Abstract
PURPOSE: To evaluate judicious antibiotic prescribing of benzathine penicillin G (BPG) after implementation of an electronic health record-based medication shortage alert during a critical drug shortage. METHODS: This was an institutional review board-approved retrospective cohort study of patients aged ≥3 months who received BPG between May 9, 2023, and February 28, 2024. The study included inpatient and outpatient visits after implementing a BPG medication shortage alert; patients with severe penicillin allergy, neurosyphilis, or congenital syphilis were excluded. Judicious BPG use was defined as use in patients diagnosed with primary, secondary, or latent syphilis or if they were prescribed a BPG alternative in response to the medication shortage alert; nonjudicious use included BPG for alternative diagnoses. Social determinants of health were assessed as exposure variables of interest. A separate cohort of syphilis patients receiving BPG or alternative therapy (i.e., doxycycline) was described. RESULTS: A total of 453 patients were included. Most patients were non-Hispanic Black (n = 273, 60%) men (n = 272, 60%) with a median (interquartile range) age of 32 (22-44) years. Of these, 318 (70%) received judicious BPG, whereas 135 (30%) received nonjudicious BPG. The most nonjudicious diagnosis was streptococcal pharyngitis (n = 128, 95%). Variables associated with judicious use included age >32 years (adjusted odds ratio [adjOR], 2.273; 95% CI, 1.488-3.472), male sex (adjOR, 1.835; 95% CI, 1.206-2.792), and black race (adjOR, 1.847; 95% CI, 1.212-2.815). Among a cohort of 128 syphilis patients who received BPG (n = 64, 50%) or doxycycline (n = 64, 50%), those who received doxycycline were more likely be uninsured (35 [54.7%] vs 43 [67.2%]; P = .15) and receive outpatient treatment (3 [4.7%] vs 12 [18.7%]; P = .13). CONCLUSION: Despite implementing an electronic health record drug shortage alert, 30% of BPG use was nonjudicious and mostly for pharyngitis.