Optimizing Test Ordering Language to Minimize Group A Streptococcus Reflex Culture for Adults

BACKGROUND: Group A Streptococcus (GAS) is a predominant bacterial cause of acute upper respiratory tract infections, with the greatest incidence of pharyngitis and possible development of acute rheumatic fever occurring in children. According to guidelines from the Infectious Diseases Society of America, negative results from rapid antigen detection testing (RADT) for GAS should be followed with reflex to pharyngeal culture for children but not generally for adults. At our institution, several departments were found to routinely order reflex culture in adults (>17 yo). 86% of RADT-negative reflex cultures were negative for GAS, as well as Groups C/G, which supports the notion that these backup cultures are unnecessary for adults. METHODS: In November 2016, a change in ordering language was implemented in the emergency department (ED), which was found to have the highest number of reflex culture orders for adults. To differentiate the two testing routes for children and adults, the word “peds” was added for RADT with reflex culture orders, and the word “adult” was added for RADT without reflex culture. At the commencement of the intervention, a brief education on the change in ordering language was provided to physicians by one of the ED providers. From November 2016 to April 2017, the number of GAS reflex culture orders for adult patients in the ED was tracked. These were compared with data from the 1-year period prior. RESULTS: Pre-intervention, the average number of GAS reflex cultures per month was 66, which fell to 34 following the change to ordering language. The percentage of total RADT tests that underwent reflex culture changed from 99.5% to 49.0% before and after the intervention. Conversely, the number of RADT tests with no reflex culture ordered showed a proportional increase. To ensure that GAS cultures were not being ordered through a different route, the number of add-on culture orders was also tracked, with no marked increase in these orders during the intervention period. CONCLUSION: While this notable decrease in reflex culture ordering following negative RADT is promising, there is ongoing room for improvement, which could be addressed by additional reminders to physicians within the ED. If successful, similar interventions will be implemented in other departments. DISCLOSURES: All authors: No reported disclosures.