BACTERIOLOGY

crossm

Incidence and Diagnostic Yield of Repeat Urine Culture in

Hospitalized Patients: an Opportunity for Diagnostic
Stewardship
Kap Sum Foong,a Satish Munigala,b Ronald Jackups, Jr.,c Melanie L. Yarbrough,c Carey-Ann D. Burnham,c
David K. Warrenb

a Section of Infectious Diseases, Department of Medicine, University of Illinois College of Medicine, Peoria, Illinois, USA
b Division of Infectious Diseases, Department of Medicine, Washington University School of Medicine, St. Louis, Missouri, USA
c
Department of Pathology and Immunology, Washington University School of Medicine, St. Louis, Missouri, USA

ABSTRACT There is limited knowledge on the incidence, diagnostic yield, and cost
associated with inappropriate repeat urine cultures. The factors that affect repeat
urine culturing practices are not well understood. We conducted a retrospective
study of adult inpatients who had ⱖ1 urine culture performed during their hospital-
ization between January 2015 and February 2018. We analyzed the proportion of in-
appropriate repeat urine cultures performed ⬍48 h after the index culture. We de-
fined an inappropriate repeat urine culture to be a repeat urine culture performed
following a negative index culture or a repeat urine specimen obtained from the
same urinary catheter. Overall, 28,141 urine cultures were performed on 21,306 pa-
tients. There were 2,060 (7.3%) urine cultures repeated in ⬍48 h. Of these, 1,120
(54.4%) urine cultures were inappropriate. Predictors for inappropriate repeat urine
cultures included collection of the initial urine sample for culture in the emergency
department (adjusted odds ratio [aOR], 5.65; 95% confidence interval [CI], 4.70 to
6.78), male gender (aOR, 1.61; 95% CI, 1.42 to 1.84), congestive heart failure (aOR,
1.20; 95% CI, 1.03 to 1.38), and a longer hospital stay (aOR, 1.01 per day; 95% CI,
1.00 to 1.01). A patient with an index urine culture obtained from an indwelling
catheter (aOR, 0.65; 95% CI, 0.53 to 0.80) was less likely to have an inappropriate re-
peat culture. Among 1,120 negative index urine cultures, only 4.7% of repeat cul-
tures were positive for bacteriuria. The estimated laboratory charges for inappropri-
ate repeat urine cultures were $16,800 over the study period. Among inpatients,
over half of all urine cultures repeated in ⬍48 h were inappropriate. This offers an
opportunity for diagnostic stewardship and optimization of antimicrobial use. Citation Foong KS, Munigala S, Jackups R, Jr,
Yarbrough ML, Burnham C-AD, Warren DK.
2019. Incidence and diagnostic yield of repeat
KEYWORDS diagnostic stewardship, diagnostic yield, inappropriate testing, urine culture in hospitalized patients: an
opportunity for diagnostic stewardship. J Clin
urine culture Microbiol 57:e00910-19. https://doi.org/10
.1128/JCM.00910-19.
Editor Nathan A. Ledeboer, Medical College of

I n recent decades, the overutilization of laboratory testing has been recognized as a

prevalent issue in health care. Inappropriate duplicative testing is an important subset of
diagnostic test overuse (1). Although duplicative laboratory testing is indicated in some
Wisconsin
Copyright © 2019 American Society for
Microbiology. All Rights Reserved.
instances, it is often unnecessary and provides limited clinically important information, Address correspondence to Kap Sum Foong,
[email protected], or David K. Warren,
especially if the laboratory test is repeated in an inappropriately short interval (2, 3). [email protected].
Urine cultures are among the most frequently performed tests in clinical microbi- Received 5 June 2019
ology laboratories (4). An accurate diagnosis of urinary tract infection is important to Returned for modification 1 July 2019
identify patients who will benefit from appropriate antimicrobial treatment. However, Accepted 25 July 2019
Accepted manuscript posted online 7
unnecessary urine culture testing can result in increased health care costs and inap-
August 2019
propriate antimicrobial treatment for asymptomatic bacteriuria, which impedes anti- Published 24 September 2019
microbial stewardship efforts and can lead to downstream antimicrobial resistance and

October 2019 Volume 57 Issue 10 e00910-19 Journal of Clinical Microbiology jcm.asm.org 1

Foong et al. Journal of Clinical Microbiology

Clostridioides difficile infection (5). Utilization practices, the diagnostic yield of repeat
urine culture among hospitalized patients, and the factors that affect ordering practices
are not well understood.
The purpose of our study was to evaluate the incidence, predictors, diagnostic yield,
and economic burden of inappropriate repeat urine cultures performed less than 48 h
after the index culture among hospitalized patients. These results have the potential to
improve diagnostic test utilization, thereby reducing health care costs and minimizing
inappropriate antimicrobial therapy.

MATERIALS AND METHODS

Setting. This was a retrospective study conducted at Barnes-Jewish Hospital, a 1,250-bed teaching
hospital, from January 2015 to February 2018. We included all hospitalized patients aged ⱖ18 years who
had at least one urine sample processed by culture during their hospital stay or within 24 h prior to
admission.
Definitions. For the purpose of this study, we defined the first urine culture performed during
admission or within 24 h prior to admission to be the “initial” culture. A sample for an initial urine culture
could be obtained in an outpatient setting (e.g., a clinic, skilled nursing facility, or long-term acute care
facility), the emergency department, or an inpatient setting. We defined an “index” urine culture to be
the first urine culture performed prior to a repeat urine culture(s). For some of the hospitalized patients,
the index urine cultures were also the initial urine cultures.
Based on previous literature (2, 6), we defined a repeat urine culture performed within less than 48 h to
be inappropriate if either (i) a repeat urine culture was performed following a negative index culture or (ii) a
repeat urine specimen was obtained from the same indwelling urinary catheter. We considered a repeat urine
culture to be clinically appropriate if (i) it was repeated for the confirmation of a positive index culture or (ii)
a specimen for a repeat catheterized urine culture was obtained from a newly inserted indwelling urinary
catheter, given concerns of initial culture contamination from an old urinary catheter (2, 6).
Pertinent data on the patients’ demographics, comorbidities, urine cultures (date of collection, type
of urine specimen, culture result, ordering service, ordering physician), results of an urinalysis associated
with the urine cultures (i.e., results of an urinalysis performed either concomitantly with or within 1 h
before the index urine culture), and disposition were extracted from the medical informatics database.
The type of urine specimen was classified into clean-catch/straight catheterized, indwelling catheterized
(e.g., the specimen was obtained from a Foley or suprapubic catheter), or procedural (e.g., the specimen
was obtained by suprapubic aspiration, percutaneous nephrostomy tube placement, or cystoscopy), as
indicated by the ordering physician. Due to the low numbers of straight catheterized urine cultures (0.3%
of all urine cultures), these were grouped together with clean-catch urine cultures. A positive urine
culture for all specimen types was defined on the basis of the established clinical breakpoints, as
previously published (7, 8). We defined a negative urine culture as having insignificant growth below a
threshold or no growth. A urine culture was considered contaminated if there were ⱖ3 different species
of organisms with colony counts recovered above a threshold. An abnormal urinalysis was defined as a
leukocyte esterase value of ⱖ1 on urine dipstick and ⬎5 white blood cells per high-power field on urine
microscopy (9).
The primary study outcome was the proportion of inappropriate repeat urine cultures performed
within less than 48 h following the index urine culture among inpatients. The secondary outcomes were
the diagnostic yield of inappropriate repeat urine cultures in detecting bacteriuria, predictors, and direct
laboratory charges associated with inappropriate repeat urine cultures performed within less than 48 h.
The gross charges estimation for a urine culture test was calculated on the basis of the Medicare Clinical
Laboratory Fee Schedule, using a national median Medicare payment rate of $15.00 per urine culture (not
adjusted to inflation) (10).
Statistical analyses. The unit of analysis in our study was reported on a per admission basis. Baseline
characteristics and urine culture data were compared on the basis of inappropriate repeat urine culturing
status using the chi-square test, the Fisher exact test, or univariable logistic regression for categorical
variables and the Kruskal-Wallis test for continuous variables. To determine the independent predictors
associated with inappropriate repeat urine cultures performed within less than 48 h, we performed
multivariable logistic regression analysis. All variables with a P value of ⬍0.20 in univariable analysis were
considered for entry into the model using forward stepwise regression, with the variable being retained
in the final model if the P value was ⬍0.05. The remaining patients in the study without inappropriate
repeat urine cultures during their admission (i.e., patients with only one urine culture or patients with an
appropriate repeat urine culture) were regarded as the reference group (controls). We also evaluated the
effect of urine culture specimen type (index versus initial) and location among the independent
predictors of inappropriate repeat urine cultures. For this, random selection of either the index culture
or the initial culture for the control group was performed to avoid a selection bias of oversampling of
urine cultures performed with specimens collected in the emergency department.
We performed a sensitivity analysis to determine if our results varied on the basis of inappropriate
criteria for repeat urine cultures. For this analysis, we considered all repeat urine cultures performed
within less than 48 h after the index culture to be inappropriate. Data were analyzed using SAS (version
9.3) software (SAS Institute Inc., Cary, NC). This study was approved by the Human Research Protection
Office at Washington University in St. Louis, MO, with a waiver of consent.

October 2019 Volume 57 Issue 10 e00910-19 jcm.asm.org 2

Repeat Urine Culture among Inpatients Journal of Clinical Microbiology

RESULTS
A total of 28,141 urine cultures were performed among 21,306 hospitalized patients
during the study period. There were 23,224 clean-catch specimens, 97 straight cathe-
terized specimens, 4,567 indwelling catheter specimens, and 253 procedural speci-
mens. Of these, 4,678 (22.0%) patients had more than one urine culture performed,
accounting for 11,513 of all urine cultures (4,678 index cultures, 6,835 repeat urine
cultures). Among the 4,678 patients with repeat urine cultures, 1,868 (39.9%) had
repeat urine cultures performed within 48 h following the index culture. We identified
that 1,120 (54.4%) of 2,060 urine cultures were inappropriately repeated within 48 h
following the index urine culture. Of all inappropriate repeat testing, the proportion
was the highest for patients who had clean-catch index urine cultures (94.5%) and who
were admitted to intensive care units (ICUs) (40.2%). Only 53 (4.7%) of 1,120 inappro-
priate repeat urine cultures performed within less than 48 h had a positive urine culture.
Of these, 41.5% were positive for Enterobacteriaceae, 18.9% were positive for entero-
cocci, and 17.0% were positive for Candida spp.
Baseline demographics were relatively similar between the control group and the
groups with inappropriate repeat urine cultures (Table 1). In univariable analysis,
patients with inappropriate repeat urine culture testing were more likely to be male
(59.1% versus 47.1%, P ⬍ 0.001), have a sample for the index urine culture obtained in
the emergency department or other outpatient settings (16.8% versus 3.6%, P ⬍ 0.001),
and have a longer median length of hospital stay (9 versus 7 days, P ⬍ 0.001). Index
urine cultures were less likely to be of an indwelling catheterized specimen in the
inappropriate repeat testing group (10.3% versus 13.6%, P ⫽ 0.002). There was no
difference in the proportion of urinalyses performed in association with the index
cultures between the two study groups. The majority of the inappropriate repeat urine
cultures were ordered by the same clinical service that ordered the index culture,
ranging from 61.5% to 96.1% (Table 2). Notably, 29% (n ⫽ 325 out of 1,120) of
inappropriate repeat urine cultures were ordered by the same physician.
Among inpatients who had at least one urine culture, the independent predictors
associated with having an inappropriate repeat urine culture performed within 48 h of
the index culture included an initial urine culture performed with a specimen collected
in the emergency department (adjusted odds ratio [aOR], 5.65; 95% confidence interval
[CI], 4.70 to 6.78), male gender (aOR, 1.61; 95% CI, 1.42 to 1.84), congestive heart failure
(aOR, 1.20; 95% CI, 1.03 to 1.38), and a longer hospital stay (aOR, 1.01 per day; 95% CI,
1.00 to 1.01 per day). An index indwelling catheterized urine culture (aOR, 0.65; 95% CI,
0.53 to 0.80) was less likely to be associated with inappropriate repeat testing. In our
sensitivity analysis, there were no significant changes in our univariable and multivari-
able results when we compared all inpatients who had repeat urine cultures performed
within less than 48 h, irrespective of their clinical appropriateness criteria (data not
shown). The estimated laboratory charges for these inappropriate repeat urine cultures
were $16,800 during the 38-month study period.

DISCUSSION
In this retrospective study, we found that, for 7.3% of all inpatient urine cultures,
repeat testing was performed within 48 h following an index culture and that 54.4% of
these repeated urine cultures were inappropriate, based on previously published
criteria. Among inpatients with a negative index urine culture, the diagnostic gain of an
inappropriate urine culture repeated within 48 h for detecting bacteriuria was only
4.7%. Several other studies have examined the utility of repeat microbiological cultures
among inpatients. However, these studies were performed in the late 1990s, and the
landscape of diagnostic test utilization and antimicrobial resistance has changed
considerably since that time. A study by Bates et al. found that 26% of the repeat urine,
sputum, and stool cultures performed within 36 h were redundant (2). Onderdonk et al.
reported that 14.6% of the repeated urine cultures were unnecessary, based on
predefined testing frequency criteria (e.g., no more than one urine culture per day) (11).
Nevertheless, the proportion of inappropriate inpatient urine cultures repeated in less

October 2019 Volume 57 Issue 10 e00910-19 jcm.asm.org 3

Foong et al. Journal of Clinical Microbiology

TABLE 1 Characteristics and comparison of risk factors for inappropriate repeat urine culture performed within 48 h following the index
urine culture among 21,306 inpatients with ⬎1 urine culturej
Value for the following patients: Multivariable analysis
Inpatients with >1
inappropriate repeat Remaining
All inpatients urine culture in patientsa
Characteristic (n ⴝ 21,306) <48 h (n ⴝ 1,040) (n ⴝ 20,266) P value aOR (95% CI) P value
No. (%) of male patients 10,155 (47.7) 615 (59.1) 9,540 (47.1) ⬍0.001 1.61 (1.42–1.84) ⬍0.001
Median (IQR) age (yr) 62 (49–72) 61 (49–71) 62 (49–72) 0.175 1.00 (0.99–1.00) 0.088
Median (IQR) length of stay (days) 7 (4–16) 9 (4–20) 7 (4–16) ⬍0.001 1.01 (1.00–1.01) ⬍0.001

No. (%) of patients by race 0.339

White 14,649 (68.8) 729 (70.1) 13,920 (68.7)
Othersb 6,657 (31.2) 311 (29.9) 6,346 (31.3)

No. (%) of patients with the following

comorbidities:
CHF 5,609 (26.3) 305 (29.3) 5,304 (26.2) 0.024 1.20 (1.03–1.38) 0.018
Chronic lung disease 4,762 (22.4) 213 (20.5) 4,549 (22.4) 0.138
ESRD 897 (4.2) 54 (5.2) 843 (4.2) 0.106
Cirrhosis 862 (4.0) 39 (3.8) 823 (4.1) 0.620
Malignancy 3,544 (16.6) 169 (16.3) 3,375 (16.7) 0.733
HIV infection 67 (0.3) 4 (0.4) 63 (0.3) 0.569
Genitourinary procedure 253 (1.2) 9 (0.9) 244 (1.2) 0.326

No. (%) of patients with an initial urine

culture by specimen type
Clean catch/straight catheterized 18,252 (85.7) 933 (89.7) 17,319 (85.4) Reference Reference
Indwelling catheterized 2,857 (13.4) 107 (10.3) 2,750 (13.6) 0.002 0.65 (0.53–0.80) ⬍0.001
Proceduralc 197 (0.9) 197 (1.0)

No. (%) of patients with an initial urine

culture by locationd
ED and other outpatient facilitye 907 (4.3) 175 (16.8) 732 (3.6) ⬍0.001 5.65 (4.70–6.78) ⬍0.001
Inpatient 20,399 (95.7) 865 (83.2) 19,534 (96.4)

No. (%) of patients with an urinalysisf 6,738 (31.6) 310 (29.8) 6,428 (31.7) 0.196
No. (%) of patients with an abnormal 3,915 (18.4) 170 (16.3) 3,745 (18.5) 0.080
urinalysisg

No. (%) of patients with the following

discharge statush:
Discharged to home 13,860 (65.5) 606 (58.9) 13,254 (65.9) Reference Reference
Discharged to facility 5,419 (25.6) 284 (27.6) 5,135 (25.5) 0.010 1.10 (0.94–1.28) 0.086
Othersi 1,879 (8.9) 139 (13.5) 1,740 (8.6) ⬍0.001 1.60 (1.31–1.95) ⬍0.001
aIncludes inpatients who had single urine cultures or appropriate repeated urine cultures.
bThis includes black (all patients ⫽ 5,447, patients with an inappropriate repeat urine culture ⫽ 251, remaining patients ⫽ 5,196) and other races (all patients ⫽
1,210, patients with an inappropriate repeat urine culture ⫽ 60, remaining patients ⫽ 1,150).
cThis included specimen collection by suprapubic aspiration, percutaneous nephrostomy tube placement, and cystoscopy.

dIn the inappropriate repeat urine culture arm, 75% of the initial cultures were index cultures, while 25% were not.

eOutpatient facility (total ⫽ 82) included a clinic, skilled nursing facility, and long-term acute care facility.

fUrinalysis was performed either concomitantly with or within 1 h before the index urine culture.

gAn abnormal urinalysis was defined as a leukocyte esterase value of ⱖ1 on urine dipstick and ⬎5 white blood cells per high-power field on urine microscopy.

hThere were 148 missing data.

iThis included leaving against medical advice and death.

jAbbreviations: aOR, adjusted odd ratio; CI, confidence interval; IQR, interquartile range; CHF, congestive heart failure; ESRD, end-stage renal disease; HIV, human

immunodeficiency virus; ED, emergency department.

than 48 h remained unclear, and the diagnostic yield of inappropriate repeat urine
cultures was also not described in these two studies (2, 11).
The diagnostic gain from repeat clinical microbiology testing that is performed
within a very short time frame is very limited (2, 3). Despite several clinical practice
guidelines with recommendations for improving urine culturing practice (6, 12, 13),
physician testing practices still vary (1, 2, 5, 11). Previous studies suggested potential
factors contributing to unnecessary repeat testing, including an unawareness of pre-
existing orders and a failure to determine when the last test was performed or to check

October 2019 Volume 57 Issue 10 e00910-19 jcm.asm.org 4

 TABLE 2 Comparison of 1,120 pairs of index and inappropriate repeat urine cultures by ordering servicea
No. (%) of samples with inappropriate repeat urine culture by ordering serviceb
General Neurology/ Gynecological Other inpatient
Ordering service medicine Surgery ICU SCT/oncology Cardiology neurosurgery oncology Obstetrics Orthopedics locationsb
Repeat Urine Culture among Inpatients

Emergency department and other 47 (31.1) 30 (19.8) 48 (31.8) 16 (10.6) 5 (3.3) 3 (2.0) 1 (0.7) 0 1 (0.7) 0
outpatient facilitiesc (n ⫽ 151)
General medicine (n ⫽ 133) 116 (87.2) 0 16 (12.0) 1 (0.8) 0 0 0 0 0 0

October 2019 Volume 57 Issue 10 e00910-19

Surgery (n ⫽ 209) 0 190 (90.9) 17 (8.1) 1 (0.5) 0 0 1 (0.5) 0 0 0
ICU (n ⫽ 357) 9 (2.5) 8 (2.2) 332 (93.0) 0 3 (0.9) 5 (1.4) 0 0 0 0
SCT/oncology (n ⫽ 127) 0 0 12 (9.5) 115 (90.5) 0 0 0 0 0 0
Cardiology (n ⫽ 48) 0 0 10 (20.8) 0 38 (79.2) 0 0 0 0 0
Neurology/neurosurgery (n ⫽ 38) 2 (5.2) 0 11 (29.0) 0 0 25 (65.8) 0 0 0 0
Gynecologic oncology (n ⫽ 26) 0 1 (3.9) 0 0 0 0 25 (96.1) 0 0 0
Obstetrics (n ⫽ 6) 0 0 1 (16.7) 0 0 0 0 5 (83.3) 0 0
Orthopedics (n ⫽ 12) 0 0 0 1 (8.3) 0 0 0 0 11 (91.7) 0
Other inpatient locationsd (n ⫽ 13) 2 (15.4) 0 3 (23.1) 0 0 0 0 0 0 8 (61.5)
aAbbreviations: ICU, intensive care unit; SCT, stem cell transplant.
bThe shaded data are for samples for which repeat urine cultures were ordered by the same clinical service that ordered the index culture.
cOutpatient facilities included a clinic, skilled nursing facility, and long-term acute care facility.

dIncluded are observation and inpatient psychiatric units.

jcm.asm.org 5
Journal of Clinical Microbiology
Foong et al. Journal of Clinical Microbiology

pending results (2, 3, 14, 15). In our current study, there was no significant difference
in the proportion of associated urinalysis orders between the inappropriate repeat
urine culture and the cultures for the control groups, suggesting a lack of awareness of
the index culture results rather than the urinalysis results as the driving force for
inappropriate repeat urine culturing practice. These factors could potentially explain
our finding that most inappropriate repeat urine cultures were ordered by the same
clinical service and that up to one-third were ordered by the same physician who
ordered the index culture. ICUs had the highest number of inappropriate repeat urine
cultures. We previously reported that isolated urine cultures, defined as urine cultures
performed without additional diagnostic testing, such as urinalysis or urine microscopy,
were more likely to be ordered among ICU patients (7). Interventions targeted to ICUs
and physicians with outlying ordering behavior may therefore have a great impact on
reducing inappropriate repeat testing.
Among the independent risk factors in our multivariable analysis, a more than 5-fold
increased risk for having inappropriate inpatient repeat urine cultures performed within
48 h following the index culture was found when the initial urine culture was per-
formed with a sample obtained in the emergency department. Other investigators have
found similar findings for cardiac enzyme testing and suggested that unnecessary
duplicative orders occurred more frequently when patients were transferred from the
emergency department to inpatient units (16). The role of computerized physician order
entry (CPOE) systems in altering physician ordering practices to reduce unnecessary labo-
ratory testing has been well described (3, 8, 17). A computerized pop-up alert screen with
pending results and the most recent completed test result with the date has resulted in a
significant reduction in duplicate orders for ordering of panels of tests for acute hepatitis
(17). A similar intervention to indicate to physicians that a recent urine culture was ordered,
is pending, or has been completed, along with the time and date, may lead to more
discerning utilization practices. A thoughtful design of CPOE systems coupled with the
implementation of diagnostic stewardship may play an important role in improving the
cost-effectiveness and quality of clinical patient care (18–20).
In this study, patients who had an index culture of a sample from an indwelling
catheter had a lower risk of having an inappropriate repeat urine culture than those
who had an index culture of clean-catch/straight catheterized and procedural speci-
mens. This finding could potentially be explained by our institutional intervention of
modifying urine testing order sets in the CPOE system, which was implemented in April
2016 (8). Urinalysis parameters of proteinuria and blood in the urine were excluded
from reflex urine culture testing, with the exception of testing for neutropenic patients
(8). This CPOE system-based intervention resulted in a significant reduction in the
number of indwelling catheterized urine cultures (75.6%) being performed compared
to the number of clean-catch urine cultures being performed (37.8%) (8). In our study,
we assigned the indwelling catheterized urine cultures to preintervention (January
2015 to April 2016) and postintervention (May 2016 to February 2018) periods, based
on our prior institutional intervention (8). We found that the proportion of all indwelling
catheterized urine cultures was significantly lower during the postintervention period
(3,217/16,308 [19.7%] versus 1,035/11,833 [8.7%]; P ⬍ 0.001). A similar finding was
observed in the proportion of inappropriate repeat indwelling catheterized urine
cultures in our cohort (219/760 [28.8%] versus 48/360 [13.3%]; P ⬍ 0.001).
A limitation of our study is that it is a retrospective study performed at a single,
academic hospital, which limits the generalizability of our results to other settings. We
did not include the data on antimicrobial use. The absence of chart review precluded
the evaluation of test indication, provider characteristic (e.g., level of training), and
longitudinal provider ordering behavior. To offset the lack of test indication by chart
review, we used a relatively conservative definition for what constitutes an inappro-
priate repeat urine culture to avoid overestimating its prevalence. We also performed
a sensitivity analysis and compared all inpatients who had repeat urine cultures
performed within less than 48 h, irrespective of their clinical appropriateness criteria,
and found similar results (data not shown). Although we used the Medicare Clinical

October 2019 Volume 57 Issue 10 e00910-19 jcm.asm.org 6

Repeat Urine Culture among Inpatients Journal of Clinical Microbiology

Laboratory Fee Schedule as a standardized method of reimbursement for urine culture,

this may not reflect the actual laboratory costs, which may vary across hospitals. We
were unable to assess the clinical consequences of reducing the number of inappro-
priate repeat urine cultures. However, previous studies postulated that elimination of
unnecessary duplicative testing is unlikely to result in clinically significant adverse
outcomes (2, 3, 11). The strengths of our study included a large sample size, a
comparison of ordering services, an assessment of the transition of care from emer-
gency department to inpatient services, and a measure of the potential financial impact
of duplicate testing on laboratories.
In conclusion, we found in a large academic medical center that over half of urine
cultures repeated on inpatients within 48 h were inappropriate. Notably, the diagnostic
gain of an inappropriate repeat urine culture performed within 48 h for detecting
bacteriuria was less than 5%. Further study is needed to evaluate provider-related
ordering behavior on urine culturing practice and the effect of repeat urine culture
testing on antimicrobial usage and clinical outcomes. Our findings highlight opportu-
nities for further optimization of CPOE systems for improving diagnostic and antimi-
crobial stewardship and patient care.

ACKNOWLEDGMENTS
We have no conflicts of interest to disclose.
This research was supported in part by the Division of Infectious Diseases at the
Washington University School of Medicine.

REFERENCES
1. Zhi M, Ding EL, Theisen-Toupal J, Whelan J, Arnaout R. 2013. The 11. Onderdonk AB, Winkelman JW, Orni᎑Wasserlauf R. 1996. Eliminating
landscape of inappropriate laboratory testing: a 15-year meta-analysis. unnecessary urine cultures to reduce costs. Lab Med 27:829 – 832.
PLoS One 8:e78962. https://doi.org/10.1371/journal.pone.0078962. https://doi.org/10.1093/labmed/27.12.829.
2. Bates DW, Boyle DL, Rittenberg E, Kuperman GJ, Ma’Luf N, Menkin V, 12. Nicolle LE, Gupta K, Bradley SF, Colgan R, DeMuri GP, Drekonja D, Eckert
Winkelman JW, Tanasijevic MJ. 1998. What proportion of common di- LO, Geerlings SE, Köves B, Hooton TM, Juthani-Mehta M, Knight SL, Saint
agnostic tests appear redundant? Am J Med 104:361–368. https://doi S, Schaeffer AJ, Trautner B, Wullt B, Siemieniuk R. 2019. Clinical practice
.org/10.1016/s0002-9343(98)00063-1. guideline for the management of asymptomatic bacteriuria: 2019 up-
3. Procop GW, Yerian LM, Wyllie R, Harrison AM, Kottke-Marchant K. date by the Infectious Diseases Society of America. Clin Infect Dis
2014. Duplicate laboratory test reduction using a clinical decision
68:e83– e110. https://doi.org/10.1093/cid/ciy1121.
support tool. Am J Clin Pathol 141:718 –723. https://doi.org/10.1309/
13. Gupta K, Hooton TM, Naber KG, Wullt B, Colgan R, Miller LG, Moran GJ,
AJCPOWHOIZBZ3FRW.
Nicolle LE, Raz R, Schaeffer AJ, Soper DE. 2011. International clinical
4. Wilson ML, Gaido L. 2004. Laboratory diagnosis of urinary tract infections
in adult patients. Clin Infect Dis 38:1150 –1158. https://doi.org/10.1086/ practice guidelines for the treatment of acute uncomplicated cystitis
383029. and pyelonephritis in women: a 2010 update by the Infectious Diseases
5. Gross PA, Patel B. 2007. Reducing antibiotic overuse: a call for a national Society of America and the European Society for Microbiology and
performance measure for not treating asymptomatic bacteriuria. Clin Infectious Diseases. Clin Infect Dis 52:e103– e120. https://doi.org/10
Infect Dis 45:1335–1337. https://doi.org/10.1086/522183. .1093/cid/ciq257.
6. Hooton TM, Bradley SF, Cardenas DD, Colgan R, Geerlings SE, Rice JC, 14. Huissoon AP, Carlton SA. 2002. Unnecessary repeat requesting of tests in
Saint S, Schaeffer AJ, Tambayh PA, Tenke P, Nicolle LE. 2010. Diagnosis, a university teaching hospital immunology laboratory: an audit. J Clin
prevention, and treatment of catheter-associated urinary tract infection Pathol 55:78. https://doi.org/10.1136/jcp.55.1.78-a.
in adults: 2009 international clinical practice guidelines from the Infec- 15. Kwok J, Jones B. 2005. Unnecessary repeat requesting of tests: an audit
tious Diseases Society of America. Clin Infect Dis 50:625– 663. https:// in a government hospital immunology laboratory. J Clin Pathol 58:
doi.org/10.1086/650482. 457– 462. https://doi.org/10.1136/jcp.2004.021691.
7. Carlson AL, Munigala S, Russo AJ, McMullen KM, Wood H, Jackups R, 16. Kim JY, Dzik WH, Dighe AS, Lewandrowski KB. 2011. Utilization manage-
Warren DK. 2017. Inpatient urine cultures are frequently performed ment in a large urban academic medical center: a 10-year experience. Am
without urinalysis or microscopy: findings from a large academic med- J Clin Pathol 135:108 –118. https://doi.org/10.1309/AJCP4GS7KSBDBACF.
ical center. Infect Control Hosp Epidemiol 38:455– 460. https://doi.org/ 17. Bridges SA, Papa L, Norris AE, Chase SK. 2014. Duplicated laboratory
10.1017/ice.2016.311. tests: evaluation of a computerized alert intervention abstract. J Healthc
8. Munigala S, Rojek R, Wood H, Yarbrough ML, Jackups RR, Burnham CD,
Qual 36:46 –53. https://doi.org/10.1111/j.1945-1474.2012.00219.x.
Warren DK. 2019. Effect of changing urine testing orderables and clini-
18. Morgan DJ, Malani P, Diekema DJ. 2017. Diagnostic stewardship—
cian order sets on inpatient urine culture testing: analysis from a large
leveraging the laboratory to improve antimicrobial use. JAMA 318:
academic medical center. Infect Control Hosp Epidemiol 40:281–286.
https://doi.org/10.1017/ice.2018.356. 607– 608. https://doi.org/10.1001/jama.2017.8531.
9. Simerville JA, Maxted WC, Pahira JJ. 2005. Urinalysis: a comprehensive 19. Sullivan KV, Morgan DJ, Leekha S. 2019. Use of diagnostic stewardship
review. Am Fam Physician 71:1153–1162. practices to improve urine culturing among SHEA Research Network
10. Centers for Medicare and Medicaid Services. 2017. Clinical laboratory hospitals. Infect Control Hosp Epidemiol 40:228 –231. https://doi.org/10
fee schedule. https://www.cms.gov/Medicare/Medicare-Fee-for-Service- .1017/ice.2018.325.
Payment/ClinicalLabFeeSched/Clinical-Laboratory-Fee-Schedule-Files- 20. Fakih MG, Advani SD, Vaughn VM. 2019. Diagnosis of urinary tract
Items/17CLAB.html?DLPage⫽1&DLEntries⫽100&DLSort⫽2&DLSortDir infections: need for a reflective rather than reflexive approach. Infect Con-
⫽descending. Accessed 15 December 2018. trol Hosp Epidemiol 40:834–835. https://doi.org/10.1017/ice.2019.98.

October 2019 Volume 57 Issue 10 e00910-19 jcm.asm.org 7