Can we predict yoga injuries?
In part one of our discussion of the context for yoga injuries, we shared all of the different individual contexts and situations where respondents from our survey project reported a yoga injury. But when yoga practice happens “in the wild”, many factors happen together. And they could all potentially contribute in a greater or lesser degree to injury. So, what happens when we look at all the possible contextual situations together? In this post, you’ll find out. We discuss the probability of experiencing a yoga injury depending the individual factors that make up the context of yoga practice and how they interact with one another.
When we first thought about doing this survey, we had many hypotheses about what causes injury. In our minds, there were so many potential factors that could influence the likelihood of an injury. We were curious if time of year, time of day, temperature of the room influenced injuries. And what about being a beginner, an advanced practitioner, or receiving adjustments? What we found out surprised us!
In this post we talk specifically about what’s driving yoga injuries. And “injury” means having at least one injury during your lifetime of yoga practice. We are not looking at what drives specific types of injuries. And we are not looking at severity of injuries reported. This is self-reported data. Injuries that required medical attention are in the same category as small tweaks that disappeared with no medical intervention in a few weeks or less.
Methods and a little more about statistics
We know that our practice happens under multiple contextual situations all at the same time. We don’t practice in a laboratory where we separate, say, temperature of the practice room from consistency of practice. For that reason, in this post we report the results of a test called multivariable logistic regression which tells us which factors, of all the factors that are occurring in practice simultaneously, are contributing to experiencing an injury. A resulting “p-value” of less than 0.05 indicates that there is a statistically significant relationship. Remember that a p-value of 0.05 is the same thing as saying there is a 95% probability that two variables are related.
We used the results of our logistic regression analysis to calculate “predicted probabilities”. This statistical method describes the probability of reporting an injury in specific contexts based on the results of our logistic regression model.
For detailed information on our methods:
Strongest relationship with injury was time spent on the mat
The primary driver for likelihood of reporting at least one injury over a lifetime of yoga practice was simply more time on the mat. The strongest driver for reporting an injury in yoga was years of practice (p<.0001). This was true regardless of what other contextual factors for practice we included in the model with years of practice. Similarly, though less strongly related, those who teach yoga also had a slightly higher likelihood of reporting an injury. You could look at teaching yoga as just another way of spending more time on the mat. This supports our finding that amount of time on the mat is the biggest influence on chances of reporting an injury in yoga.
Predicted probabilities of yoga injuries
The probability of reporting an injury did not change when we included contextual factors of practice other than years of practice and teaching yoga in our model.
When a model was evaluated with injury yes/no as our independent variable and all potential factors related to practice context included as dependent variables, the following did NOT significantly influence the likelihood of reporting an injury (either positively or negatively):
- Number of days that you practice with a teacher each week
- Time of day that you typically practice
- Duration (number of minutes) of your typical practice
- Number of teachers you’ve practiced with in the last 30 days
- The number of years that your teacher has been practicing
- Number of days per week that you typically practice without a teacher
- Whether you have ever received hands-on adjustments in a yoga class
- Your gender
- Your age
- Whether you currently have a chronic health condition that impacts daily life and/or causes pain
- If you have had chronic physical pain for any reason prior to beginning a yoga practice
- Whether your teacher was registered with Yoga Alliance at the 200 or 500 hour level
Conclusion
Our study results remind us that yoga asana is a physical activity. Any physical activity, regardless of how carefully we do it, has some risk of injury. We have shared in previous posts some of the contexts in which specific injuries, for example knee injuries or shoulder injuries, were more likely to occur. Our study does point to aspects of yoga practice that we could modify to potentially reduce specific kinds of injuries. However, if we are speaking of injury in general, our study suggests that if we spend a lot of time at a physical activity, like yoga, our chances of having an off-day, taking a misstep, or having an accident increase. While this understanding doesn’t need to keep us away from yoga, it could change our approach to yoga and our expectations of what yoga can do.