Friday, September 21, 2012

Does anyone bother to read more than the abstract?


Training plan I made in college that mostly follows the 10% rule.
Scholarship is sometimes a lonely endeavor.  Especially when it feels like everyone else is “doing it wrong.”  Today was one of those days—all the buzz online is about a new study titled “Can GPS be used to detect deleterious progression in training volume among runners?” in the Journal of Strength and Conditioning Research that purportedly shows that the oft-cited (and obviously arbitrary) “10% rule”—that is, you should increase your mileage by no more than 10% per week—is more conservative than necessary, and that novice runners may increase their mileage by 22% per week with no increased risk of injury.  This stat made it onto several Twitter accounts I follow, the front page of LetsRun.com, and garnered a brief write-up in Runner’s World.  Seeing all this, I did what I always do when a study comes out and makes some noise: I read the article. 

The abstract, which is freely available, is as follows:

There is a need to ascertain if an association exists between excessive progression in weekly volume and development of running related injuries. The purpose of this study was to investigate if GPS can be used to detect deleterious progression in weekly training volume among 60 novice runners included in a 10 week prospective study. All participants used GPS to quantify training volume while running. In case of injury participants attended a clinical examination. The thirteen runners sustaining injuries during follow-up, had a significantly higher weekly progression in total training volume in the week prior to the injury origin of 86 % [95 % CI: 12.9 to 159.9], p = 0.026 compared with other weeks. Although not significant, participants with injuries had an increase in weekly training volume of 31.6 % compared with a 22.1 % increase among the healthy. [...] Based on the results from current study, increases in weekly training progression may become deleterious at a weekly increase above 30 % which is more than the 10 % rule, currently used as guideline for correct progression in weekly volume by runners and coaches. Still, no clear evidence for safe progression of weekly volume exists. But is seems like some individuals may tolerate weekly progressions around 20 to 25 %, at least for a short period of time.
Now, I could whine about the usual problems—the small number of runners who became injured, the problems of applying studies on complete novices to experienced athletes, and so on.  But I’ll bite my tongue for now, because that’s not what wrecked my day today.  When I said “I read the article,” I did mean the whole thing.  Let’s take a closer look at the protocols of this study as outlined in the all-important methods section.

The study itself was mostly well-done: 60 novice runners from Denmark, none of which had any underlying health issues or injuries, and none of which had run further than 10km, were issued GPS watches and turned loose for 10 weeks of running training. 13 came down with a running injury during the course of the study, which was defined as “any musculoskeletal complaint of the lower extremity or back causing a restriction of running for at least one week,” which in my book is a fairly strict distinction.  The injured runners had been increasing their weekly mileage (or kilometerage, as this was a European study) at approximately 33% per week, while the healthy group had averaged 22%.  Notably, the injured runners had a statistically significantly higher BMI than the healthy runners—27.6 vs. 24.8 (a BMI of 25.0 is considered “overweight”; for a 5’9 person like me, that’s 170 lbs!).  The average weight of the subjects in this study should give us a few clues that we aren’t exactly dealing with world-beaters.  Unfortunately, because of the small sample size, the authors were not able to statistically extract the effects of weight and mileage on injury risk.

The most important detail, however, and the sentence that ruined my day, was this one:

The participants ran a total distance of 4556.5 km in 1172 training sessions during the 10 weeks

Doing the math, we get some troubling numbers.  1172 training sessions ÷ 60 participants ÷ 10 weeks = 1.95 training sessions per participants per week.  And the average weekly mileage? 7.6 kilometers.  That’s 4.75 miles per week.  Incoming freshmen at my high school were asked to run six times as much for the entire 10-week summer!  To picture things a bit more clearly, I created an example chart of the average weekly mileage of a healthy runner in this study (keeping in mind 1.95 training sessions per week):


Week
Miles per week
Miles per run
% increase
1
1.65
0.85

2
2.01
1.03
22.0%
3
2.46
1.26
22.0%
4
3.00
1.54
22.0%
5
3.66
1.87
22.0%
6
4.46
2.29
22.0%
7
5.44
2.79
22.0%
8
6.64
3.40
22.0%
9
8.10
4.15
22.0%
10
9.88
5.07
22.0%
Avg
4.73
2.42
22.0%


Now, I hope I don’t offend any readers of this blog, but even 10 miles a week is more jogger territory than “real running.”  And percent increases at such low mileage are almost meaningless—a one or two minute difference in the duration of a run has a substantial effect on the week’s mileage!

In all, I can only chide the authors of the study for not making it more clear how low the volume of running was in this study.  And for not including all the relevant information in the abstract (which is always a big misstep, in my opinion). Happily, there are some lessons to be learned from the study, too: being overweight is definitely a risk factor for injury.  More impact per footstrike is the likely culprit, but the general lack of athleticism associated with being overweight probably doesn’t help either.  Additionally, mileage increases in such a population should be constrained to around 20% per week.  However, I suspect it would be more helpful to discuss mileage increases at low volumes in terms of absolute distance, not percent.  For competitive runners in training, I have always liked “5mi per week or 10%, whichever is more.”  After all, as my college coach said, if you start from zero, ten percent more than zero is still zero.  Finally, GPS watches are a great research tool.  I would love to lead a study much like this, except giving GPS devices to a cadre of 50 or 100 collegiate cross country runners at the beginning of the summer.  That could be a treasure trove of data, given that GPS watches can collect information about not only distance, but speed too—a benefit not utilized by this study.  Garmin execs, are you reading this?

Anyhow, one more point I should make is that a 21.6% injury rate among runners averaging less than five miles a week is (philosophically speaking) absolutely unacceptable.  To me, that indicates that sedentary people have underlying biomechanical or strength defects that need to be addressed before or while they begin a fitness program.  I don't believe that everyone was "born to run" 100 mile weeks, but nearly everybody ought to be able to jog around the block a few times a week without getting injured.  
I hope that this paper spurs more research into training loads and injury rates, but I certainly hope it doesn’t encourage high school and college runners to jack up their mileage by 15 or 20 miles in a single week.  There is a big, big difference between moving from eight to ten miles a week and jumping from 65 to 80.
Lastly, I have some stern words for the folks who’ve already jumped on this “out-with-the-10%-rule” bandwagon: next time, please read more than the abstract.  Or at least talk to someone who has.  Perhaps LetsRun doesn’t have journal access (though any coach at a University ought to—just saying), but Runner’s World definitely does.  Or should.  Maybe more journals need to be open access.  Or maybe exercise physiology and biomechanics needs an ArXiv equivalent.  But even without, just because you’re out of school doesn’t mean you don’t have to do your homework...

6 comments:

  1. Those are minutes on that chart of mine from college, by the way, not miles. Nominal mileage assumes 7:30 pace, which is on the conservative side.

    ReplyDelete
  2. Hallelujah! Preach on brother. Great post. As an experienced runner and experienced researcher I can't stress the importance of truly 'digesting' some of these studies that keep coming out... whether they are on minimalist running, injury rates, etc etc.. read the methods! I am interested to read this full article and see what they talk about in their discussion.

    ReplyDelete
  3. Thank you for doing this. Too often an academic study's conclusions, based on a small amount of very constrained data, are taken to be more than they are in the popular press. And journalists wonder why academics get annoyed with them...

    ReplyDelete
    Replies
    1. Absolutely true... but on this occasion the academics seem to be to blame. To me, they seem to be courting publicity by means of a sensationalist abstract.

      Delete
  4. Nice review! This is my first time reading your blog, but I'm quite a fan, and will be back. It's nice to see someone break things down and actually look beyond the title/abstract.

    ReplyDelete