# P-804. Dulling Reflexes: Refining Criteria for Reflex Urine Cultures in Outpatient Areas and Emergency Departments to Decrease Inappropriate Antimicrobial Use

## TL;DR

This study shows how changing urine test rules in clinics and ERs can reduce unnecessary cultures and antibiotic use.

## Contribution

The paper introduces refined criteria for reflex urine cultures to decrease antimicrobial overuse in outpatient and ED settings.

## Key findings

- Reflex urine cultures decreased from 51% to 40% post-intervention.
- Removing the positive bacteria criterion averted 91% of unnecessary reflex cultures.
- Only 6% of non-reflexed cases developed UTIs within 28 days.

## Abstract

Urinary tract infections (UTIs) are overdiagnosed and overtreated. Too-sensitive lab criteria for urine culture reflex contribute to the problem, leading to possible patient harm stemming from misdiagnoses and antibiotic overuse.Percent of Urinalysis with Culture Reflex Orders that Reflexed by LocationUAs that reflexed to culture fell from 51% (CI: 50-52%) pre-intervention to 40% [CI: 38-41%] post by refining culture reflex criteria.

Percent of Urinalysis with Culture Reflex Orders that Reflexed by Location

UAs that reflexed to culture fell from 51% (CI: 50-52%) pre-intervention to 40% [CI: 38-41%] post by refining culture reflex criteria.

In July 2024, UNC Medical Center made urinalysis (UA) with culture reflex criteria changes in outpatient areas (OP) and the emergency department (ED): increase white blood cell counts (WBC) from 4 to 10/hpf; exclude positive bacteria; add new triggers for positive pregnancy test or absolute neutrophil count of < 0.5 x 10^9/L; lower patient age for exclusion from < 5 years to < 3 months; and keep leukocyte esterase or positive nitrites. Any single inclusion criterion in the UA triggered a reflex culture.

We conducted a retrospective pre- and post-study of patients with an outpatient- or ED-initiated urinalysis with culture reflex between 1/1-6/30/2024 (pre) and 8/1-12/31/2024 (post). EMR data was analyzed and descriptive statistics were calculated.

For all OP and ED, UAs that reflexed to culture fell from 51% [CI: 50-52%] in the pre-period to 40% [CI: 38-41%] post-period. In the ED, reflex cultures decreased from 51% [CI: 50-52%] to 40% [CI: 38-41%]. In OP, reflex culture decreases from 47.6% [44.7-50.6%] to 42% [CI: 38.9-45.5%] were not significant.

Changes averted 27% (1,472) of reflex culture orders that would have reflexed under the pre-period criteria. Of those, 91% were averted due to removal of the positive bacteria criterion, 21% due to WBC count, and 1% due to age.

For patients whose UAs did not reflex, 6% (279/5,372) were diagnosed with cystitis, UTI, or pyelonephritis within 28 days (51% were cystitis, 28% UTI, and 27% pyelonephritis). New urine culture orders were placed within 24 hours for 2% (119).

Modifying reflex triggers for UA to reflex culture orders in ED decreased the number of inappropriate cultures performed.

Kevin Alby, PhD, Gradientech: Grant/Research Support