A very recent research-article (May 1st) on aafp about the FM/a-test is unfortunately behind a paywall (40$ for the issue) apart from the beginning where it says
"However, this type of study design, which uses a control group of healthy patients, overestimates accuracy. In the same study, when using a more appropriate comparison group of patients with known rheumatoid arthritis (RA) or systemic lupus erythematosus (SLE) who did not have co-occurring fibromyalgia, the specificity was only 70%, with a positive likelihood ratio of 3.1 and a negative likelihood ratio of 0.1. The most appropriate study design would enroll patients with clinically suspected fibromyalgia, but this type of study has not been performed."