1681. Location, location, location: A change in urine testing order sets on culturing practices at an academic medical center emergency department
Session: Oral Abstract Session: Studies that will Impact your Practice
Friday, October 28, 2016: 2:30 PM
Room: 275-277

Limited knowledge exists regarding emergency department (ED) provider ordering practices for urine testing. We evaluated the impact of a change in electronic order sets on urine testing practices in the ED of an urban academic medical center.


On 10/21/2015, urinalysis with reflex to culture (cx) was moved from an electronic order set of “commonly ordered tests” to category-specific order sets (deeply embeded) for hematology/coagulation/urinalysis and microbiology. Urinalysis with reflex to microscopy was kept on the common order set and urine cx was kept in the microbiology order set. We conducted a retrospective analysis of ED patients (pts) seen between 9/1/2015 – 11/30/2015 with ≥ 1 urine test (urine cx, urinalysis and/or microscopy) ordered. Patient and lab data were abstracted from hospital informatics database. We compared patient and urine testing data in the 50 days before (9/1 – 10/20/2015) and 41 days after (10/21 – 11/30/2015) the change. We used auto regressive integrated moving average to assess % change in daily urine tests per 1,000 ED patient (pt) visits.


During the study period, 6,466 (28.2%) of 22,948 ED pts had ≥ 1 urine test ordered [2805 (27.8%) post- vs. 3661 (28.5%) pre-change; p =0.22]. Post-change pts with urine testing were more likely to be admitted to the hospital (45.1% vs. 33.6%; p<0.001). There were no differences in gender (37.0% male post- vs. 35.7% pre-change; p=0.52) or Black race (58.0% vs. 56.1%; p=0.12). Urine tests per 1000 pt visit decreased in the post-change period (urine dipstick: 270.4 post- vs. 284.1 / 1000 pt visits pre-change, p=0.022; urine microscopy: 182.8 vs. 194.2 / 1000 pt visits, p=0.03; urine cx: 22.5 vs. 37.3 / 1000 pt visits, p <.001). When adjusted for overall temporal trend the daily culture rate decreased by 40.4% [-40.4%, 95% confidence interval (CI): -62.2, -6.1%], but dipstick (1.7%, 95%CI:-29.0, 45.7%) and microscopy rates (-6.8%, 95%CI: -37.4, 38.6%) were unchanged. The percent of positive urine cxs was unchanged [142 (62.6%) post- vs. 323 (67.4%) pre-change; p=0.20].


A single change to an electronic order set resulted in a 40% reduction in urine cultures ordered in the ED. There was no change in the ordering of urine dipstick or microscopy tests or proportion of positive urine cultures.

Satish Munigala, MBBS, MPH1, Robert Poirier, MD1, Stephen Liang, MD, MPHS1, Helen Wood, RN, BSN, CIC2, Ronald Jackups Jr., MD3 and David Warren, MD, MPH1, (1)Internal Medicine, Washington University in St. Louis, St. Louis, MO, (2)Barnes Jewish Hospital, St. Louis, MO, (3)Laboratory & Genomic Medicine, Washington University in St. Louis, St. Louis, MO


S. Munigala, None

R. Poirier, None

S. Liang, None

H. Wood, None

R. Jackups Jr., None

D. Warren, None

Findings in the abstracts are embargoed until 12:01 a.m. CDT, Wednesday Oct. 26th with the exception of research findings presented at the IDWeek press conferences.