Running Linked To Lower Arthritis Rates Than Walking

Massive study defies common logic and anti-runner bias.

Published

February 05, 2013

Running epidemiologist Paul Williams, Ph.D., often doesn’t get the respect he deserves. That’s because Williams doesn’t conduct randomized, controlled, double-blind studies. Those “gold standard” studies are the ones that usually make the headlines.

Instead, Williams collects and analyzes self-reported data from runners. Lots of it. He was a “big data” guy long before the term gained its present currency due to folks like Nate Silver and the “quants” on Wall Street. Indeed, Williams launched the National Runners Health Study in the pages of Runner’s World in 1991, and he’s been happily amassing runner data ever since. (He also collects data for a smaller National Walkers Health Study.)

Every year, Williams runs several of his impressive data sets through a strict statistical analysis, and tells us more about ourselves than we could possibly learn anywhere else. Today the journal Medicine & Science In Sports & Exercise has released his latest effort, a look into osteoarthritis (OA) and hip replacement (HR) rates among 74,752 runners (after 7.1 years of follow-up) and 14,625 walkers (after 5.7 years of follow-up).

The first major result is that the runners have roughly half the risk of osteoarthritis and hip replacement as the walkers. This is the opposite of what much common-logic and anti-runner bias would predict. Williams says that running lowers body mass index(BMI) more than walking, and that the lower body weight probably prevents arthritis. See the first table below for details.

Second, he looked at the OA and HR rates of runners logging different amounts of weekly mileage, and divided them into four groups: under 7; 7-14; 14-21; 21+. He set the first group as his standard, and found that the other groups (doing more mileage) had 15% to 18 % lower rates of OA, and 35% to 50% lower rates of HR. In other words, more running reduced the rates of OA and HR, again likely through lower BMI, but contrary to common belief. See the second table below.

Third, Williams found another surprising result: Extra hours of non-running exercises increased the rates of OA and HR where running had reduced them. This seems to support the hypothesis that humans are well-designed for the straight-ahead motion of running and walking compared to the twisting and torquing of many other sports activities.

Conclusion: “Whereas other exercise increased OA and hip replacement risk, running significantly reduced their risk due, in part, to running’s association with lower BMI.”

RUNNERS VS WALKERS: ARTHRITIS AND HIP REPLACEMENT RATES

Respondents	Osteoarthritis	Hip Replacement
74,752 runners*	2,004 (2.7%)	259 (.35%)
14,625 walkers#	696 (4.8%)	114 (.78%)

* 7.1 years of follow-up
# 5.7 years of follow-up

RUNNERS ONLY: OA AND HR RATES BY MILES/WEEK

	<7 miles/wk	7-14 miles/week	14-21 miles/week	21+ miles/week
Osteoarthritis	1.0*	-18.1%	-16.1%	-15.6%
Hip Replacement	1.0*	-35.1%	-50.4%	-38.5%

*set as 1.0 for a comparison standard with the other groups

Note: Williams reported his exercise data as MET hours/day. We have converted to miles/week of running using the best-available tool. See here for more.