Great thoughts and take your time, the upcoming content sounds exciting to get into.
I’m glad you elaborated on the Meehl article about statistical prediction being superior to clinical judgement. It’s convincing that this occurs so consistently across diverse conditions. I read it a few weeks ago and really forgot about it and then read it again. There were 10 examples he used in the 1993 publication (predicting MI, academic success, police termination, etc.) and I’m sure there has been a lot more published since. I think that physical therapy has been trying to capitalize on this kind of thinking given the number of clinical prediction rules that are out there however a lot of them have just been in the derivation stage and I’ve been skeptical to apply them. I really trust the diagnostic CPRs but the ones that forecast a response to a certain intervention make me dubious. The past few months has made it easier to appraise how useful they can be.
Here are some examples that I thought about …
Using ROM to predict full recovery following rotator cuff repair
Tonotsuka H, Sugaya H, Takahashi N, Kawai N, Sugiyama H, Marumo K. Target range of motion at 3 months after arthroscopic rotator cuff repair and its effect on the final outcome. Journal of Orthopaedic Surgery. 2017;25(3)
Researchers identified that having at least 120 degrees of forward flexion and at least 20 degrees of ER was predictive of full recovery as defined by not having a re-tear. I remember reading this and thinking that 120 degrees at 13 weeks is pretty limited but patients followed for 2 years that obtained this range at the 3-month mark did well.
Return to work following a back injury
Wynne-Jones G, Cowen J, Jordan JL, Uthman O, Main CJ, Glozier N, van der Windt D. Absence from work and return to work in people with back pain: a systematic review and meta-analysis. Occup Environ Med. 2014 Jun;71(6):448-56
I couldn’t find the original publication but I’ve repeatedly read the statistics on how the duration of missed work following a back injury is associated with a higher likelihood of delayed return to work. If an individual doesn’t return to work after two years, there is a very high chance of them never working again.
Return to sport battery (quadriceps index, hop test scores, knee specific outcome tests)
Plenty of articles here in prediction of graft failure or second injury after ACLr.
I think there are some areas in PT practice where statistical prediction is excellent and the data to draw from is rich and there are of course other areas where we have more uncertainty. It reminds me of Schön’s description of “technical rationality” vs. “muddling” in professional practice. The first part of his book the Reflective Practitioner is fantastic where he explores these two processes. He describes technical rationality as how professionals apply established theories, research and techniques to solve well defined problems. He uses the analogy of practicing on the high, hard ground. Treating BPPV comes to mind here. He then describes a concept of muddling through difficult concepts where professionals navigate messier, ill-defined problems. He uses the analogy of practicing on a muddy swamp. Managing nonspecific LBP comes to mind here.
I drew a picture to describe this but I can't upload it into Substack comments, I'll try another way if anyone is interested :)
Great thoughts and take your time, the upcoming content sounds exciting to get into.
I’m glad you elaborated on the Meehl article about statistical prediction being superior to clinical judgement. It’s convincing that this occurs so consistently across diverse conditions. I read it a few weeks ago and really forgot about it and then read it again. There were 10 examples he used in the 1993 publication (predicting MI, academic success, police termination, etc.) and I’m sure there has been a lot more published since. I think that physical therapy has been trying to capitalize on this kind of thinking given the number of clinical prediction rules that are out there however a lot of them have just been in the derivation stage and I’ve been skeptical to apply them. I really trust the diagnostic CPRs but the ones that forecast a response to a certain intervention make me dubious. The past few months has made it easier to appraise how useful they can be.
Here are some examples that I thought about …
Using ROM to predict full recovery following rotator cuff repair
Tonotsuka H, Sugaya H, Takahashi N, Kawai N, Sugiyama H, Marumo K. Target range of motion at 3 months after arthroscopic rotator cuff repair and its effect on the final outcome. Journal of Orthopaedic Surgery. 2017;25(3)
Researchers identified that having at least 120 degrees of forward flexion and at least 20 degrees of ER was predictive of full recovery as defined by not having a re-tear. I remember reading this and thinking that 120 degrees at 13 weeks is pretty limited but patients followed for 2 years that obtained this range at the 3-month mark did well.
Return to work following a back injury
Wynne-Jones G, Cowen J, Jordan JL, Uthman O, Main CJ, Glozier N, van der Windt D. Absence from work and return to work in people with back pain: a systematic review and meta-analysis. Occup Environ Med. 2014 Jun;71(6):448-56
I couldn’t find the original publication but I’ve repeatedly read the statistics on how the duration of missed work following a back injury is associated with a higher likelihood of delayed return to work. If an individual doesn’t return to work after two years, there is a very high chance of them never working again.
Return to sport battery (quadriceps index, hop test scores, knee specific outcome tests)
Plenty of articles here in prediction of graft failure or second injury after ACLr.
I think there are some areas in PT practice where statistical prediction is excellent and the data to draw from is rich and there are of course other areas where we have more uncertainty. It reminds me of Schön’s description of “technical rationality” vs. “muddling” in professional practice. The first part of his book the Reflective Practitioner is fantastic where he explores these two processes. He describes technical rationality as how professionals apply established theories, research and techniques to solve well defined problems. He uses the analogy of practicing on the high, hard ground. Treating BPPV comes to mind here. He then describes a concept of muddling through difficult concepts where professionals navigate messier, ill-defined problems. He uses the analogy of practicing on a muddy swamp. Managing nonspecific LBP comes to mind here.
I drew a picture to describe this but I can't upload it into Substack comments, I'll try another way if anyone is interested :)
If you want to write a post - I can invite you as an author (I think) and then you can embed the picture in the post)
Sure, try it. The picture has flies and stink lines on it.