Agreed... it requires more than one test to support a differential diagnosis. I have always gone by the basic rule to look at at least 2 of 3 selected tests being positive.
using single test to rule in or out is weak. we are seeing more and more clinical prediction rules and it is probably more appropriate to do the same here instead of relying on a Neer or Hawkins.
http://www.ncbi.nlm.nih.gov/pubmed/19887215 serves as nice guide.