Evaluation of antimicrobial therapy orders circumventing an antimicrobial stewardship program: investigating the strategy of "stealth dosing"

Abstract

Objective: Prior-approval antimicrobial stewardship programs (ASPs) improve patient outcomes and decrease antimicrobial resistance. These benefits would be limited if physicians circumvented ASP efforts. We evaluated whether prescribers wait until after the prior-approval period to order restricted antimicrobial therapy that is in conflict with guidelines or unnecessary.

Design: A cross-sectional study design and a retrospective cohort study design.

Setting: A tertiary care, academic medical center with a prior-approval ASP that was active between 8 am and 10 pm.

Methods: We evaluated whether there was an increase in the proportion of orders for antimicrobial therapy that involve restricted (vs nonrestricted) antimicrobials during the first hour that the ASP is inactive (ie, the first hour that prior approval is not required), compared with the remainder of the day. We also evaluated whether restricted antimicrobial therapy ordered during this first hour is less likely to be continued when the ASP becomes active the next day, compared with that ordered during the preceding hour.

Results: A greater proportion of the antimicrobial therapy orders placed between 10:00 pm and 10:59 pm were for restricted agents, compared with orders placed during other periods (57.0% vs 49.9%; P=.02). Surgical patients for whom antimicrobial therapy orders were placed between 10:00 pm and 10:59 pm were less likely to have that antimicrobial therapy continued, compared with patients whose therapy was ordered between 9:00 pm and 9:59 pm (60.0% vs 98.1%; P<.001). Nonsurgical patients whose therapy orders were placed between 10:00 pm and 10:59 pm were also less likely to have the ordered antimicrobial therapy continued, compared with patients whose therapy was ordered between 9:00 pm and 9:59 pm (70.8% vs 84.2%; P=.01).

Conclusion: Physicians avoid having to obtain prior approval for therapy involving restricted antimicrobials by waiting until restrictions are no longer active to place orders. Compared with restricted antimicrobial therapy ordered when the ASP is active, these courses of therapy are less often continued by the ASP, suggesting that they are more likely to be in conflict with guidelines or unnecessary.

Figure 1. Frequency of Restricted and Unrestricted Antimicrobial Orders by Hour

Note: frequency was calculated by hour; the plot lines were smoothed to more easily compare the two ordering patterns.

Figure 2. Proportion of Orders for Restricted Antimicrobials by Hour