. That analysis found that as little as five minutes of daily running was associated with prolonged life spans.

Over all, this new review reinforced the findings of the earlier research, the scientists determined. Cumulatively, the data indicated that running, whatever someone's pace or mileage, dropped a person's risk of premature death by almost 40 percent, a benefit that held true even when the researchers controlled for smoking, drinking and a history of health problems such as hypertension or obesity. Perhaps most interesting, the researchers calculated that, hour for hour, running statistically returns more time to people's lives than it consumes. Figuring two hours per week of training, since that was the average reported by runners in the earlier study, the researchers estimated that a typical runner would spend less than six months actually running over the course of almost 40 years, but could expect an increase in life expectancy of 3.2 years, for a net gain of about 2.8 years. In concrete terms, an hour of running statistically lengthens life expectancy by seven hours, the researchers report.

The gains in life expectancy are capped at around three extra years, however much people run. The good news is that prolonged running does not seem to become counterproductive for longevity, according to the data. Improvements in life expectancy generally plateaued at about four hours of running per week, but they did not decline. Meanwhile, other kinds of exercise also reliably benefited life expectancy, the researchers found, but not to the same degree as running. Walking, cycling and other activities, even if they required the same exertion as running, typically dropped the risk of premature death by about 12 percent. Why running should be so uniquely potent against early mortality remains uncertain, but it raises aerobic fitness, and high aerobic fitness is one of the best-known indicators of an individual's long-term health.

Running as a Key Lifestyle Medicine for Longevity

Running is a popular and convenient leisure-time physical activity (PA) with a significant impact on longevity. In general, runners have a 25-40% reduced risk of premature mortality and live approximately 3 years longer than non-runners. Recently, specific questions have emerged regarding the extent of the health benefits of running versus other types of PA, and perhaps more critically, whether there are diminishing returns on health and mortality outcomes with higher amounts of running. This review details the findings surrounding the impact of running on various health outcomes and premature mortality, highlights plausible underlying mechanisms linking running with chronic disease prevention and longevity, identifies the estimated additional life expectancy among runners and other active individuals, and discusses whether there is adequate evidence to suggest that longevity benefits are attenuated with higher doses of running.


This is the best article Ive read on exercise. My long held question was answered:

Perhaps most interesting, the researchers calculated that, hour for hour, running statistically returns more time to people's lives than it consumes.

Posted by: Norse at April 18th, 2017 4:16 AM

Some people have reported early-onset arthritis of the hip and knee due to frequent running. That IMO is enough to counteract the other benefits of running. And I'm not saying that every runner will necessarily get this kind of injury.

Posted by: Matheus at April 18th, 2017 6:35 AM

I'm an active hiker, skier and kayaker, but gave up on running at all. However, I prefer resistance exercise three times a week and at the age of 62 I feel this is more beneficial for my overall health.

Posted by: bardu at April 18th, 2017 10:51 PM

Suspecting that sprinting 5x100 meter discplined and all might confer similar protection with less danger to knees and joints.

Posted by: at April 19th, 2017 7:32 AM

All of these studies are confounded by underlying genetic health. People that are genetically healthier, have lower mutational load are more likely to do sports. If you do not compensate for this fact your study is completely WORTHLESS.

I was aware of this problem before but had assumed that researchers in the field would know about it. I decided to follow through on the links and look at one of the studies cited. You can look for yourself:

It does not mention genetic confounding AT ALL. The sheer incompetence is staggering.

In fact, there is an extreme theory, that is nonetheless consistent with the facts (as far as I have been able to determine, please let me know if you found more rigorous studies!) that the correlation between sports and health is completely spurious. Moreover, from the viewpoint of an evolutionary biologist it seems most likely that sports is an evolved habit that serves to signal the sportsmans genetic fitness.

I have mentioned this before on this blog and got only a snarky reply.

Posted by: Lee Wang at April 20th, 2017 11:54 AM

@Lee Wang: I think the point about the direction of causation is valid for studies on elite atheletes. Less so for more general studies on more average people, where there is a lot more data to crunch, and the corresponding animal studies that do show direction of causation are more applicable as comparisons.

Posted by: Reason at April 20th, 2017 3:27 PM

Dear @Reason,

I thought so as well and agree that corresponding animal studies are suggestive. Nonetheless, numerous studies have turned out negative. With the onslaught of low quality research in nutrition science it is important to personally read the studies involved.
For instance, the study cited in the news paper article of the original post does not study elite runners but looks at various groups of runners that are more or less fanatic about their sport. Yet it does not take into account genetic confounding which means nothing can be concluded from this study.

Do you know of a study that does take into account genetic confounding?

See e.g.

On another note, I have been reading about the mitochrondial theory of aging. Is the theory as espoused in Ending aging or Aubrey de Greys thesis still the most up to date? Or is there some other source that is more up to date?

Posted by: Lee Wang at April 21st, 2017 9:23 AM

Post a comment; thoughtful, considered opinions are valued. Comments incorporating ad hominem attacks, advertising, and other forms of inappropriate behavior are likely to be deleted.

Note that there is a comment feed for those who like to keep up with conversations.