Background: International Classification of Diseases, Tenth Edition (ICD-10) data help track outpatient antibiotic prescribing but lack validation in immunocompromised populations or subspecialty clinics for this purpose. Asymptomatic bacteriuria (ASB) and urinary tract infection (UTI) are important stewardship targets in renal transplant (RT) patients, but they may require alternative metrics to best monitor prescribing patterns. We describe ICD-10 utilization for RT clinic encounters in which antibiotics were prescribed. We developed a metric classifying “acute urinary antibiotics” (AUA) to track antibiotic use for ASB and UTI, and we validated systematic identification of AUA to enable practical implementation. Methods: We examined RT clinic visit and telemedicine encounters from 2018 to 2021 conducted 1 month after transplant. This project was deemed non–human-subjects research by the Stanford Panel on Human Subjects in Medical Research. Results: The analytic cohort included 420 antibacterial prescriptions from 408 encounters (Fig. 1). Of 238 patients, 136 (57%) were male and 112 (47%) were Hispanic or Latino. The most common primary ICD-10 code was Z94.0 (kidney transplant status) (N = 302 of 408 encounters, 75%); 26 encounters (6%) were coded for UTI (eg, N39.0, urinary tract infection, site not specified); and 214 encounters (53%) had multiple ICD-10 codes. The R82.71 code (bacteriuria) was never used. However, 215 prescriptions (51%) were classified as AUA (Fig. 2). The validation cohort included 130 prescriptions; 59 (45%) were classified as AUA and 51 (39%) had documented intent to treat ASB or UTI (positive percent agreement, 83%; negative percent agreement, 97%) (Table 1). For patients >1 month after transplant, the positive percent agreement was 95% and the negative percent agreement was 98%. Of 51 patients receiving AUA, 32 (63%) were asymptomatic despite frequently having a code for UTI (Fig. 3). Conclusions: ICD-10 coding may not be helpful in monitoring antibiotic prescribing in RT patients. The AUA metric offers a practical alternative to track antibiotic prescribing for urinary syndromes and reliably correlates with physician intent. Monitoring AUA prescribing rates could help identify opportunities to optimize antibiotic use in this complex outpatient setting.

Disclosures: None

Background: Using patient data from the electronic health record (EHR) and computer logic, an “electronic phenotype” can be created to identify patients with community-acquired pneumonia (CAP) in real time to assist with syndrome-specific antimicrobial stewardship efforts.1 We adapted and validated the performance of an inpatient CAP electronic phenotype for antimicrobial stewardship interventions. Methods: An automated scoring system was created within the EHR (Epic Systems) to identify hospitalized patients with CAP based on the variables and logic listed in Fig. 1B. We adapted a score used by the Michigan Hospital Medicine Safety Consortium (HMS) to identify patients with CAP, with additions made to improve sensitivity (Fig. 1).1 The score can be displayed in a column within the EHR patient list (Fig. 2). We validated the electronic phenotype via chart review of all hospitalized patients on systemic antimicrobials admitted to a medicine team consecutively between November 8 and 18, 2021. Patients who were readmitted within the validation time frame were excluded. We assessed the performance of the electronic phenotype by comparing the score to manual chart review, where “CAP diagnosis” was defined as (1) mention of “pneumonia” or “CAP” as part of the differential diagnosis in the admission documentation, (2) antimicrobials were started within 48 hours of admission, and (3) radiographic findings were suggestive of pneumonia. After initial evaluation, the scoring system was adjusted, and performance was re-evaluated during prospective audit and feedback performed on EHR CAP–positive patients over 13 days between July 2022 and December 2022. Results: We included 191 patients in our initial validation cohort. The CAP score had high sensitivity (95.83%), specificity (92.2%), and negative predictive value (99.35%), though lower positive predictive value (63.89%) was noted (Table 2). The rules were further refined to include bloodstream infection only with Haemophilus influenza or Streptococcus pneumoniae in rule 2B, and azithromycin was removed from “CAP antibiotics.” After these changes, repeated evaluation of 88 patients with positive CAP EHR score was performed, and only 20 (23%) were considered false-positive results. Conclusions: Electronic phenotypes can be used to create automated tools to identify patients with CAP with reasonable performance. Data from this tool can be used to guide more focused antimicrobial stewardship interventions and clinical decision support in the future. Reference: Vaughn VM, et al. A statewide collaborative quality initiative to improve antibiotic duration and outcomes in patients hospitalized with uncomplicated community-acquired pneumonia. Clin Infect Dis 2022;75:460–467.

Disclosures: None

Background: Billing data have been used in the outpatient setting to identify targets for antimicrobial stewardship. However, COVID-19 ICD-10 codes are new, and the validity of using COVID-19 ICD-10 codes to accurately identify COVID-19 encounters is unknown. We investigated COVID-19 ICD-10 utilization in our urgent care clinics during the pandemic and the impact of using different COVID-19 encounter definitions on antibiotic prescribing rates (APRs). Methods: We included all telemedicine and office visits at 2 academic urgent-care clinics from January 2020 to September 2021. We extracted ICD-10 encounter codes and testing data from the electronic medical record. We compared encounters for which COVID-19 ICD-10 codes were present with encounters for which SARS-CoV-2 nucleic acid amplification testing (NAAT) was performed within 5 days of and up to 2 days after the encounter (Fig. 1). We calculated the sensitivity of the use of COVID-19 ICD-10 codes against a positive NAAT. We calculated the APR as the proportion of encounters in which an antibacterial drug was prescribed. This quality improvement project was deemed non–human-subjects research by the Stanford Panel on Human Subjects in Medical Research.

Funding: None

Disclosures: None

Background: Emerging evidence supports the use of billing data to identify stewardship targets in primary care. Standardizing an approach to antibiotic prescribing rate (APR) calculations could facilitate external benchmarking. Methods: Using methodology and an ICD-10 dictionary validated in urgent care clinics,1 we created an expanded ICD-10 dictionary to incorporate additional ICD-10 codes from primary care associated with antibiotic prescriptions (Fig. 1). We then compared antibiotic prescribing rates using the urgent care and expanded dictionaries. We included all primary care visits from 2019 to 2020 and extracted ICD-10 codes and antibiotic order data. Using the urgent care and expanded ICD-10 dictionary, we classified each encounter by prescribing tier based on whether antibiotics are almost always (tier 1), sometimes (tier 2), or almost never (tier 3) indicated. For encounters with ICD-10s in multiple tiers, we chose the lowest tier. For multiple ICD-10 codes within the same tier, we chose the first extracted ICD-10 code. We calculated antibiotic prescribing rates as the proportion of encounters associated with ≥ 1 antibacterial prescription. This quality improvement project was deemed non–human subjects research by the Stanford Panel on Human Subjects in Medical Research. Results: The urgent care dictionary has 1,400 ICD-10 codes. We added 1,439 ICD-10 codes derived from primary care encounters to create the expanded ICD-10 dictionary (8.5% tier 1, 9.1% tier 2, and 82.4% tier 3) (Fig. 1). We identified 177,531 encounters; 74% had ≥ 2 associated ICD-10 codes (Fig. 2). In total, 147,085 encounters (82.9%) were classified into a tier using the urgent care dictionary. An additional 22,039 encounters were classified with the expanded dictionary (Table 1). Most added encounters were tier 3 with low 0.7% APR (Tables 1 and 3). In total, 41,473 (28.2%) encounters were classified differently depending on the ICD-10 dictionary used, most commonly changing from tier 3 to tier 2 without an increase in overall tier 2 antibiotic prescribing rate (Tables 2 and 3). Overall antibiotic prescribing rates were similar when using either the urgent care or expanded ICD-10 dictionary (Table 2). Conclusions: The expanded ICD-10 dictionary allowed for classification of more encounters in primary care; however, it did not meaningfully change antibiotic prescribing rates. Antibiotic prescribing rates were likely diluted by classifying more encounters without identifying an associated increase in antibiotic prescribing. A more sophisticated classification system may help to accommodate the diversity and volume of ICD-10 codes used in primary care.

1. Stenehjem E, et al. Clin Infect Dis 2020;70:1781–1787.

Funding: None

Disclosures: None