Okay, I can’t take credit for this one entirely. Actually at all. I got into the Bayesian thing from The Shoulder Physio’s Instagram. But hey, isn’t that just all of academia? Borrowing ideas and messing with them? Anyway, let’s take another reductionist look at complicated topics for the fun of learning! Yay!
So, Bayesian statistics. Yeah, I said it: STATS.
I’ve read some stuff & listened to some other stuff. But I really liked this broad explanation that Wikipedia has:
is a theory in the field of statistics based on the Bayesian interpretation of probability where probability expresses a degree of belief in an event. The degree of belief may be based on prior knowledge about the event, such as the results of previous experiments, or on personal beliefs about the event.
Let that sink in a bit. The probability of the coin landing heads-up or tails-up may depend upon your belief or prior experience…
What I am really talking about is less like Bayesian Statistics and more like Bayesian Epistemology. I.e., the way we know truth is likely relative to our degree of belief or prior experience. But I include the statistics part because our brain is a pink, mushy Bayesian Statistics computer. The human brain is constantly running the odds based on our prior knowledge/experience/belief.
Pause! Rewind! The human brain is constantly running the odds based on our prior knowledge/experience/belief.
This is one of the keys to working with people in pain. Your brain is running a statistics computation on how much pain it expects based on its prior knowledge/experience/beliefs. And for the movements it believes will be painful, it places bets. Those bets are sometimes felt as pain to prevent you from potentially injuring yourself. Isn’t that NUTS?!? It also makes a lot of sense for survival…like don’t grab that red, glowing piece of wood from the fire or you’ll get burned like last time…
Peter O’Sullivan & JP Caneiro wrote in a paper that clinicians should use movement experiences to violate patient expectancy when it comes to pain & then draw their attention to it. Or in the context of the Trog Blog series:
Allow emergent human movement, in a dynamical system, to change the (Bayesian) statistical prediction model your patient is currently using to solve movement problems. And then draw their attention to it–this is key–to change the model.
This concept was one of those mind-bending moments that left me permanently deformed for the better. People’s central nervous systems are just placing bets based on the last game of “when will my back hurt”. And it makes sense that people who suffer from chronic pain tend to not improve much with any intervention. Their statistical models are running on 30 years of painful data. The horses are doped, the race is fixed, and the bookies are crooked.
The clinical pearl?
See how people solve a movement problem that is historically painful. Recognize that you’ll have to alter it enough that it changes the Bayesian Statistical modeling but not so much that their brain chooses a different model entirely. Theoretically, with enough repetitions, the statistical model can be re-written. Make sure to draw their attention to it if they haven’t noticed, because their conscious perception is the key to changing the model. Give them the principles & repetitions that they can practice independently. And then let your patient’s explore movement for themselves. You’ll be amazed when they come back for their discharge!
Get buckets, get W’s, and get championships people…and don’t live in caves…
Comments