Posted by Dave on August 25, 2011 | 10 Comments
The U.S. Department of Health and Human Services currently recommends that adults engage in exercise at least 150 minutes per week. They also say that more exercise is even better, but that also suggests that doing somewhat less than 150 minutes per week might be better than nothing.
If you’re an active runner, you might feel like 30 minutes per day, 5 days per week is the bare minimum you’d consider exercising, but if you’re doing no exercise at all, that much exercise might seem like an insurmountable hurdle. Back in the stone age when I was in my 20s and my kids were babies, I took a couple years off running. When I started back up again, it was very difficult for me to run even 10 minutes at a time. If a doctor had told me I needed to work out 30 minutes a day—especially if I had never exercised before—I might have given up after the first day!
Amazingly, few studies have systematically examined the benefits of exercising less than 150 minutes per week. That’s why a study published last week in Lancet has started to attract some attention. A Taiwanese team led by Chi Pang Wen tracked over 400,000 Taiwanese adults for 8 years, surveying them several times about physical activity levels and their health. Most importantly, their study allowed participants to indicate much lower amounts of exercise: as low as 75 minutes per week. Can just 15 minutes a day, 5 days a week, actually have a measurable impact on health?
One of the impressive things about this study’s data is how comprehensive it is. A major problem with many long-term studies is that participants get lost: They sign up for the study at the outset, but when the moment comes to find out if there has been any long-term impact, the researchers can’t locate them. In a health study, this might mean they are sick or dead, which would be a very important thing to know! In Taiwan, there is a national system of identification, government health care, and a national death registry, so this study actually lost very few participants.
Participants indicated what types of exercise they did and how many minutes per week they did each exercise. The researchers then calculated hazard ratios (HR) over the 8-year period of each person’s participation in the study. An HR of 1 was assigned to the inactive group — those reporting less than an hour a week of exercise. Any HR lower than one means you have a better chance of avoiding death. Ideally, your HR would be zero, but then you’d be immortal and probably wouldn’t bother with exercise. So what did the study find?
Amazingly, the researchers found that low-volume, moderate exercisers, who exercised just 75 minutes per week, had significantly better HRs than inactive people: 0.86 on average (compared to 1 for the inactive group). The result held up even after controlling for a number of other factors, like age, sex, smoking, drinking, and other health issues. For a typical 30-year-old, this means an increase of 2.5 to 3 years in life-expectancy! Does more, or more vigorous exercise improve your health even more? Yes, as this graph illustrates quite nicely:
As you can see, for every level of exercise the Wen’s team measured, there was a significant decrease in mortality: 30 minutes a day is better than 15, 60 is better than 30, and 90 is better than 60. The gains occur even faster with vigorous exercise (running and jogging count as vigorous; moderate exercise includes activities like gardening and walking).
What Wen’s team was most interested in were the low numbers: Doctors could recommend just 15 minutes a day of moderate exercise and expect to see improved life expectancy in their patients (Note that all the data in this study is correlational; we can’t be positive that the exercise causes a longer life-span. Maybe people who would naturally live longer are just more inclined to exercise). In this study, more than 50 percent of respondents were inactive, so such a modest change could have a huge impact.
One point Wen et al. didn’t emphasize is what happens to that curve at the other end. Notice that the blue curve representing vigorous exercise flattens out and trails off around 46 percent reduction in mortality, at about 50 minutes per day. What about runners (like myself) who exercise vigorously more than 50 minutes per day? I track my workouts somewhat obsessively, and I’ve run about 250 hours this year in just 237 days, or over 60 minutes per day. Given the research suggesting that extreme marathoners could be at risk of heart disease, what does this study add to that?
In the article I link above, I said there wasn’t enough evidence to conclude that moderately active runners, who aren’t running dozens of marathons, are at either an increased or decreased risk of heart disease. What about the high end of this study, the folks averaging 90 minutes or more of exercise per day? This graph offers a good summary of the results:
The error bars in this graph are 95 percent confidence intervals, which means there can be about a half-bar’s worth of overlap and the data points are still significantly different from each other. Graph A shows that all-cause death rates diminish for every exercise category, although the two final data points, high and very high, aren’t significantly different from each other. It’s only in cardiovascular disease that we see a leveling off of the hazard ratios for very high exercise levels, but there’s no evidence here of any actual harm from exercising as much as 90 minutes per day.
A table in the report does discuss ischaemic heart disease, the type discussed in my post on marathoning and heart disease. While the HR for this disease does go up from .39 to .57 when comparing medium-volume and high-volume exercise, the difference is not significant. Furthermore, deaths from ischaemic heart disease represent just a tiny portion of all deaths recorded in this study.
What does this mean for you?
Even low amounts of moderate exercise, as little as 15 minutes of walking per day, are associated with improved health and decreased chance of death. More, and more vigorous exercise, up to as much as 90 minutes per day, only improves these measures. If running is hazardous to your health, this study doesn’t offer any evidence that those hazards occur when you average 90 minutes or less of running per day.