Is CrossFit Safe?

January 2021. John Casey MISCP MScSportsMed.

I get this question often in SportsPlus Physiotherapy. The answer is ‘not entirely’. But what form of aerobic exercise is completely safe? (Here’s where the pedantics among us take a minute)! So, the question that really needs to be asked is what is the injury risk with CrossFit and, following on from that, how does this risk of injury compare with that of other sports?

CrossFit, as a branded fitness regimen, is a relatively new movement being first incorporated by Greg Glassman and Lauren Jenai in 2000. It’s first gym was located in California and it rapidly spread around the world with 13,000 CrossFit ‘boxes’ in 120 countries by 2017. It is a strength, conditioning and overall fitness program consisting of a mix of aerobic exercise, body weight exercise, Olympic lifting and plyometrics. It also has an individual and group competitive element to it with exercises being timed to either see how many you can do in a certain time or how long it takes you to do a set number of repetitions of an exercise.

There have been injury concerns regarding CrossFit from its beginnings. Many of the movements involved are very technical. Injury can result from a mix of contributing factors including, inexperienced athletes, inadequate coaching and inappropriate levels of training intensity. Although CrossFit is relatively new there have been several research studies carried out on investigating its injury incidence. Unfortunately, some of these have been biased, some of poor quality and quite a few conflicting.

I recently read a very interesting study carried out in 2020 by Szeles et al in Brazil. Szeles and his colleagues were all orthopaedic surgeons and many of them were CrossFit enthusiasts themselves. They began to see higher injury rates in their clinical and exercise practice in people who trained in CrossFit than previous research had indicated.

They conducted a prospective 12-week study where participants had to complete a questionnaire online every 2 weeks. (A prospective study is one in which participants are enrolled before they develop an injury). They sent out questionnaires to 800 people who were training out of 13 CrossFit ‘boxes’ in an area of Brazil. 406 people met the inclusion criteria they had set out, including that they could not be injured at baseline. Crucially they defined an injury as ‘any injury or pain in joints, bones, ligaments, tendons, or muscles that prevented an athlete from exercising for at least 24 hours’.

Results showed that the average training exposure was 3.9 days per week. Each session lasted approximately 1 hour. 32% of the 406 people that completed the survey reported at least 1 injury over the 12 weeks. There were 247 injuries reported over the 12 weeks which was an exposure time of 13,041 CrossFit hours. This breaks down to 18.9 injuries per 1000 hours. This figure is important as it allows us to compare CrossFit injuries against other sports.

The trouble with comparing injury rates of different sports is that there is a wide range of definitions of injury being used in research. These vary from an incident that requires medical consultation to missing a training session or a match. The definition being mostly used is an injury is defined as a traumatic or non-traumatic incident that results in pain or reduced function causing a withdrawal from training for 24 hours or more. This is the definition that was used in the CrossFit study. Now we can compare this study with studies on other sports which used the same definition of injury.

Blake et al (2014) studied injuries in inter-county hurling and found the injury incidence rate to be 2.99 per 1000 training hours and 61.75 per 1000 match hours.

Murphy et al (2012) studied Gaelic football and found the injury incidence rate to be 4.05 per 1000 training hours and 61.86 per 1000 playing hours.

Hagglund et al (2013) looked at the injury incidence rate in professional soccer and found the figures to be 4.0 per 1000 training hours and 26.6 per 1000 playing hours.

Brooks et al (2005) looked at professional club rugby in England and studied injury over the 2002/03 and the 2003/04 seasons in the Premiership in players from 12 of the 13 clubs. They found an injury rate of 2 per 1000 training hours and 91 per 1000 playing hours.

Please note that the figures above do not say anything regarding the severity of the injuries sustained so they incorporate a wide range of injury ranging from 1 day off sport to several months or career-ending injuries.

What was interesting in this CrossFit study was that a 1-year increase in CrossFit experience reduced the odds of sustaining an injury by approximately half. One could assume that, with experience, managing load improves as does technique which helps reduce injury.

Limitations of this study are that it was only over 12-weeks and that it was specific to a region of Brazil. Further study is required into injury incidence in CrossFit, as it is in all other sports. This allows for the development of methods to reduce injury targeting especially risky practices and training techniques.

From my own point of view, other factors also need to be considered when discussing injuries and CrossFit. One is that people usually arrive to CrossFit in their late 20s or 30s (or older) and, while they may have participated in other sports previously, CrossFit is new to them and new to their bodies. If they do come to CrossFit from another sport, they are usually competitive people and competition, whether it be against others, against yourself or against the clock, is a risky way to learn new skills and techniques from an injury point of view. Quality coaching, training at an appropriate level and having adequate rest and recovery between sessions are vital to minimizing injury in any sport, including CrossFit.

So, to summarise. What is the injury risk associated with CrossFit? It varies greatly from individual to individual, but this study indicates that the general risk of injury from CrossFit training is higher than the injury risk associated with inter-county hurling training sessions, inter-county Gaelic football training sessions, professional soccer training sessions, and rugby training sessions but lower than playing matches in these codes.

References.

Szeles et al (2020) CrossFit and the Epidemiology of Musculoskeletal Injuries; A prospective 12-week cohort study, The Orthopaedic Journal of Sports Medicine, 8 (3), 1-9.

Blake et al (2014) Epidemiology of Injuries in Hurling: A Prospective Study 2007-2011, BMJ Open,  4, e005059.

Murphy et al (2012) Incidence if Injury in Gaelic Football: A 4-year prospective study, The American Journal of Sports Medicine, 40, 9, 2113-2120.

Hagglund et al (2013) Injuries Affect Team Performance Negatively in Professional Football: an 11-year follow-up of the UEFA Champions League injury study, British Journal of Sports Medicine, 47, 738-742.

Brooks et al (2005) Epidemiology of Injuries in English Professional Rugby Union: part 1 match injuries, British Journal of Sports Medicine, 39, 757-766.