Special Tests

I'm not going to be able to reference this very well because the journal is in long term parking at the airport.  The recent JOSPT article regarding manipulation of the cervical spine that many of you were involved in writing has me thinking in a more broader, bigger picture way. 

How is it that we as a profession have so many special tests for various body parts or diagnoses but it actually seems that a lot of them are worthless?  At the same time, clinical decisions are being made every day based upon many of these special tests that may not be valid and reliable, may not have high specificity or sensitivity qualities.  When will it be that as clinicians we will begin to have a test or a set of tests that provide consistent results to help make better, more accurate clinical decisions?  Maybe it's just me, but it seems that if we have a shaky foundation that is the core of the beginnings of our clinical decision making process in which we build... then what happens to the treatment aspect of what we do when it is based on something that isn't solid and strong?  Just some thoughts in my head that maybe some of you wiser therapists would happen to know the answer.

Comments

[Jason Silvernail]

I think the use of special tests varies quite a bit from PT to PT.
Keep in mind that many of these are from the early Orthopedic literature, and we share them with other providers, they are not unique to PT.

Certainly some special tests have some good evidence supporting their predictive ability (eg Obrien's active compression in the shoulder) and others do not.

In our clinic, we do make an effort to show preference for those tests which have some established validity or predictive ability, and to practice accordingly.

I have found (as many others have as well) that through time, we narrow the number and types of tests we do with patients, by way of trying to match patient to treatment.
Much of the classification work lately has pointed up the fact that we really need a much smaller regimen of evaluative measures in order to help match a given patient to a given intervention.

I am unsure of the number of therapists out there still doing 12 special tests on every shoulder that comes into the office, but I would venture to guess that it is very small. Just as I know the overutilization of modalities is slowly decreasing.
Education and evidence are slowly changing practice, though not as quickly as they could...or should. :)
J

[Ben Hando]

Selena/Jason- I think you bring up an important subject.

I think Jason's example of O'Brien's test is a great example of why you need to stay current in the lit. even with commonly used tests. His (O'Brien's) original article had great results, and it seemed like the end all be all for AC dysfunction and SLAP tears. However, in subsequent studies, the specificity values weren't so hot (.42). Certainly still a valuable test though for screening purposes.

I was at a course this past week where we talked about the fact that the boom in EBP has largely involved treatment, and in the future the focus will shift or expand to include diagnostics more than it does today.

A side note: I have a good cheat sheet for shoulder Dx tests with sens./specificity and LR's if anyone would like it, you can shoot me an email. I didn't assemble it so I don't feel comfortable posting to the blog.

Ben Hando

[John Groves]

Selena

I am not the "wiser" PT that you were hoping to hear from and I am sure John Childs and others will provide a more in depth answer than I am about to attmept...

Your frustration with special tests is warranted as we all know. The posts to this point have used the shoulder as an example, but think of the lumbar spine for a second. Which tests should we use? The Clinical Prediction Rule for spinal manip derived by Flynn and validated by Childs gives a different look at how important special tests are for deciding which side to treat. While this CPR doesn't answer all the questions (i.e shoulder tests), for back pain it gives us the idea not to worry to much about the special test, just move the SIJ (or whatever actaully moves) and move on. Doesn't matter. The prediction rule becomes the special test. Sound evidence based foundation with that approach.