CAP PT vs Quality Cross Checks

This seems very extreme to me. For PT and for quality cross-checks.
I read the directions. For the departments that I know have things handled- I often just highlight the important instructions. 
For departments with enough staff that almost no one ever runs one, I talk them through anything that’s different than running a patient (special mixing directions, different adapters, etc.)

That seems like a rather adversarial stance. We know the goal of the PT is to treat it as closely to analyzing a patient sample as possible. Running our PT samples is often NOTHING like running a patient sample. We're running Prime Plus blood gas analyzers here. The PT directions include "shaking vigorously".  Clearly that is not how someone would treat a patient specimen. 
I can't even get most non-Laboratory personnel to understand what a Proficiency Test is in the first place, so I always help getting into the proper mode, etc.. I don't think "letting them just figure it out" is the spirit of what the CAP intended.
 

We treat the quality cross-checks the same way we do the PT, with the exception of course that the cross-check gets the same samples repeated multiple times.

I instruct my facilities to manage the prep-work of the Survey samples, ensuring they are properly prepared prior to giving them to the staff on the floors for testing.  I also instruct them to NOT guide them on how to run the survey sample (outside of potentially reviewing proficiency mode).  If they do it wrong... it fails, but you know **why** it failed.

Handing the kit to the floors and walking away is asking for trouble.  Staff on the floors are primarily focused on patient care and not running the surveys.  I've had facilities do exactly this (give the kit to the floors and walk-away) and the survey fails.  Now you're stuck... what happened? did they hydrate the QC incorrectly, did they run the sample cold or over-heat it?  Did they read the UA dipstick immediately after dipping it?  Or perhaps there is an actual test system issue.   If you're not there watching (and not guiding them) you will not know these issues, and it makes troubleshooting much more difficult.  Preparing survey samples is NOT part of normal patient testing.  The intent of these surveys is to detect issues in the test itself and the user's ability to run the test correctly.

What your lab director advised was extreme, and is basically a setup for failure for both the operator and the laboratory, as many on this thread have stated.  One of the (unfortunate) tenets of POCT is that performing the testing is not the clinical staff's primary duty, nor do they typically possess a laboratorian's analytical science background where assumptions can be made how to handle a sample without being told.  While I understand what your lab director was "after" by instructing you to let them sink or swim, it is really not applicable outside the laboratory.  Those of us in the POC field know this all too well.

I think it is a good idea to treat cross-check samples like PT. It is important to note that PT (and cross-check) samples are NOT anything like patient samples. Though CMS requires that operators treat the samples like patient tests, that is sometimes impossible. We do send revised instructions to the units when we send the PT, and we have written in our QA policy, where we discuss PT, that when the samples or process is unlike the normal patient test process, we can give them some assistance. Sending PT or cross-check with no assistance is just asking for trouble, and potentially failed QC and cease testing orders!