Using 4+ to grade near-normal muscle strength does not improve agreement

Contrary to the original hypothesis, introduction of a '4+' grade did not clearly improve agreement or variability of ratings, despite eliminating the physical muscle testing by providing written descriptions of test findings and specifically designing these to suggest a weakness of grade '4+'.