If you have eyes, you’ve noticed journalists and struggling trainers speculating about the dangers of CrossFit. If you have a brain, you’ve noticed the hysteria fails to meet basic logical standards. Meaningful claims about risk must meet at least the following three points, to start:
It is not enough to look at a workout program and assume it’s dangerous. One would hope this was obvious, but some American College of Sports Medicine fellows keep falling short. They have repeatedly published bad science on the basis of non-existent or misinterpreted evidence. Sources: http://journal.crossfit.com/2012/09/…sensus-pap.tpl and http://journal.crossfit.com/2013/05/acsm.tpl. And the bad science doesn’t end with papers about CrossFit. More on this, later.
Contrary to what Patrick Mccarty has claimed, it is possible to gather data on CrossFit, both online and in gyms. Data need not encompass the entire CrossFit community to count.
2. Time frame:
With a long enough time frame, the fatality rate for all activities is 100%. An injury rate without a time frame is meaningless.A 74% injury rate over one workout is very different from a 74% rate over years of training.
Outside Magazine is a first-class source for bad reporting on fitness. They made this mistake recently:
“Studies have pegged the CrossFit injury rate from as low as 16 percent to as high as 74 percent.” (The 16 percent figure has never been substantiated, but the Outside reporter failed to seriously investigate that fact.)
Furthermore, the 16 percent figure comes from a 6-week study, whereas the 74 percent figure comes from a study where the average CrossFit experience was 18.6 months. In other words, the 74 percent figure came from a study with a time frame over 13 times as long as the other study. It’s not a fair comparison.
One way to address time is find the number of injuries per 1000 hours of participation. This is the incidence rate. Dr. Yuri Feito explains the concept here.
At least 52 Americans have died competing in triathlons since 2007. (I trust that Eric Robertson has already started writing “USA Triathlon’s Dirty Little Secret). Are triathlons high-risk? To answer this question, you would have to find an incidence rate, as discussed above. Next, you would need to compare it with other sports and fitness training.
As Dr. Lon Kilgore has said:
“Even if a trainer takes care of every screening detail, exquisitely explains the benefits and risks of training within his system, and obtains an assumption of risk and a waiver of liability from a trainee, and even if every cleanliness, equipment- safety, exercise-technique and supervisory standard is met, someone will get hurt. The statistics tell us that. It is a certainty. We just can’t say who or when.” (source)
This is not just true of CrossFit – it is true of all human movement. And still, we dare to leave the couch. Consider a study published in the Journal of Strength and Conditioning Research entitled “The nature and prevalence of injury during CrossFit training.”
Their online survey found an incidence rate of 3.1 injuries per 1000 hours of CrossFit. Jeff Barnett capably addressed the study’s possible flaws and biases here, but the study is still notable. They are the first researchers in all of exercise science to publish a study on CrossFit that met the three standards listed here.
The researchers note that the rate is “similar to those seen in other high intensity and technically demanding sports such as Olympic weight lifting and power lifting as well as general fitness training.”
The “general fitness training” source the researchers cite found a rate of 5.92 injuries per 1000 hours of training. That’s nearly twice what the researchers found for CrossFit.
Here are some more studies for comparison:
Running (multiple studies have found an injury rate 10X what this study found for CrossFit): http://www.ncbi.nlm.nih.gov/pubmed/8536050
The headlines are right; they just got the victims wrong. CrossFit is dangerous – for less effective training programs (and much of the food industry).
(Originally posted on the CrossFit Message Board here)