Once again lazy health journalists have fallen down on the job and performed a disservice to the public. The new outrage: a multitude of media reports about a small study on the effect of jogging on mortality. Here are just a few headlines, published minutes after the study was published in the Journal of the American College of Cardiology:
- Fast running is as deadly as sitting on couch, scientists find (The Telegraph, UK)
- New study shows casual joggers outlast ironmen over long term (Chicago Tribune)
- When Exercise Does More Harm than Good (Time)
But a careful reading of the paper shows that this small study provides nowhere near enough evidence to support these headlines.
Let's first take a quick look at the study. Researchers in Denmark followed 1,098 healthy joggers and 413 sedentary but otherwise healthy non-joggers for 12 years. They found that light joggers had a significantly lower risk of death than sedentary non-joggers. This difference remained significant after adjustment for other risk factors. Moderate joggers had only two-thirds of the benefit observed in light joggers while strenuous joggers had nearly twice the risk of non-joggers, though in both cases these differences were not statistically significant after adjusting for other risk factors. The overall pattern remained consistent when the researchers sorted the data by the number of hours, the frequency, and the pace of jogging each week.
The authors conclude: "On the basis of current knowledge, if the goal is to decrease the risk of death and improve life expectancy, going for a leisurely jog a few times per week at a moderate pace is a good strategy. Higher doses of running are not only unnecessary but may also erode some of the remarkable longevity benefits conferred by lower doses of running."
But in an accompanying editorial Duck-chul Lee, Carl J. Lavie, and Rajesh Vedanthan write that "we still need more data to truly determine 'is more actually worse?' regarding exercise dose and prognosis. (It should be noted that these authors are among the leading proponents of the "less is more" exercise philosophy, so they would be inclined to support this study if the support were at all warranted.)
They point out several very important weaknesses in the study. First there is the general issue that this is just an observational study. There are a multitude of differences between the different groups in the study, and it is impossible to know with any certainty whether the jogging dose had any important causal relationship with the deaths that occurred in the study. Of course the researchers attempted to correct for many of the known differences but this is a highly imperfect science at best. And this was not a best case scenario. The mean age of the non-joggers in the study was 61.3 years while the mean age of all the joggers in the study ranged from the late 30s to the mid 40s. So this isn't just comparing apples and oranges, it's comparing a young juicy apple with a shriveled old lemon.
But even if it were possible to compare the groups and adjust for the differences there would still be another insuperable problem. The study simply had no statistical power to detect differences between the jogging groups. Although there were 128 deaths among the 413 non-joggers there were only 17 deaths in all the 1,098 joggers, including only 2 deaths among the 36 strenuous joggers. The authors calculated that those 2 deaths represented a two-fold increase in risk for the strenuous joggers compared to the non-joggers, but the enormous confidence interval, ranging from less than half the risk to an 8-fold increase, illustrates the futility of obtaining any sort of reasonable estimate of risk based on so few data points. Quite simply, there's just no way to estimate the comparative effect of strenuous jogging with no jogging based on the data in this study.
Now it is reasonable to propose that moderate exercise may be preferable to strenuous exercise, but it is also entirely possible that the opposite is true. Journalists and scientists have an obligation to fairly and accurately report the results of individual studies, and they have the further obligation to place those results in the context of what is already known in the field. By reporting the results of this one quite limited study with little or no critical perspective of its details or the larger context of the research, they have once again helped perpetuate the scientific illiteracy and innumeracy that is fast becoming one of the hallmarks of our time.
Here's another sharp take on this silly study by Alex Hutchinson in Runner's World.
Here's the journal's summary of the trial results:
