It is a good point that psychometrics are usually based on reference to “gold standard” diagnostics, but that even these are not great for specific identification of structures that are driving pain- leaving us with even more unknown about the true validity of special testing.

For the shoulder I don’t think I can say that special tests alone help me to feel confident in completely ruling in or out. I have found that having one specific structure at fault likely isn’t the case with a lot of shoulder pain b/c of the proximity of so many structures and likelihood of one specific movement to stress multiple things.

I feel like my pattern recognition is based a lot more on specific subjective complaints, functional assessment, ROM, and resisted testing with less of an emphasis on special testing. To be honest at times I feel like I am just doing it because that is what we learned but don’t put a whole lot of stock into the findings- something to evaluate about my practice and improve clinical reasoning on to determine if and how these tests can be used better.

I haven’t in the past used special testing as an objective asterisk and/or to identify and reassess irritability; is this something that y’all do often?

Steve- Did this article change your thoughts on utility of shoulder special tests in clinical practice? Is there any part of the exam you feel like you weigh the findings more to help with diagnosis and pattern recognition?

I actually use special tests as objective *’s frequently. While Hawkins-kennedy is a “junk” test, it is useful when it recreates the pt’s symptoms. If we then improve the HK, then I feel more confident that we treated the appropriate tissues to have less symptoms with other tasks. Per what we talked about on zoom, I look at a lot of different objective *’s from different systems (something active, something resistant, something ROM, something that’s a special test) when it’s time for reassessment. This drives what interventions I proceed with next when I see what does/doesn’t improve. This, of course, fits well into a Maitland/pain provocation model where identifying the structure at fault can sometimes be less relevant…

…however, I think there is benefit from being as specific as possible and I think special tests do add some relevance. The author is being a bit ridiculous. Yes, multiple muscles probably contract during a full can, however how many of those muscles are clinically relevant? What do we know from biomechanics/anatomy to identify the most probable tissue at fault? Does our APR and palpation reflect that assumption?

I agree that this article move special tests even further down the list. I also loved that the author made a point that I had thought to myself before: If pathological changes are present on imaging of non-painful tissue, how do we truly identify how valid a test is?

In terms of abandoning tests all together, I think that is a mistake. Some tests provide valuable information such as specific positions of discomfort, an approximation of tissue irritability, etc. I think the difference is what YOU think the test is telling you. They aren’t telling you a specific pathology is present, but perhaps they give you a better clinical picture of movement and patient experience.

First of all, this author is definitely on a soapbox…

Anna/Steve – totally agree, I usually do the tests but don’t put a whole ton of stake in them

Anna/Barrett – also agree to the point about identifying a specific structure at fault. In some body regions I think this is very appropriate but for the shoulder, with as interconnected as everything is, I’m not sure it’s appropriate to attempt to identify one specific structure at fault. I find myself spending SO MUCH TIME trying to identify a structure, that I often wonder if I had just started treating the impairments if they would be getting better faster…

All in all, I think special tests DO have a place in a shoulder evaluation. It’s a cluster just like everything else in PT is, and within the cluster there is a hierarchy of information. Special tests are lower on that totem pole for me than some other subjective/objective info as Taylor mentioned. I think I tend to use special test results broadly rather than using them to point to a specific tissue at fault. For example, I may think “this shoulder does not like compression” with the crank test or “this shoulder does not like internal rotation” with Hawkins Kennedy.

Lastly, to the author’s point about imaging: isn’t it true for most body regions that imaging findings don’t correlate to pain? I think this speaks to a more comprehensive approach to an evaluation, taking into consideration pain neuroscience and other psychosocial factors. Again, just more data points for the cluster…