Background: Research on the association between the standardized antimicrobial administration ratio (SAAR) and clinical outcomes is lacking. Objective: We compared SAAR and patient outcomes in 97 acute-care facilities affiliated with a large healthcare system. Methods: Facilities were classified using the broad-spectrum hospital-onset (BSHO) SAAR for medical, surgical, and medical-surgical wards as low, moderate, or high antimicrobial use: low use SAAR, <0.8; moderate use SAAR, 0.95–1.05; and high-use SAAR, >1.2. Data were included from patients aged ≥18 years who were discharged between the first quarter of 2018 and the second quarter of 2019, had nonmissing matching criteria, BMI between 10 and 90, and at least 1 BSHO medication administered in a medical, surgical, or medical-surgical ward. Patients were matched for gender, age group, BMI category, year and quarter of discharge, ICU stay, and diagnosis-related group (DRG). Eligible drugs included all routes for cefepime, ceftazidime, doripenem, imipenem/cilastatin, meropenem, and piperacillin/tazobactam and IV only for amikacin, aztreonam, gentamicin, and tobramycin. Outcomes were evaluated in a pairwise manner using t tests or χ2 tests. Results: Each of the 3 study groups consisted of 6,327 patients, 51% of whom were men; average age, 63 years; 70% of whom were obese or overweight, and 19% of whom had an ICU stay. The most common DRG code was infectious and parasitic diseases (57%) followed by digestive system (9%), respiratory system (7%), and kidney and urinary tract (6%). High antibiotic use was associated with longer length of stay and a higher estimated cost per visit. Low antibiotic use was associated with higher rate of mortality and a lower rate of readmissions compared to moderate use. The low-usage group did not exhibit a statistically significant difference in mortality, readmissions, or rate of C. difficile compared to the high-usage group. Conclusions: The optimal antibiotic utilization group varied among outcomes. Further evaluation of outcomes is needed for the SAAR to understand the ranges and the relationship between the measure and clinical outcomes.

Funding: None

Disclosures: None