Background/Purpose:   As part of the American Board of Internal Medicine’s Choosing Wisely Campaign®, the ACR published recommendations in 2013 advising clinicians against ordering anti-nuclear antibody (ANA) subserologies in the absence of a positive ANA and a clinical suspicion for an underlying immune mediated disease.  Subserology testing for SSA and Jo-1 was considered a reasonable exception this rule.  Previously published guidelines also recommend against obtaining serial ANAs.   In this study, we sought to assess the current practice patterns for ordering ANAs and ANA subserologies at an urban community hospital and its affiliated outpatient clinics.

Methods:   Laboratory data was compiled for all patients who underwent testing for an ANA or any of several ANA subserologies (dsDNA, Smith, RNP, SSA, SSB, Scl-70, Centromere, and Jo-1) at the San Francisco General Hospital clinical laboratory between January 2012 and January 2013.  All ANAs were measured by immunofluorescence, and all subserologies were measured by enzyme linked immunosorbant assay.  Data were analyzed to generate descriptive information regarding the frequency and origin of ANA and ANA subserology ordering

Results: A total of 2514 patients underwent testing for either an ANA or an ANA subserology during the study period.  A total of 2350 patients underwent initial ANA testing of whom 123 (5.2%) underwent serial ANA evaluations.  Repeated ANA tests were ordered either from primary care clinics or on inpatient services and not from subspecialty clinics (Table 1).  A positive initial ANA (defined as a titer >1:40) was found in 524 (23%) patients, and a negative initial ANA was found in 1810 (77%) patients.  Of those patients whose initial ANA was negative, 117 (6%) went on to have a subserology checked.  From this group, 58 (50%) patients were tested for SSA and 12 (10%) were tested for Jo-1.  ANA subserologies were therefore ordered inappropriately in 47 (2%) patients whose initial ANA was negative.  Over a one year period, 170 (7%) patients with an initial ANA test underwent subsequent ANA or ANA subserology testing that was counter to current recommendations. 

Conclusion:   Inappropriate ANA and ANA subserology testing were relatively infrequent in our safety net health system, which provides care to poor and medically indigent patients.  Serial ANA testing was three times more common than inappropriate subserology testing.  Measures aimed at reducing repeated ANAs ordered from primary care clinics and on inpatient services would likely be most efficacious in reducing unnecessary testing.