The following was previously posted by Dr. Marcelin to SHEA Journal Club published online in February 2019.
Electronic clinical decision support tools and rapid diagnostic testing have significantly impacted the way we practice Infectious Diseases. Despite these scientific gains, Antimicrobial Stewardship still requires an understanding of the behavioral science of prescribing. Prior studies have demonstrated that antibiotic prescribing may be influenced by specific behavioral interventions, such as peer comparison, or “nudge” theory.
Yadav et al. conducted a quasi-experimental quality improvement study to determine the impact of implementing an “Expected Practice” (EP) method to alter antibiotic prescribing practices towards favoring shorter duration of therapy (DOT). This method leverages the prescriber’s desire to meet their own institutional expectations, which may be viewed as more authoritative than external medical society guidelines. The authors chose DOT as the endpoint because clinicians were concerned about potential poor outcomes with shorter DOT.
The EP document included a wealth of evidence supporting shorter treatment durations for the uncomplicated infections, including urinary tract infections (UTI, 1-5 days), pyelonephritis (5-7 days), skin/soft tissue infections (SSTI, 5-6 days), community-acquired pneumonia (CAP, 5 days), and ventilator-acquired pneumonia (VAP, 7 days). The authors already had an established antimicrobial stewardship program, whose practices did not change during the study, with the exception of implementing procalcitonin testing.
In the 12 months after implementing EP, the average antibiotic DOT decreased by 10%, 11%, 11%, and 27% for UTI, SSTI, CAP and VAP respectively, with concomitant decreases in total antibiotic exposure measured in total milligrams administered. This impact was statistically sustained by the end of that year for UTI and CAP, but seemed to wane toward the end of the year for SSTIs and VAP.
The procalcitonin test that was also implemented during that period was associated with statistically significant increase of antibiotic dose exposure and DOT, which was thought to be due to patients with more complex illness having the test performed and subsequently requiring longer treatment durations. Mortality was unchanged post-intervention.
The only condition in the EP where antibiotics were not recommended was asymptomatic bacteriuria, and though there is no way to be certain, it is possible that this particular recommendation contributed to the decreased DOT for UTI, by reducing the number of antibiotic starts for “UTI” that was really asymptomatic bacteriuria and didn’t require antibiotics at all. This study demonstrated that providing expectations of practice regarding specific conditions can have important clinical impact on prescribing.
The influence of this change goes beyond simple DOT or dose exposure, to other important factors not measured by this study, like impact on C. difficile infection rates, impact on antimicrobial resistance rates, reduction of antibiotic-related adverse drug events, hospital length of stay.
Kabir Yadav, Eriko Masuda, Emi Minejima, Brad Spellberg; Expected Practice as a Novel Antibiotic Stewardship Intervention, Open Forum Infectious Diseases, Volume 6, Issue 1, 1 January 2019, ofy319, https://doi.org/10.1093/ofid/ofy319