I have this question that I have been pondering. It's about shoulder internal rotation and range of motion. First of all, we've got two ways of assessing it, right? Abducted shoulder internal rotation and back scratch test.
With the back scratch test I've not yet seen a clear classification of what arm position corresponds to what range of motion. General classification is : if your arm goes horizontal - that's a good start; if you can't get horizontal - bad; if you can go above horizontal - you have good flexibility. Always look for the inferior border of the scapulae and scapular winging - the easiest way to cheat the test. Other ways are shoulder abduction and thoracic lateral bend.
That one's fine with me, if you say that it's a good test.
The one that's bothering me is the abducted arm internal rotation test.
Here's the thing - I can rarely get anyone to go into internal rotation without scapular elevation and/or anterior tilting.
What Eric Cressey is showing in his pictures usually corresponds with my observation - on most of the pictures he post, he allows people to hike the shoulder up and get more ROM. Now don't get me wrong - he's a really smart guy, but is that ok with you?
Best of wishes.
A: In general, we've got definite ROMs that we would use for shoulder internal/external rotation that we consider to be "accurate" (90 & 70 degrees, respectively, by goniometric measure according to the American Academy of Orthopeadic Surgeons). Personally, I prefer the supine position (the subject remains on his/her back with the arm brought out 90 degrees), but any position is fine relative to patient needs (from a clinical standpoint, there may be a reason that the patient can't get into the position that you need them to, so you have to use different strategies). That can include the arm held at the side or with the shoulder abducted to 90 degrees, etc. The scratch test is fine too, but it's more of a gross movement test vs. a single, isolated movement test and you won't get a "pure" number corresponding to just one movement. In the gym, this is fine. In the clinic, this may or may not be ok...it just depends on the case.
If we're interested in the technical word of law, then no: you shouldn't allow any additional movements such as scapular tilting (or any other compensation) when taking the ROM of the shoulder. However, it probably isn't that important in a non-pathological shoulder, which is what you're likely to see in a gym client.
In terms of "real world" function, observing compensations by allowing gross movement can be extremely valuable, so I understand why Eric would say that. I generally do the same thing for a non-pathological client, too. I would just add that if there's a difference, you probably need to break down the movement into it's constituents (IR, extension, elbow flexion, etc) and see what's really going on.
Which leads to the last point: how do you measure things in the "real world" outside of goniometers and the AAOS? Compare both sides: it's more important that a client is symmetrical then if they have the appropriate ROM in most (if not all) cases. Problems will occur if there is a difference between limbs, and that's what you should be looking for.
Yes, if a client presents with bilaterally short hip flexors, for instance, you would probably look to address that through various stretching and mobility work, but if you have the case of one hip flexor being significantly stiffer or shorter than the other, that's where you will see non-symmetric, unilateral alterations to the kinetic chain, which can cause major problems over time. That's like driving with one flat tire: eventually, your car's alignment will also need fixing, too.