Background/Purpose:  
The American College of Rheumatology published its
Top 5 List of Things Physicians and Patients Should Question as part of the
American Board of Internal Medicine’s Choosing Wisely campaign.  First on the
list was “Don’t test ANA subserologies without a positive ANA and clinical
suspicion of immune-mediated disease. We instituted a system to promote more
cost effective subserology ordering.

Methods:  
We identified 101 of 1100 providers who were over utilizers (more than 2
subserologies ordered when ANA negative in a 26 month period). These providers
received an email informing them of their ordering practice with explanation of
ANA reflex testing, and a link to an online education module. Providers were
surveyed pre and post intervention on their preferred quality improvement method.
We prospectively examined for 1 year subsequent instances of unnecessary
subserology ordering defined as anti-dsDNA, anti-SM/RNP, and SCL70 ordered when
simultaneous ANA was negative. This was compared with the prior 26 months.  If
SLE was suspected, the ANA would reflex if positive to anti-ds DNA, anti-cardiolipin
ab, and anti-Smith/RNP ab.  For Sjogren’s, the positive ANA would reflex to the
SSA/SSB ab. 29/101 providers completed the non-required online education. Generalized estimating equations
models were used to obtain estimates of odds ratios (OR) with their 95%
confidence intervals (CI).  A cluster effect by provider was included in the
model to take into account the correlation within each provider.  Finally, a cost analysis based on the Medicare fee
schedule was performed. This analysis assumed that when reflex testing was ordered, unnecessary
subserologies were avoided.

Results: Unnecessary subserology ordering across all system
providers decreased significantly after the intervention (OR 0.625, CI 0.425, 0.918,
P-value 0.0167). Over the 26 month pre intervention
period, the mean percentage of unnecessary subserology ordering episodes was
44.5% when the ANA was negative.  In the other 55.6%, the SSA/SSB or anti Jo1
antibodies were ordered which can be appropriate when the ANA is negative. In the
10 month post intervention period, the number of unnecessary instances
decreased to 35.1% of the total.    Surveyed providers preferred reflex testing
and education over other practice improvement strategies. Of the 101 providers
who over utilized subserologies, 26 started using ANA reflex testing.  Cost
analysis showed that ANA reflex testing potentially saved $8,462.50 by reducing
unnecessary subserology utilization.

Conclusion:  The introduction of optional ANA
reflex testing with optional online education led to significant decreases in unnecessary
subserology ordering when the ANA was negative across the entire integrated
health system.  Although, the calculated potential cost savings was not large,
we only examined the instance of simultaneous ANA and subserology ordering when
the ANA was negative, and not the entire spectrum of ordering practices. ANA
reflex testing potentially saves providers time and patients inconvenience for
2^nd phlebotomies. We expect that ANA reflex testing will be adopted
by additional providers and are developing a scleroderma reflex test.