Abstract
Introduction
Urinary tract infections (UTIs) and sexually transmitted infections (STIs) can have overlapping signs, symptoms, and findings on urinalysis. Our objective was to determine if patient or provider demographics are associated with differences in the diagnosis and management of UTIs and STIs in the emergency department (ED).
Methods
We analyzed 38,062 ED patient encounters from a single healthcare system between April 18, 2014, and March 7, 2017. All encounters were women ≥18 years of age and not admitted to the hospital. We performed logistic regression using patient and provider demographics, laboratory testing results, ED triage data, and ED diagnoses.
Results
The patient's age, race, and marital status were not associated with having an ED UTI diagnosis with a urine culture ≥10,000 colony forming units (CFUs)/mL (vs. <10,000 CFUs/mL). Patient race and the sex of the ED provider were not associated with differences in empiric antibiotic treatment for gonorrhea and chlamydia during the ED encounter. Patient's race and the sex of the ED provider were also not associated with discordance between empiric antibiotic therapy given in the ED and the results of gonorrhea and chlamydia tests that resulted following the ED encounter.
Conclusion
In our multivariate analyses, we did not observe that the patient's race resulted in significant differences in the diagnosis of UTIs with bacteriuria ≥10,000 CFU/mL or differences in the empiric treatment of gonorrhea and chlamydia infections among those tested for the infection in the ED. The patient's age and marital status, but not the provider's sex, were significantly associated with differences in the management of gonorrhea and chlamydia.
1. Introduction
Rates of gonorrhea and chlamydia are rising in the United States, and an increasing number of sexually transmitted infections (STIs) are being diagnosed in the emergency department (ED) [1–4]. Approximately, 20% of Americans have an STI and new STIs result in $16 billion in direct medical care [5]. Managing STIs in the ED can be challenging because patients can be asymptomatic or present with varied signs and symptoms. STIs in women can be challenging to differentiate clinically from urinary tract infections as both can have overlapping signs, symptoms, and laboratory findings on urinalysis. Furthermore, the results of nucleic acid amplification tests (NAATs) for gonorrhea, chlamydia, and trichomonas may not be available to the clinician during the patient's ED encounter requiring providers to either empirically treat for STIs in the ED or wait for the results of diagnostic testing. If a woman goes untreated for an STI, the risks may include chronic pelvic pain, infertility, genital cancers, and an increased risk for HIV [2]. These risks must be balanced against the harms of administering unnecessary antibiotics.
Urinary tract infections (UTIs) are among the most common bacterial infections diagnosed in the emergency department (ED) [6]. An estimated one fifth of ambulatory medicine encounters involving UTI occurs in the ED [6]. The diagnosis of a bacterial UTI traditionally relies on the patient's signs, symptoms, or both, of a UTI plus a positive urine culture. However, urine culture results are unavailable during the clinical encounter, and ED clinicians are frequently noncompliant with evidence-based recommendations for UTI management [7]. Not all ED patients diagnosed with a UTI receive a urine culture. Patients at risk for sexually transmitted infections (STIs) and with concurrent genitourinary complaints do not always receive ED testing for STIs [8]. As a result, ED clinicians likely overdiagnose UTIs and underdiagnose STIs [8].
Racism, racial disparities, and implicit bias exist within our healthcare system, adversely affecting the quality of care [9, 10]. Previous studies have shown racial inequality in STI testing, although there are few studies related to STIs in the ED [11–15]. Little has been published on disparities in diagnosing and managing UTIs, particularly in the ED [16–18]. Our primary study objective was to determine if a patient's race is associated with differences in the diagnosis and management of genitourinary tract infections. Secondarily, we looked at other patient and provider demographics to determine if they were also associated with differences in the diagnosis and management of STIs and UTIs. Specifically, we looked for differences in bacteriuria in the urine culture among women diagnosed with a UTI, the empiric treatment of gonorrhea and chlamydia during the ED encounter, and the frequency with which there was discordance with gonorrhea and chlamydia test results and the administration of empiric antibiotic for those infections during the ED encounter.
2. Materials and Methods
The <removed for blinding> institutional review board approved the study with a waiver of informed consent. We utilized an existing data set of 75,000 ED patient encounters, but our analysis only included 38,254 women who were not admitted to the hospital. All patient encounters in the data set were at least 18 years of age and seen at university hospitals between April 18, 2014, and March 7, 2017. All patient encounters in the data set were tested for gonorrhea, chlamydia, or trichomonas or they had a urinalysis plus urine culture during their ED visit. Patients who received empiric antibiotics for STIs but did not receive STI testing were not included in the data set. Similarly, the data set did not include patients who received a urinalysis without a urine culture or STI testing. The data set contained retrospective data extracted from the electronic medical records by an institutional information technology team. Data from this data set have been published previously, including how the data set was created [19–28].
Women who received azithromycin and ceftriaxone in the ED or ceftriaxone in the ED, plus an outpatient prescription for doxycycline, were considered to be treated for gonorrhea and chlamydia. These were the management recommendations for gonorrhea and chlamydia when clinical care was administered [29]. Women not receiving those antibiotics during their ED encounter were considered not empirically treated for gonorrhea and chlamydia. We considered ED testing and treatment for gonorrhea and chlamydia to be concordant when (1) a patient was empirically treated for gonorrhea and chlamydia and had a positive NAAT for gonorrhea, chlamydia, or both or (2) a patient was not empirically treated for gonorrhea and chlamydia and they had both negative gonorrhea and chlamydia NAATs. We considered ED testing and treatment for gonorrhea and chlamydia to be discordant when (1) a patient was empirically treated for gonorrhea and chlamydia, but they were negative for both gonorrhea and chlamydia by NAAT or (2) a patient was not empirically treated for gonorrhea and chlamydia but had a positive NAAT for gonorrhea, chlamydia, or both. Indeterminant or erroneous NAAT results for gonorrhea and chlamydia were not included in the analysis.
A urinalysis was performed if any test from the urinalysis was reported. The urine source was categorized as clean catch/voided, “other” (i.e., bladder catheter, straight catheter, ileostomy, nephrostomy, suprapubic, or urostomy, etc.), or it was not documented. The urinalysis was categorized as follows: bacteria (0–4+), blood (0–3+), glucose (positive or negative), ketones (positive or negative), leukocyte esterase level (0–3+), mucous (0–4+), nitrite (positive or negative), pH [5–9], protein (positive or negative), red blood cells (RBCs) 0–101 cells/high powered field (cells/HPF), trichomonas (positive or not reported), white blood cell (WBC) clumps (positive or negative), WBCs 0–101 (cells/high powered field (HPF)), and yeast (present or none). The clinical laboratories reported urine RBCs and WBCs in various ways. The mean was used in the analysis if a range was reported. If >100 cells/HPF were reported, 101 cells/HPF were used for analysis. The urine culture was classified as ≥10,000 colony forming units (CFUs)/mL or <10,000 CFUs/mL. All laboratory testing and cultures were ordered at the treating clinician's discretion.
If any wet mount results were reported, women were categorized as having a vaginal wet mount. Vaginal wet mount clue cells, trichomonas, and yeast were either present or absent. Vaginal wet mount WBCs were categorized as ≤10 or 11–100 cells/HPF [24].
There were 1,913 women who tested positive for T. vaginalis, and this was by urinalysis 384/26,880 (1.4%), vaginal wet mount 1,354/15,572 (8.7%), and NAAT 418/4,923 encounters (8.5%); however, the women may have had more than 1 positive test during the encounter. ED clinicians do not use the urinalysis to diagnose T. vaginalis because of low sensitivity (but does have high specificity). The vaginal wet mount has moderate sensitivity and high specificity, while the NAAT has high sensitivity and specificity [30–32]. Both the urine microscopy and vaginal wet mount results are available to the clinician during the ED encounter, and to avoid multicollinearity in the multivariable analysis, we combined these T. vaginalis results into a single variable, indicating whether T. vaginalis infection was known during the ED encounter as follows: (1) positive for T. vaginalis during the ED encounter: positive result by urine microscopy, vaginal wet mount, or both; (2) negative for T. vaginalis infection during ED encounter: negative T. vaginalis by vaginal wet mount and negative by urine microscopy if the urinalysis was performed; and (3) vaginal wet mount not performed: vaginal wet mount not performed and urine microscopy negative for T. vaginalis if a urinalysis was performed. The T. vaginalis, N. gonorrhea, and C. trachomatous NAAT Aptima (Hologic) were considered separately because these test results did not result until after the ED visit. When examining women empirically treated for gonorrhea and chlamydia in the ED and for concordance and discordance between empiric antibiotics for gonorrhea and chlamydia during the ED encounter and the testing results, we only examined women who had both a positive or negative NAAT for gonorrhea and chlamydia.
The demographic variables used in the analysis included the patient's age (18–28, 29–39, 40–50, 51–61, and ≥62 years), race (Black/African American (n = 25,429), White (11,995), and Asian (n = 163) which we combined with “other” (n = 475) in our analysis), and marital status. We considered age a categorical variable given the nonlinear distribution of STIs across the age spectrum [27, 29]. There were 192 persons of unknown race and 131 persons with the unknown marital status. The sex and training level of the primary clinician who was principally caring for the patient were considered in our multivariable modeling as follows: (1) the resident: for any patient encounter involving a resident; (2) the attending: for any attending physician-only encounters; and (3) the advanced practice practitioner (APP): for any encounter involving an APP or an APP plus an attending physician. The principal clinician caring for the patient was considered to be a resident for encounters involving a resident, APP, and an attending physician because residents would occasionally sign out unfinished patients to an APP but APPs would not sign out cases to residents. Both APPs and residents would occasionally sign out unfinished patient encounters to attending physician or to another APP or resident, respectively.
The following triage and encounter data were included in the analysis if it was present for the clinical encounter: whether the patient has a primary care physician documented, the method of ED arrival, and the triage emergency severity index (ESI). Women were diagnosed with a UTI if they had an International Classification of Diseases (ICDs), Ninth Revision or Tenth Revision, code of N30.90, O86.22, N30.00, N30.91, N30, N30.0, N30.01, N30.9, O23.10, O86.20, N39.0, O08.83, O03.38, O04.88, O03.88, O86.2, O86.29, O23.40, 646.64, 599.0, 639.8, 646.60, 595.0, 595.9, or 595.89. Women were considered pregnant if they had a documented positive pregnancy test or an ICD Ninth Revision or Tenth Revision, code of O00.1, O00.8, O21.9, O00.90, Z32.01, O00, O20, V72.42, 643.93, or 633.90.
2.1. Statistical Analysis
Patient demographics, clinical characteristics, and provider characteristics were summarized as count (percentage) for categorical variables and median (interquartile range) for continuous variables. Univariate and multivariable logistic regression analyses were used to investigate the associations between our outcomes of interest and patient/provider characteristics. Variables for the regression analysis were chosen based on available patient and provider data (Table 1). Not all women undergoing STI testing had a urinalysis or a vaginal wet mount. However, we included these variables in our regression analyses because genitourinary tract infections can cause nonspecific inflammatory changes on both tests, which could have influenced clinician management decisions [8]. The analyses were conducted by BlueSky Statistics software v7.40 (BlueSky Statistics LLC). All tests were two sided, and p values <0.05 were considered significant.
Table 1.
Descriptive analysis of patient demographics and clinical characteristics for all encounters by patient's race.
| White (N = 11,995) | Asian/other race (N = 638) | Black/African American (N = 25,429) | Total (N = 38,062) | |
|---|---|---|---|---|
| Age (years) | ||||
| 18–28 | 2,845 (23.7%) | 277 (43.4%) | 13,244 (52.1%) | 16,366 (43.0%) |
| 29–39 | 2,025 (16.9%) | 165 (25.9%) | 5,898 (23.2%) | 8,088 (21.2%) |
| 40–50 | 1,434 (12.0%) | 80 (12.5%) | 2,484 (9.8%) | 3,998 (10.5%) |
| 51–61 | 1,343 (11.2%) | 57 (8.9%) | 1,446 (5.7%) | 2,846 (7.5%) |
| ≥62 | 4,348 (36.2%) | 59 (9.2%) | 2,357 (9.3%) | 6,764 (17.8%) |
| Marital status | ||||
| Missing | 45 | 8 | 47 | 100 |
| Single | 4,320 (36.2%) | 333 (52.9%) | 20,603 (81.2%) | 25,256 (66.5%) |
| Married/life partner | 4,102 (34.3%) | 247 (39.2%) | 2,546 (10.0%) | 6,895 (18.2%) |
| Separated/divorced | 1,448 (12.1%) | 25 (4.0%) | 1,257 (5.0%) | 2,730 (7.2%) |
| Widowed | 2,080 (17.4%) | 25 (4.0%) | 976 (3.8%) | 3,081 (8.1%) |
| Documented primary care physician | ||||
| No | 6,576 (54.8%) | 469 (73.5%) | 19,949 (78.4%) | 26,994 (70.9%) |
| Yes | 5,419 (45.2%) | 169 (26.5%) | 5,480 (21.6%) | 11,068 (29.1%) |
| Pregnant | ||||
| No | 11, 406 (95.1%) | 543 (85.1%) | 21,421 (84.2%) | 33,370 (87.7%) |
| Yes | 589 (4.9%) | 95 (14.9%) | 4,008 (15.8%) | 4,692 (12.3%) |
| Emergency severity index | ||||
| Missing | 207 | 32 | 1,189 | 1,428 |
| 1 and 2 | 1,063 (9.0%) | 46 (7.6%) | 1,104 (4.6%) | 2,213 (6.0%) |
| 3 | 9,111 (77.3%) | 448 (73.9%) | 17,996 (74.2%) | 27,555 (75.2%) |
| 4 and 5 | 1,614 (13.7%) | 112 (18.5%) | 5,140 (21.2%) | 6,866 (18.7%) |
| Mechanism of ED arrival | ||||
| Missing | 112 | 4 | 131 | 247 |
| Private vehicle | 9,194 (77.4%) | 534 (84.2%) | 21,034 (83.1%) | 30,762 (81.3%) |
| EMS/Police | 2,596 (21.8%) | 81 (12.8%) | 3,153 (12.5%) | 5,830 (15.4%) |
| Public transport/on foot | 93 (0.8%) | 19 (3.0%) | 1,111 (4.4%) | 1,223 (3.2%) |
| Sex of primary ED provider | ||||
| Missing | 741 | 24 | 203 | 968 |
| Female | 3,469 (30.8%) | 194 (31.6%) | 8,228 (32.6%) | 11,891 (32.1%) |
| Male | 7,785 (69.2%) | 420 (68.4%) | 16,998 (67.4%) | 25,203 (67.9%) |
| Training level of primary ED provider | ||||
| Missing | 84 | 13 | 348 | 445 |
| Attending only | 7690 (64.6%) | 338 (54.1%) | 11282 (45.0%) | 19310 (51.3%) |
| APP | 2499 (21.0%) | 119 (19.0%) | 5335 (21.3%) | 7953 (21.1%) |
| Attending + Resident | 1722 (14.5%) | 168 (26.9%) | 8464 (33.7%) | 10354 (27.5%) |
| Urinalysis performed | ||||
| Not performed | 748 (6.2%) | 56 (8.8%) | 2,716 (10.7%) | 3,520 (9.2%) |
| Performed | 11,247 (93.8%) | 582 (91.2%) | 22,713 (89.3%) | 34,542 (90.8%) |
| Source of urine sample | ||||
| Missing | 728 | 56 | 2,653 | 3,437 |
| Clean catch/void urine | 6616 (58.7%) | 309 (53.1%) | 8,838 (38.8%) | 15,763 (45.5%) |
| Not documented | 3,349 (29.7%) | 249 (42.8%) | 13,259 (58.2%) | 16,857 (48.7%) |
| Other | 1,302 (11.6%) | 24 (4.1%) | 679 (3.0%) | 2,005 (5.8%) |
| Bacteria, urine (0–4+) | ||||
| Missing | 2,882 | 178 | 8,145 | 11,205 |
| Median (Q1, Q3) | 1.0 (0.0, 2.0) | 1.0 (0.0, 2.0) | 1.0 (0.0, 2.0) | 1.0 (0.0, 2.0) |
| Blood, urine (0–3+) | ||||
| Missing | 985 | 67 | 3,177 | 4,229 |
| Median (Q1, Q3) | 0.0 (0.0, 2.0) | 0.0 (0.0, 2.0) | 0.0 (0.0, 1.0) | 0.0 (0.0, 2.0) |
| Glucose, urine | ||||
| Missing | 809 | 56 | 2,856 | 3,721 |
| Negative | 10,369 (92.7%) | 536 (92.1%) | 21,256 (94.2%) | 32,161 (93.7%) |
| Positive | 817 (7.3%) | 46 (7.9%) | 1,317 (5.8%) | 2,180 (6.3%) |
| Ketones, urine | ||||
| Missing | 818 | 56 | 2,882 | 3,756 |
| Negative | 9,339 (83.6%) | 483 (83.0%) | 18,475 (81.9%) | 28,297 (82.5%) |
| Positive | 1,838 (16.4%) | 99 (17.0%) | 4,072 (18.1%) | 6,009 (17.5%) |
| Leukocyte esterase, urine (0–3+) | ||||
| Missing | 1,049 | 66 | 3,185 | 4,300 |
| Median (Q1, Q3) | 1.0 (0.0, 3.0) | 1.0 (0.0, 2.0) | 1.0 (0.0, 2.0) | 1.0 (0.0, 2.0) |
| Mucous, urine (0–4+) | ||||
| Missing | 2,905 | 178 | 8,142 | 11,225 |
| Median (Q1, Q3) | 0.0 (0.0, 1.0) | 0.0 (0.0, 1.0) | 1.0 (0.0, 2.0) | 0.0 (0.0, 1.0) |
| Nitrite, urine | ||||
| Missing | 805 | 56 | 2,826 | 3,687 |
| Negative | 10,020 (89.5%) | 546 (93.8%) | 21,102 (93.4%) | 31,668 (92.1%) |
| Positive | 1,170 (10.5%) | 36 (6.2%) | 1,501 (6.6%) | 2,707 (7.9%) |
| Protein, urine | ||||
| Missing | 820 | 57 | 2,867 | 3,744 |
| Negative | 6,878 (61.5%) | 383 (65.9%) | 14,734 (65.3%) | 21,995 (64.1%) |
| Positive | 4,297 (38.5%) | 198 (34.1%) | 7,828 (34.7%) | 12,323 (35.9%) |
| RBCs, urine (0–101) | ||||
| Missing | 2,910 | 180 | 8,136 | 11,226 |
| Median (Q1, Q3) | 2.5 (2.5, 12.5) | 2.5 (2.5, 12.5) | 2.5 (2.0, 12.5) | 2.5 (2.5, 12.5) |
| Trichomonas, urine | ||||
| Missing | 2,917 | 178 | 8,209 | 11,304 |
| Not present | 9033 (99.5%) | 456 (99.1%) | 16886 (98.1%) | 26375 (98.6%) |
| Present | 45 (0.5%) | 4 (0.9%) | 334 (1.9%) | 383 (1.4%) |
| WBC clumps, urine | ||||
| Missing | 3,026 | 182 | 8332 | 11,540 |
| None | 8, 121 (90.5%) | 425 (93.2%) | 15,817 (92.5%) | 24,363 (91.9%) |
| Present | 848 (9.5%) | 31 (6.8%) | 1,280 (7.5%) | 2,159 (8.1%) |
| WBC, urine (0–101) | ||||
| Missing | 2,969 | 180 | 8,120 | 11,269 |
| Median (Q1, Q3) | 12.5 (2.5, 36.0) | 8.0 (2.5, 24.8) | 8.0 (2.5, 29.0) | 12.5 (2.5, 36.0) |
| Yeast, urine | ||||
| Missing | 3,025 | 180 | 8,238 | 11,443 |
| None | 8,635 (96.3%) | 446 (97.4%) | 16,706 (97.2%) | 25,787 (96.9%) |
| Present | 335 (3.7%) | 12 (2.6%) | 485 (2.8%) | 832 (3.1%) |
| Clue cells, wet mount | ||||
| No result | 10,577 (88.2%) | 426 (66.8%) | 11,358 (44.7%) | 22,361 (58.7%) |
| None | 1,038 (8.7%) | 152 (23.8%) | 7,603 (29.9%) | 8,793 (23.1%) |
| Present | 380 (3.2%) | 60 (9.4%) | 6,468 (25.4%) | 6,908 (18.1%) |
| WBC, wet mount | ||||
| No result | 10,509 (87.6%) | 423 (66.3%) | 11,032 (43.4%) | 21,964 (57.7%) |
| ≤10 | 1,050 (8.8%) | 147 (23.0%) | 9,632 (37.9%) | 10,829 (28.5%) |
| 11–100 | 436 (3.6%) | 68 (10.7%) | 4,765 (18.7%) | 5,269 (13.8%) |
| Yeast, wet mount | ||||
| No result | 10,587 (88.3%) | 426 (66.8%) | 11,549 (45.4%) | 22,562 (59.3%) |
| None | 1,355 (11.3%) | 195 (30.6%) | 12,925 (50.8%) | 14,475 (38.0%) |
| Present | 53 (0.4%) | 17 (2.7%) | 955 (3.8%) | 1,025 (2.7%) |
| Trichomonas, wet mount | ||||
| No result | 10,588 (88.3%) | 428 (67.1%) | 11,539 (45.4%) | 22,555 (59.3%) |
| None | 1,361 (11.3%) | 203 (31.8%) | 12,596 (49.5%) | 14,160 (37.2%) |
| Present | 46 (0.4%) | 7 (1.1%) | 1,294 (5.1%) | 1,347 (3.5%) |
| Gonorrhea, NAAT | ||||
| No test result | 10,489 (87.4%) | 423 (66.3%) | 10,923 (43.0%) | 21,835 (57.4%) |
| Negative | 1,489 (12.4%) | 211 (33.1%) | 14,051 (55.3%) | 15,751 (41.4%) |
| Positive | 17 (0.1%) | 4 (0.6%) | 455 (1.8%) | 476 (1.3%) |
| Chlamydia, NAAAT | ||||
| No test result | 10,489 (87.4%) | 422 (66.1%) | 10,931 (43.0%) | 21,842 (57.4%) |
| Negative | 1,426 (11.9%) | 203 (31.8%) | 13,291 (52.3%) | 14,920 (39.2%) |
| Positive | 80 (0.7%) | 13 (2.0%) | 1,207 (4.7%) | 1,300 (3.4%) |
| Trichomonas, NAAT | ||||
| No test result | 11,539 (96.2%) | 569 (89.2%) | 21,053 (82.8%) | 33,161 (87.1%) |
| Negative | 439 (3.7%) | 66 (10.3%) | 3,978 (15.6%) | 4,483 (11.8%) |
| Positive | 17 (0.1%) | 3 (0.5%) | 398 (1.6%) | 418 (1.1%) |
| Diagnosed with a urinary tract infection (UTI in the ED) | ||||
| No | 8,384 (69.9%) | 496 (77.7%) | 20,895 (82.2%) | 29,775 (78.2%) |
| Yes | 3,611 (30.1%) | 142 (22.3%) | 4,534 (17.8%) | 8,287 (21.8%) |
^There were 192 encounters with no race provided.
3. Results
38,254/75,000 (51%) encounters in the data set met our inclusion criteria of being a woman who was not admitted to the hospital and were included in the analysis. Descriptive statistics are summarized in Table 1. Within our cohort, UTI (n = 8,330) was diagnosed in 1,437 (17.3%) women with no urine culture performed, 3,915 (47.0%) with <10,000 CFU/mL bacterial growth, and 2,978 (35.8%) with ≥10,000 CFU/mL bacterial growth. 6,893 (82.7%) women diagnosed with a UTI received a urine culture. A total of 22,668 women received a urine culture, of which 5,655 (24.9%) had ≥10,000 CFU/mL bacterial growth. Among those women with ≥10,000 CFU/mL bacterial growth on urine culture, 2,978 (52.7%) were diagnosed with a UTI.
Among those being diagnosed with a UTI in the ED and with a documented patient race, they were Black/African American (n = 4,534), White (n = 3,611), and Asian/others (n = 142) (Table 2). The rates of being diagnosed with a UTI and having a urine culture ≥10,000 CFU/mL were 1552 (34.3%) for Black/African American, 1355 (37.5%) for White, and 53 (37.3%) for Asian/others.
Table 2.
Urine culture results among women diagnosed with a UTI in the ED.
| White (N = 3,611) | Asian/other race (N = 142) | Black/African American (N = 4,534) | Total (N = 8,287) | p value | |
|---|---|---|---|---|---|
| Colony forming units (CFU)/mL of bacteria in urine culture (binary) | <0.001 [1] | ||||
| (i) <10,000 | 1,857 (51.4%) | 70 (49.3%) | 1,967 (43.4%) | 3,894 (47.0%) | |
| (ii) ≥10,000 | 1,355 (37.5%) | 53 (37.3%) | 1,552 (34.2%) | 2,960 (35.7%) | |
| (iii) No urine culture | 399 (11.0%) | 19 (13.4%) | 1,015 (22.4%) | 1,433 (17.3%) |
There were 16,317 patient encounters with positive or negative test results for gonorrhea, chlamydia, or both. Among these, they were Black/African American (n = 14,522; 89.0%), White (1,509; 9.2%), Asian/others (n = 217; 1.3%), and unknown race (69; 4.2%). Their marital status was reported as married/life partner (n = 1,432; 8.8%), single (14,210; 87.1%), widowed (72; 0.4%), divorced/separated (556; 3.4%), and unknown marital status 47 (0.3%). There were 1,300 (8.0%) who tested positive for chlamydia, 473 (2.9%) who tested positive for gonorrhea, and 149 (0.9%) who tested positive for both gonorrhea and chlamydia. There were 1,516/14,481 (10.5%) tested Black/African American women infected with gonorrhea, chlamydia, or both compared with 90/1,503 (6.0%) of White women (p < 0.001).
There were 16,199 encounters with both NAAT results being positive or negative for gonorrhea and chlamydia (Table 3). Of these, 3,068 (18.9%) were given empiric antibiotics for gonorrhea and chlamydia during the ED encounter compared to 13,131 (81.1%) not given empiric antibiotics for gonorrhea and chlamydia. Women had concordance with being given empiric antibiotics in the ED for gonorrhea and chlamydia and their NAAT results for one or both of those infections (n = 12,526; 77.3%) compared with discordance between empiric antibiotics in the ED and their NAAT result (n = 3,673; 22.7%).
Table 3.
Empiric antibiotic treatment for gonorrhea and chlamydia during the ED encounter and gonorrhea and chlamydia testing result.
| White (N = 1,503) | Asian/other race (N = 214) | Black/African American (N = 14,482) | Total (N = 16,199) | |
|---|---|---|---|---|
| Negative for both gonorrhea and chlamydia, and no empiric antibiotics given for gonorrhea and chlamydia during ED encounter (concordance) | 1,237 (10.3%) | 171 (1.4%) | 10,610 (88.3%) | 12,018 |
| Negative test for both gonorrhea and chlamydia, and empiric antibiotics given for gonorrhea and chlamydia during ED encounter (discordance) | 176 (6.9%) | 28 (1.1%) | 2,356 (92.0%) | 2,560 |
| Positive test for gonorrhea, chlamydia, or both, and no empiric antibiotics given for gonorrhea and chlamydia during ED encounter (discordance) | 62 (5.6%) | 11 (1.0%) | 1,040 (93.4%) | 1,113 |
| Positive test for gonorrhea, chlamydia, or both, and empiric antibiotics given for gonorrhea and chlamydia during ED encounter (concordance) | 28 (5.5%) | 4 (0.8%) | 476 (93.7%) | 508 |
3.1. UTI Diagnosis and ≥10,000 CFU/mL Bacteriuria vs. <10,000 CFU/mL Bacteriuria on Urine Culture
Our univariate analysis is available as Supplementary file 1. The patient's race was not associated with a UTI diagnosis and having a urine culture grow ≥10,000 CFU/mL bacterial growth compared to being diagnosed with a UTI and having <10,000 CFU/mL (Table 4). The following were associated with being diagnosed with a UTI and having a urine culture ≥10,000 CFU/mL bacterial growth when compared to being diagnosed with a UTI and having <10,000 CFU/mL bacterial growth: ESI 4-5 (vs. 1-2); the patient was seen by a resident and attending (vs. an APP); urine source was clean catch/voided and “other” (vs. not documented); on urinalysis: higher urine bacteria, more blood, fewer ketones, less mucus, nitrite positive, more WBC clumps, more urine WBCs, and on vaginal wet mount: having ≤10 WBCs/HPF (vs. 11–100 WBCs/HPF); negative T. vaginalis NAAT (vs. no test); negative for T. vaginalis infection during ED encounter (vs. positive for T. vaginalis during the ED encounter); and a negative T. vaginalis NAAT (vs. no test) (vs. 11–100 cells/HPF) (p < 0.03 for all) (Table 4).
Table 4.
Multivariate OR of being diagnosed with a UTI and having a positive urine culture (vs. negative urine culture).
| Variables | Comparison groups | Reference levels | OR (95% CI) | p value |
|---|---|---|---|---|
| Age (years) | 18–28 | ≥62 | 0.93 (0.74, 1.18) | 0.56 |
| 29–39 | ≥62 | 0.89 (0.71, 1.13) | 0.35 | |
| 40–50 | ≥62 | 0.86 (0.67, 1.10) | 0.22 | |
| 51–61 | ≥62 | 0.94 (0.74, 1.20) | 0.64 | |
|
| ||||
| Race | Black/African American | Asian/other | 1.09 (0.69, 1.73) | 0.70 |
| White | Asian/other | 0.84 (0.53, 1.33) | 0.44 | |
|
| ||||
| Marital status | Separated/divorced | Married/life partner | 0.88 (0.69, 1.12) | 0.31 |
| Single | Married/life partner | 1.01 (0.84, 1.21) | 0.91 | |
| Widowed | Married/life partner | 1.10 (0.87, 1.38) | 0.42 | |
|
| ||||
| Documented primary care physician | Yes | No | 0.89 (0.78, 1.02) | 0.08 |
|
| ||||
| Pregnant | Yes | No | 0.80 (0.57, 1.13) | 0.21 |
|
| ||||
| Emergency severity index | 3 | 1-2 | 1.04 (0.79, 1.40) | 0.77 |
| 4-5 | 1-2 | 1.44 (1.05, 1.98) | 0.02 | |
|
| ||||
| Mechanism of ED arrival | Private vehicle | EMS/police | 1.19 (0.98, 1.45) | 0.09 |
| Public transport/on foot | EMS/police | 1.52 (0.95, 2.41) | 0.08 | |
|
| ||||
| Sex of primary ED provider | Male | Female | 1.0257 (0.89, 1.18) | 0.72 |
|
| ||||
| Training level of primary ED provider | Attending only | APP | 1.08 (0.92, 1.26) | 0.35 |
| Attending plus resident | APP | 1.27 (1.05, 1.54) | 0.02 | |
|
| ||||
| Source of urine sample | Not documented | Clean catch/voided | 0.62 (0.54, 0.72) | <0.001 |
| Other | Clean catch/voided | 1.38 (1.08, 1.78) | 0.01 | |
|
| ||||
| Bacteria (urinalysis) | 1-unit increase | 1.31 (1.25, 1.38) | <0.001 | |
|
| ||||
| Blood (urinalysis) | 1-unit increase | 1.24 (1.15, 1.32) | <0.001 | |
|
| ||||
| Glucose (urinalysis) | Positive | Negative | 0.97 (0.77, 1.23) | 0.80 |
|
| ||||
| Ketones (urinalysis) | Positive | Negative | 0.82 (0.68, 0.98) | 0.03 |
|
| ||||
| Leukocyte esterase (urinalysis) | 1-unit increase | 0.99 (0.93, 1.07) | 0.89 | |
|
| ||||
| Mucous (urinalysis) | 1-unit increase | 0.89 (0.84, 0.95) | <0.001 | |
|
| ||||
| Nitrite (urinalysis) | Positive | Negative | 6.42 (5.40, 7.67) | <0.001 |
|
| ||||
| Protein (urinalysis) | Positive | Negative | 1.09 (0.95, 1.25) | 0.23 |
|
| ||||
| RBCs (urinalysis) | 1-unit increase | 1.00 (1.00, 1.00) | 0.12 | |
|
| ||||
| WBC clumps (urinalysis) | Present | Negative | 1.50 (1.26, 1.78) | <0.001 |
|
| ||||
| WBCs (urinalysis) | 1-unit increase | 1.02 (1.01, 1.02) | <0.001 | |
|
| ||||
| Yeast (urinalysis) | Present | Absent | 0.74 (0.53, 1.03) | 0.08 |
|
| ||||
| Clue cells (vaginal wet mount) | Not performed | Negative | 0.32 (0.02, 3.58) | 0.37 |
| Present | Negative | 0.81 (0.56, 1.16) | 0.25 | |
|
| ||||
| WBC (vaginal wet mount) (cells/HPF) | 11–100 | ≤10 | 0.58 (0.40, 0.82) | 0.002 |
| Not performed | ≤10 | 1.29 (0.28, 8.19) | 0.77 | |
|
| ||||
| Yeast (vaginal wet mount) | Not performed | Negative | 1.11 (0.13, 9.70) | 0.93 |
| Present | Negative | 0.78 (0.37, 1.56) | 0.50 | |
|
| ||||
| T. vaginalis NAAT | No test | Negative | 0.61 (0.42, 0.90) | 0.01 |
| Positive | Negative | 0.60 (0.21, 1.55) | 0.31 | |
|
| ||||
| Positive for gonorrhea, chlamydia, or both; negative for both gonorrhea and chlamydia; or not tested for gonorrhea and chlamydia | Not tested for gonorrhea and chlamydia | Negative for both gonorrhea and chlamydia | 1.03 (0.59, 1.77) | 0.93 |
| Positive for gonorrhea, chlamydia, or both | Negative for both gonorrhea and chlamydia | 1.06 (0.65, 1.73) | 0.81 | |
|
| ||||
| Trichomonas found during the ED encounter | Positive for T. vaginalis during the ED encounter | Negative for T. vaginalis infection during ED encounter | 0.41 (0.20, 0.78) | 0.009 |
| Vaginal wet mount not performed | Negative for T. vaginalis infection during ED encounter | 2.16 (0.69, 7.31) | 0.20 | |
3.2. Empiric Treatment for Gonorrhea and Chlamydia in Those Tested for the Diseases
Our univariate analysis is available as Supplementary file 2. The following were significantly associated with being empirically treated for gonorrhea and chlamydia in the ED (vs. not being empirically treated in the ED): age 18–28, 29–39, 40–50, and 51–61 years (vs. ≥62 years); separated/divorced (vs. married/life partner); single (vs. married/life partner); not being pregnant (vs. pregnant); APP (vs. attending only); urine source clean catch/voided (vs. not documented); on urinalysis: less blood urine, higher leukocyte esterase, lower RBCs, and no yeast present (vs. present); and on vaginal wet mount: clue cells negative (vs. not performed), WBC 11–100 cells/HPF (vs. <10 cells/HPF); on NAAT: no gonorrhea test result and a positive result (vs. negative), trichomonas positive (vs. negative); not diagnosed with a UTI in the ED; and positive for T. vaginalis during the ED encounter (vs. negative for T. vaginalis infection during ED encounter) (p < 0.05 for all) (Table 5). The patient's race and the sex of the primary ED provider were both not significant.
Table 5.
Multivariable logistic regression for associations between empiric treatment for gonorrhea and chlamydia (vs. no treatment in the ED) among those tested.
| Variables | Comparison groups | Reference levels | OR (95% CI) | p value |
|---|---|---|---|---|
| Age (years) | 18–28 | ≥62 | 3.96 (1.51, 13.65) | 0.01 |
| 29–39 | ≥62 | 3.81 (1.45, 13.15) | 0.01 | |
| 40–50 | ≥62 | 3.86 (1.45, 13.40) | 0.01 | |
| 51–61 | ≥62 | 4.40 (1.60, 15.65) | 0.009 | |
|
| ||||
| Race | Black/African American | Asian/other | 1.85 (1.00, 3.75) | 0.07 |
| White | Asian/other | 1.598 (0.84, 3.32) | 0.18 | |
|
| ||||
| Marital status | Separated/divorced | Married/life partner | 1.49 (1.02, 2.18) | 0.04 |
| Single | Married/life partner | 1.60 (1.26, 2.06) | <0.001 | |
| Widowed | Married/life partner | 0.71 (0.16, 2.20) | 0.59 | |
|
| ||||
| Documented primary care physician | Yes | No | 0.89 (0.76, 1.04) | 0.14 |
|
| ||||
| Pregnant | Yes | No | 0.16 (0.12, 0.19) | <0.001 |
|
| ||||
| Emergency severity index | 3 | 1-2 | 0.95 (0.63, 1.46) | 0.80 |
| 4-5 | 1-2 | 1.07 (0.70, 1.66) | 0.77 | |
|
| ||||
| Mechanism of ED arrival | Private vehicle | EMS/police | 1.10 (0.87, 1.42) | 0.42 |
| Public transport/on foot | EMS/police | 1.06 (0.75, 1.49) | 0.73 | |
|
| ||||
| Sex of primary ED provider | Male | Female | 1.10 (0.98, 1.25) | 0.12 |
|
| ||||
| Training level of primary ED provider | Attending only | APP | 0.76 (0.65, 0.89) | <0.001 |
| Attending plus resident | APP | 0.99 (0.85, 1.16) | 0.94 | |
|
| ||||
| Source of urine sample | Not documented | Clean catch/voided | 0.72 (0.56, 0.91) | 0.007 |
| Other | Clean catch/voided | 0.77 (0.27, 1.84) | 0.58 | |
|
| ||||
| Bacteria (urinalysis) | 1-unit increase | 1.02 (0.97, 1.08) | 0.43 | |
|
| ||||
| Blood (urinalysis) | 1-unit increase | 0.84 (0.79, 0.89) | <0.001 | |
|
| ||||
| Glucose (urinalysis) | Positive | Negative | 0.86 (0.63, 1.15) | 0.32 |
|
| ||||
| Ketones (urinalysis) | Positive | Negative | 1.05 (0.90, 1.23) | 0.54 |
|
| ||||
| Leukocyte esterase (urinalysis) | 1-Unit increase | 1.17 (1.10, 1.25) | <0.001 | |
|
| ||||
| Mucous (urinalysis) | 1-Unit increase | 1.00 (0.96, 1.05) | 0.83 | |
|
| ||||
| Nitrite (urinalysis) | Positive | Negative | 1.06 (0.80, 1.39) | 0.68 |
|
| ||||
| Protein (urinalysis) | Positive | Negative | 0.95 (0.83, 1.08) | 0.45 |
|
| ||||
| RBCs (urinalysis) | 1-unit increase | 0.99 (0.99, 1.00) | <0.001 | |
|
| ||||
| WBC clumps (urinalysis) | Present | Negative | 1.19 (0.90, 1.56) | 0.21 |
|
| ||||
| WBCs (urinalysis) | 1-unit increase | 1.00 (1.00, 1.01) | 0.12 | |
|
| ||||
| Yeast (urinalysis) | Present | Absent | 0.43 (0.26, 0.66) | <0.001 |
|
| ||||
| CFU/mL on urine culture | ≥10,000 | <10,000 | 0.86 (0.66, 1.11) | 0.25 |
| No urine culture done | <10,000 | 1.24 (0.99, 1.57) | 0.07 | |
|
| ||||
| Clue cells (vaginal wet mount) | Not performed | Negative | 0.35 (0.12, 0.97) | 0.05 |
| Present | Negative | 0.98 (0.87, 1.10) | 0.74 | |
|
| ||||
| WBC (vaginal wet mount) (cells/HPF) | 11–100 | ≤10 | 1.38 (1.22, 1.56) | <0.001 |
| Not performed | ≤10 | 1.09 (0.51, 2.48) | 0.83 | |
|
| ||||
| Yeast (vaginal wet mount) | Not performed | Negative | 2.01 (0.69, 6.27) | 0.21 |
| Present | Negative | 0.79 (0.62, 1.01) | 0.06 | |
|
| ||||
| N. gonorrhea NAAT | No test result | Negative | 3.70 (0.97, 13.81) | 0.048 |
| Positive | Negative | 1.95 (1.50, 2.52) | <0.001 | |
|
| ||||
| C. trichomonas NAAT | No test result | Negative | 1.55 (0.46, 4.50) | 0.44 |
| Positive | Negative | 1.49 (1.24, 1.78) | <0.001 | |
|
| ||||
| T. vaginalis NAAT | No test result | Negative | 1.10 (0.95, 1.27) | 0.22 |
| Positive | Negative | 1.25 (0.89, 1.74) | 0.19 | |
|
| ||||
| Diagnosed with a UTI in the ED | Yes | No | 0.76 (0.65, 0.90) | 0.001 |
|
| ||||
| Trichomonas found during the ED encounter | Positive for T. vaginalis during the ED encounter | Negative for T. vaginalis infection during ED encounter | 3.88 (3.29, 4.58) | <0.001 |
| Vaginal wet mount not performed | Negative for T. vaginalis infection during ED encounter | 1.27 (0.49, 3.20) | 0.62 | |
3.3. Discordance between Testing Results for Gonorrhea and Chlamydia and Empiric Treatment in the ED
The following were significantly associated with being empirically treated for gonorrhea and chlamydia in the ED if testing negative or not being treated for gonorrhea or chlamydia and testing positive for those infections (vs. not being empirically treated for gonorrhea and chlamydia in the ED if testing negative or empirically treated for gonorrhea or chlamydia and testing positive for those infections): age ≥62 years (vs. 18–28, 29–39, 40–50, and 51–61 years); married/life partner (vs. separated/divorced); married/life partner (vs. single); pregnant (vs. not); urine source not documented (vs. clean catch/voided); on urinalysis: higher blood, lower leukocyte esterase, and yeast present (vs. not); urine culture <10,000 CFU/mL (vs. not urine culture done); on vaginal wet mount: clue cell negative (vs. positive), WBC ≤10 WBC/HPF (vs. 11–100 WBC/HPF), yeast preset (vs. absent); and T. vaginalis NAAT negative (vs. positive) (p < 0.02 for all) (Table 6). Neither the patient's race, sex of the primary ED provider, nor the training level of the primary ED provider was associated with our outcome (Table 6).
Table 6.
Multivariable logistic regression for associations between discordance between gonorrhea and chlamydia test results and empiric ED treatment (vs. concordance).
| Variables | Comparison groups | Reference levels | OR (95% CI) | p value |
|---|---|---|---|---|
| Age (years) | 18–28 | ≥62 | 0.19 (0.06, 0.49) | 0.002 |
| 29–39 | ≥62 | 0.25 (0.07, 0.65) | 0.01 | |
| 40–50 | ≥62 | 0.27 (0.08, 0.71) | 0.02 | |
| 51–61 | ≥62 | 0.24 (0.07, 0.65) | 0.01 | |
|
| ||||
| Race | Black/African American | Asian/other | 0.59 (0.32, 1.00) | 0.06 |
| White | Asian/other | 0.75 (0.40, 1.31) | 0.33 | |
|
| ||||
| Marital status | Separated/divorced | Married/life partner | 0.57 (0.41, 0.81) | 0.002 |
| Single | Married/life partner | 0.65 (0.52, 0.81) | <0.001 | |
| Widowed | Married/life partner | 2.5 (0.70, 16.30) | 0.23 | |
|
| ||||
| Documented primary care physician | Yes | No | 1.09 (0.95, 1.25) | 0.24 |
|
| ||||
| Pregnant | Yes | No | 2.66 (2.30, 3.10) | <0.001 |
|
| ||||
| Emergency severity index | 3 | 1-2 | 0.96 (0.65, 1.39) | 0.84 |
| 4-5 | 1-2 | 0.94 (0.63, 1.38) | 0.77 | |
|
| ||||
| Mechanism of ED arrival | Private vehicle | EMS/police | 1.08 (0.88, 1.33) | 0.45 |
| Public transport/on foot | EMS/police | 1.11 (0.83, 1.50) | 0.48 | |
|
| ||||
| Sex of primary ED provider | Male | Female | 0.99 (0.88, 1.10) | 0.82 |
|
| ||||
| Training level of primary ED provider | Attending only | APP | 1.05 (0.91, 1.20) | 0.53 |
| Attending plus resident | APP | 0.91 (0.79, 1.05) | 0.21 | |
|
| ||||
| Source of urine sample | Not documented | Clean catch/voided | 1.29 (1.04, 1.61) | 0.02 |
| Other | Clean catch/voided | 1.27 (0.57, 3.24) | 0.58 | |
|
| ||||
| Bacteria (urinalysis) | 1-unit increase | 1.04 (0.99, 1.09) | 0.16 | |
|
| ||||
| Blood (urinalysis) | 1-unit increase | 1.15 (1.09, 1.22) | <0.001 | |
|
| ||||
| Glucose (urinalysis) | Positive | Negative | 1.27 (0.97, 1.69) | 0.09 |
|
| ||||
| Ketones (urinalysis) | Positive | Negative | 1.04 (0.91, 1.20) | 0.57 |
|
| ||||
| Leukocyte esterase (urinalysis) | 1-unit increase | 0.84 (0.79, 0.89) | <0.001 | |
|
| ||||
| Mucous (urinalysis) | 1-unit increase | 1.00 (0.96, 1.04) | 0.98 | |
|
| ||||
| Nitrite (urinalysis) | Positive | Negative | 0.85 (0.67, 1.09) | 0.19 |
|
| ||||
| Protein (urinalysis) | Positive | Negative | 0.98 (0.87, 1.10) | 0.77 |
|
| ||||
| RBCs (urinalysis) | 1-unit increase | 1.00 (1.00, 1.00) | 0.19 | |
|
| ||||
| WBC clumps (urinalysis) | Present | Negative | 0.91 (0.71, 1.17) | 0.47 |
|
| ||||
| WBCs (urinalysis) | 1-unit increase | 1.00 (0.9964, 1.0016) | 0.46 | |
|
| ||||
| Yeast (urinalysis) | Present | Absent | 1.77 (1.22, 2.63) | 0.004 |
|
| ||||
| CFU/mL on urine culture | ≥10,000 | <10,000 | 1.11 (0.88, 1.40) | 0.38 |
| No urine culture done | <10,000 | 0.77 (0.62, 0.94) | 0.01 | |
|
| ||||
| Clue cells (vaginal wet mount) | Not performed | Negative | 1.86 (0.75, 4.64) | 0.18 |
| Present | Negative | 0.83 (0.75, 0.93) | <0.001 | |
|
| ||||
| WBC (vaginal wet mount) (cells/HPF) | 11–100 | ≤10 | 0.71 (0.64, 0.79) | <0.001 |
| Not performed | ≤10 | 1.00 (0.50, 1.91) | >0.99 | |
|
| ||||
| Yeast (vaginal wet mount) | Not performed | Negative | 0.77 (0.28, 2.04) | 0.60 |
| Present | Negative | 1.38 (1.11, 1.7) | 0.004 | |
|
| ||||
| T. vaginalis NAAT | No test result | Negative | 1.02 (0.90, 1.16) | 0.78 |
| Positive | Negative | 0.60 (0.45, 0.80) | <0.001 | |
|
| ||||
| Diagnosed with a UTI in the ED | Yes | No | 1.11 (0.96, 1.29) | 0.16 |
|
| ||||
| Trichomonas found during the ED encounter | Vaginal wet mount not performed | Negative for T. vaginalis infection during ED encounter | 1.55 (0.64, 3.79) | 0.33 |
| Positive for T. vaginalis during the ED encounter | Negative for T. vaginalis infection during ED encounter | 0.63 (0.26, 1.52) | 0.30 | |
4. Discussion
Our study examined whether the patient's race was associated with differences in diagnosing and managing genitourinary tract infections. We found that patient's race was not associated with differences in bacteriuria on urine culture among women diagnosed with a UTI. Our findings concord with a survey of implicit bias among pediatricians and found no significant relationships for treating UTI in Black patients [10]. Another study found no significant differences between ED patients receiving inappropriate treatment for UTIs when examining age, patient's race, and the professional level of training by the treating clinician [7].
We also found that patient's race was not associated with the empiric treatment of gonorrhea and chlamydia in the ED nor concordance or discordance between empiric antibiotic treatment for gonorrhea and chlamydia in the ED and resultant gonorrhea and chlamydia NAAT result.
Our study also examined whether other patient and provider demographics were associated with our outcomes of interest. We found that neither patient's age, marital status, nor the pregnancy status was associated with ≥10,000 CFU/mL of bacteriuria on urine culture in women diagnosed with UTI. We found that patient encounters principally involving a resident plus attending, compared with encounters involving APPs, were more likely to have ≥10,000 CFU/mL on urine culture in patients diagnosed with a UTI. These findings may reflect selection bias as the APPs tend to care for less severely ill patients in the ED compared with residents.
Black/African American women are significantly more likely to test positive for gonorrhea, chlamydia, and trichomonas than other races [29]. However, when accounting for other triage, demographic, and laboratory factors, patient's race was not significantly associated with higher rates of empiric antibiotic treatment for gonorrhea and chlamydia in the ED. Patient's race was also not associated with discordance between empiric antibiotics for gonorrhea and chlamydia administered in the ED and the NAAT testing results. Previously, it was shown that non-Hispanic race and ethnicity, treatment by a nurse practitioner, and older age were associated with the empiric treatment of STIs [33]. In addition, the empiric treatment of STIs in a pediatric ED was more common among non-Hispanic White patients than among non-Hispanic Black patients despite lower disease prevalence in the non-Hispanic White population [33]. A survey of primary care physicians showed that screening for STIs in asymptomatic adolescent women was significantly more likely if the practitioner was a woman and the practice had more Black/African American patients [34]. Other studies have shown that race was not a significant predictor of undertreatment for STIs, but Black patients were significantly more likely to be overtreated and to receive nonrecommended treatments for STIs than White patients [35, 36].
Younger age, being single (compared to married/life partner), and having higher vaginal WBCs on the wet mount were significantly associated with receiving empiric antibiotics for gonorrhea and chlamydia, which have also been reported previously [24, 27, 29, 37, 38]. ED clinicians were more likely to empirically treat gonorrhea and chlamydia in the ED when trichomonas was identified during the ED encounter. Patient encounters involving an APP (vs. attending only) were significantly more likely to treat for gonorrhea and chlamydia empirically, which may reflect the lower acuity patients that the APPs were more likely to encounter in a “fast track” area of the ED. However, our regression models did account for ESI. The sex of the primary ED provider was not associated with being more likely to receive empiric antibiotics for gonorrhea and chlamydia.
While race was not associated with discordance between being given empiric antibiotics for gonorrhea and chlamydia and the eventual NAAT results, the following demographics were younger age, being single (vs. married/life partner), and being treated in the ED by an APP (vs. attending). These associations are likely because there was approximately five times the number of patients empirically treated for gonorrhea and chlamydia who tested negative compared to those who tested positive for one or both of those infections. There may have been social factors that required empiric antibiotics for gonorrhea and chlamydia in the ED in some of these encounters. However, our findings highlight an opportunity for improved antibiotics stewardship and shared decision making with the patient about the appropriateness of empiric treatment for STIs.
5. Limitations
Our data set only included patients tested for an STI or who received a urinalysis and urine culture. We missed patients diagnosed with a UTI based on urinalysis only and who did not get a urine culture. We did not examine patients empirically treated for STIs but who did not undergo STI testing. Our data set did not include history or physical exam findings which could have been included in our regression models; however, the history and physical examination findings may be unreliable for predicting STIs [27, 39]. We recognize that UTIs are frequently mismanaged and that ICD codes are an imperfect strategy for identifying disease. However, our UTI outcome focused on the diagnosis the ED provider thought was accurate at the time of the ED encounter. In addition, we could not differentiate complicated from uncomplicated UTIs or whether some patients had multiple ED encounters, which could have influenced prescriber decisions on ordering a urine culture or testing and treating for an STI. Our data set comprised data from a single healthcare system, so the results may not be generalizable. In addition, the data were collected years ago, so more current data could influence the results.
We only examined women from a single health system who were not admitted to the hospital, so our results are not generalizable to all demographics and all ED encounters. While we only examined bacterial urine cultures, viruses, fungi, and some noninfectious conditions can also cause UTI symptoms. We used 10,000 CFU/mL as our cutoff value, but no consensus exists for a specific CFU/mL that presents all UTIs.
Our data set had limited racial diversity, which was incompletely dichotomized. Adding other known social determinants of health could have improved our models.
Alternative antibiotics to the ones chosen in our models could have been used to treat gonorrhea and chlamydia in the ED. The U.S. Centers for Disease Control and Prevention updated STI testing and treatment recommendations in 2021, which occurred after the patient encounters in our data set [29]. We did not have information on the dose or duration of the antibiotics reported in the study, so some patients may have been treated with ceftriaxone plus azithromycin or doxycycline for a non-STI indication, but this is very unlikely in the setting of concurrent STI testing. We did not account for ED diagnoses that could also have affected our results, such as pelvic inflammatory disease (PID), which may be treated by antibiotics with activity against gonorrhea and chlamydia even though non-STI bacteria may have caused the condition. Regardless, patients with PID routinely get tested for STIs in the ED.
History or physical examination findings could have influenced the provider's decision to treat STIs in the ED. However, the history and physical examination may be less important for correctly diagnosing gonorrhea and chlamydia than demographic and laboratory findings [27, 39]. Information on a patient's sexual orientation, number of sexual partners, and the sex of sexual partners could have influenced STI treatment decisions. However, ED providers have historically done poorly in taking detailed sexual histories or adhering to STI testing and treatment guidelines [39, 40]. Other demographic variables not included in our data set could have affected the decision to treat STIs in the ED, such as homelessness or limited means to return to the ED for treatment if STI testing was positive. Poverty and lower socioeconomic status are both associated with higher rates of STIs [41, 42]. We did not include male patients in our analysis because those with penile discharge may be more likely to undergo preemptive STI treatment without concurrent testing. Therefore, they would not be included in our data set.
6. Conclusion
We did not observe significant differences between patient's race and the presence or absence of bacteriuria among patients diagnosed with a UTI or in managing gonorrhea and chlamydia infections. The patient's age and marital status, but not the provider's sex, were significantly associated with differences in the management of gonorrhea and chlamydia.
Acknowledgments
We thank Dr. Adetolu Odufuye for her early discussions on this project and Dr. Sheele for performing the multinomial regression analysis.
Data Availability
The dataset is not available for dissemination.
Conflicts of Interest
The authors declare that they have no conflicts of interest.
Supplementary Materials
Supplementary file 1: the univariate analysis for UTI diagnosis and ≥10,000 CFU/mL bacteriuria vs. <10,000 CFU/mL bacteriuria on urine culture. Supplementary file 2: the univariate analysis for the empiric treatment for gonorrhea and chlamydia in those tested for the diseases.
References
- 1.Barrow R. Y., Ahmed F., Bolan G. A., Workowski K. A. Recommendations for providing quality sexually transmitted diseases clinical services, 2020. Morbidity and Mortality Weekly Report Recommendations and Reports . 2020;68(5):1–20. doi: 10.15585/mmwr.rr6805a1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Vermund S. H., Geller A. B., Crowley J. S., editors. National Academies of Sciences. Engineering, and Medicine. Sexually Transmitted Infections: Adopting a Sexual Health Paradigm . Washington, DC, USA: The National Academies Press; 2021. p. p. 750. [PubMed] [Google Scholar]
- 3.Chesson H. W., Spicknall I. H., Bingham A., Brisson M., Eppink S. T., Farnham P. G., et al. The estimated direct lifetime medical costs of sexually transmitted infections acquired in the United States in 2018. Sexually Transmitted Diseases . 2021;48(4):215–221. doi: 10.1097/olq.0000000000001380. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Kreisel K. M., Spicknall I. H., Gargano J. W., Lewis F. M. T., Lewis R. M., Markowitz L. E., et al. Sexually transmitted infections among US women and men: prevalence and incidence estimates. Sexually Transmitted Diseases . 2021;48(4):208–214. doi: 10.1097/olq.0000000000001355. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Centers for Disease Control and Prevention, Us Department of Health & Human Services. Sexually Transmitted Infections Prevalence, Incidence, and Cost Estimates in the United States Atlanta (GA) Centers for Disease Control and Prevention; 2021. https://www.cdc.gov/std/statistics/prevalence-incidence-cost-2020.htm . [Google Scholar]
- 6.Foxman B. Urinary tract infection syndromes: occurrence, recurrence, bacteriology, risk factors, and disease burden. Infectious Disease Clinics of North America . 2014;28(1):1–13. doi: 10.1016/j.idc.2013.09.003. [DOI] [PubMed] [Google Scholar]
- 7.Zatorski C., Zocchi M., Cosgrove S. E., Rand C., Brooks G., May L. A single center observational study on emergency department clinician non-adherence to clinical practice guidelines for treatment of uncomplicated urinary tract infections. BMC Infectious Diseases . 2016;16(1):p. 638. doi: 10.1186/s12879-016-1972-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Tomas M. E., Getman D., Donskey C. J., Hecker M. T. Overdiagnosis of urinary tract infection and underdiagnosis of sexually transmitted infection in adult women presenting to an emergency department. Journal of Clinical Microbiology . 2015;53(8):2686–2692. doi: 10.1128/jcm.00670-15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Khan A., Plummer D., Hussain R., Minichiello V. Does physician bias affect the quality of care they deliver? Evidence in the care of sexually transmitted infections. Sexually Transmitted Infections . 2008;84(2):150–151. doi: 10.1136/sti.2007.028050. [DOI] [PubMed] [Google Scholar]
- 10.Sabin J. A., Greenwald A. G. The influence of implicit bias on treatment recommendations for 4 common pediatric conditions: pain, urinary tract infection, attention deficit hyperactivity disorder, and asthma. American Journal of Public Health . 2012;102(5):988–995. doi: 10.2105/ajph.2011.300621. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Goyal M. K., Hayes K. L., Mollen C. J. Racial disparities in testing for sexually transmitted infections in the emergency department. Academic Emergency Medicine . 2012;19(5):604–607. doi: 10.1111/j.1553-2712.2012.01338.x. [DOI] [PubMed] [Google Scholar]
- 12.Wiehe S. E., Rosenman M. B., Wang J., Fortenberry J. D. Disparities in chlamydia testing among young women with sexually transmitted infection symptoms. Sexually Transmitted Diseases . 2010;37(12):751–755. doi: 10.1097/olq.0b013e3181e50044. [DOI] [PubMed] [Google Scholar]
- 13.Masonbrink A. R., Richardson T., McCulloh R. J., Hall M., Bettenhausen J. L., Walker J. M., et al. Sexually transmitted infection testing in adolescents: current practices in the hospital setting. Journal of Adolescent Health . 2018;63(3):342–347. doi: 10.1016/j.jadohealth.2018.04.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Ricks T. N., Abbyad C., Polinard E. Undoing racism and mitigating bias among healthcare professionals: lessons learned during a systematic review. Journal of Racial and Ethnic Health Disparities . 2021;9(5):1990–2000. doi: 10.1007/s40615-021-01137-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Polhemus S. M., Pickett M. L., Liu X. (J., Fraser R., Ferguson C. C., Drendel A. L. Racial disparities in the emergency department evaluation of adolescent girls. Pediatric Emergency Care . 2022;38(7):307–311. doi: 10.1097/pec.0000000000002675. [DOI] [PubMed] [Google Scholar]
- 16.Casey J. A., Rudolph K. E., Robinson S. C., Bruxvoort K., Raphael E., Hong V., et al. Sociodemographic inequalities in urinary tract infection in 2 large California health systems. Open Forum Infectious Diseases . 2021;8(6):p. ofab276. doi: 10.1093/ofid/ofab276. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Middelkoop S. J. M., van Pelt L. J., Kampinga G. A., Ter Maaten J. C., Stegeman C. A. Influence of gender on the performance of urine dipstick and automated urinalysis in the diagnosis of urinary tract infections at the emergency department. European Journal of Internal Medicine . 2021;87:44–50. doi: 10.1016/j.ejim.2021.03.010. [DOI] [PubMed] [Google Scholar]
- 18.Yoon G. M. V. Probing automated treatment of urinary tract infections for bias: a case-study where machine learning perpetuates structural differences and racial disparities. Proceedings of the 12th ACM Conference on Bioinformatics, Computational Biology, and Health Informatics; August 2021; Gainesville, FL, USA. p. p. 1. [Google Scholar]
- 19.Bonner M., Sheele J. M., Cantillo-Campos S., Elkins J. M. A descriptive analysis of men diagnosed with epididymitis, orchitis, or both in the emergency department. Cureus . 2021;13(6) doi: 10.7759/cureus.15800.e15800 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Campos S. C., Elkins J. M., Sheele J. M. Descriptive analysis of prostatitis in the emergency department. The American Journal of Emergency Medicine . 2021;44:143–147. doi: 10.1016/j.ajem.2021.01.054. [DOI] [PubMed] [Google Scholar]
- 21.Elkins J. M., Hamid O. S., Simon L. V., Sheele J. M. Association of Bartholin cysts and abscesses and sexually transmitted infections. The American Journal of Emergency Medicine . 2021;44:323–327. doi: 10.1016/j.ajem.2020.04.027. [DOI] [PubMed] [Google Scholar]
- 22.Elkins J. M., Cantillo-Campos S., Sheele J. M. Frequency of coinfection on the vaginal wet preparation in the emergency department. Cureus . 2020;12(11) doi: 10.7759/cureus.11566.e11566 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Elkins J. M., Cantillo-Campos S., Thompson C., Mohseni M., Sheele J. M. Descriptive evaluation of male emergency department patients in the United States with gonorrhea and Chlamydia. Cureus . 2020;12(10) doi: 10.7759/cureus.11244.e11244 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Sheele J. M., Elkins J. M., Mohseni M. M., Monas J., Campos S. C., Benard R. B., et al. Vaginal leukocyte counts for predicting sexually transmitted infections in the emergency department. The American Journal of Emergency Medicine . 2021;49:373–377. doi: 10.1016/j.ajem.2021.06.070. [DOI] [PubMed] [Google Scholar]
- 25.Mohseni M., Simon L. V., Sheele J. M. Epidemiologic and clinical characteristics of tubo-ovarian abscess, hydrosalpinx, pyosalpinx, and oophoritis in emergency department patients. Cureus . 2020;12(11) doi: 10.7759/cureus.11647.e11647 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Mohseni M. M., Benard R. B., Mead-Harvey C., Mi L., Lindor R. A., Sheele J. M. Sexually transmitted infections in the emergency department are not associated with holidays or school breaks. The American Journal of Emergency Medicine . 2021;45:642–644. doi: 10.1016/j.ajem.2020.11.020. [DOI] [PubMed] [Google Scholar]
- 27.Sheele J. M., Niforatos J. D., Elkins J. M., Campos S. C., Thompson C. L. Prediction model for gonorrhea, chlamydia, and trichomoniasis in the emergency department. The American Journal of Emergency Medicine . 2022;51:313–319. doi: 10.1016/j.ajem.2021.11.004. [DOI] [PubMed] [Google Scholar]
- 28.Sheele J. M., Libertin C. R., Fink I., Jensen T., Dasalla N., Lyon T. D. Alkaline urine in the emergency department predicts nitrofurantoin resistance. Journal of Emergency Medicine . 2022;62(3):368–377. doi: 10.1016/j.jemermed.2021.10.022. [DOI] [PubMed] [Google Scholar]
- 29.Workowski K. A., Bachmann L. H., Chan P. A., Johnston C. M., Muzny C. A., Park I., et al. Sexually transmitted infections treatment guidelines, 2021. Morbidity and Mortality Weekly Report Recommendations and Reports . 2021;70(4):1–187. doi: 10.15585/mmwr.rr7004a1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Lawing L. F., Hedges S. R., Schwebke J. R. Detection of trichomonosis in vaginal and urine specimens from women by culture and PCR. Journal of Clinical Microbiology . 2000;38(10):3585–3588. doi: 10.1128/jcm.38.10.3585-3588.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Blake D. R., Duggan A., Joffe A. Use of spun urine to enhance detection of Trichomonas vaginalis in adolescent women. Archives of Pediatrics and Adolescent Medicine . 1999;153(12):1222–1225. doi: 10.1001/archpedi.153.12.1222. [DOI] [PubMed] [Google Scholar]
- 32.Hill J. T., Starr T., Ash V. D., Liao R. S. Evaluation of the gen-probe APTIMA trichomonas vaginalis assay with the APTIMA combo 2 assay for detection of Chlamydia trachomatis (CT) and/or Neisseria gonorrhoeae (GC) Journal of Molecular Diagnostics . 2011;13(6):p. 749. [Google Scholar]
- 33.Breslin K., Tuchman L., Hayes K. L., Badolato G., Goyal M. K. Sensitivity and specificity of empiric treatment for sexually transmitted infections in a pediatric emergency department. The Journal of Pediatrics . 2017;189:48–53. doi: 10.1016/j.jpeds.2017.05.050. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Cook R. L., Wiesenfeld H. C., Ashton M. R., Krohn M. A., Zamborsky T., Scholle S. H. Barriers to screening sexually active adolescent women for chlamydia: a survey of primary care physicians. Journal of Adolescent Health . 2001;28(3):204–210. doi: 10.1016/s1054-139x(00)00152-x. [DOI] [PubMed] [Google Scholar]
- 35.Dretler A. W., Trolard A., Bergquist E. P., Cooper B., Liang S. Y., Stoner B. P., et al. The influence of race and sex on gonorrhea and chlamydia treatment in the emergency department. The American Journal of Emergency Medicine . 2020;38(3):566–570. doi: 10.1016/j.ajem.2019.05.054. [DOI] [PubMed] [Google Scholar]
- 36.Lechtenberg R. J., Samuel M. C., Bernstein K. T., Lahiff M., Olson N., Bauer H. M. Variation in adherence to the treatment guidelines for Neisseria gonorrhoeae by clinical practice setting, California, 2009 to 2011. Sexually Transmitted Diseases . 2014;41(5):338–344. doi: 10.1097/olq.0000000000000113. [DOI] [PubMed] [Google Scholar]
- 37.Fox H. T., Sheele J. M. Association of marital status in the testing and treatment of sexually transmitted infections in the emergency department. Cureus . 2021;13(8) doi: 10.7759/cureus.17489.e17489 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Sheele J. M., Bragg K. J., Bragg B., Campos S. C., Elkins J. M., Niforatos J. D. Descriptive epidemiology of women in the emergency department with gonorrhea and chlamydial infection in the United States. Advanced Emergency Nursing Journal . 2022;44(2):144–157. doi: 10.1097/tme.0000000000000408. [DOI] [PubMed] [Google Scholar]
- 39.Sheele J. M., Smith J., Niforatos J. D., Wessling E., Hilliker B., Bragg B., et al. History, physical examination, and laboratory findings associated with infection and the empiric treatment of gonorrhea and chlamydia of women in the emergency department. Cureus . 2019;11(12) doi: 10.7759/cureus.6482.e6482 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Niforatos J. D., Nowacki A. S., Avery A., Gripshover B. M., Yax J. A. Clinical knowledge of human immunodeficiency virus and sexually transmitted infections among emergency medicine providers. International Journal of STD & AIDS . 2019;30(13):1290–1297. doi: 10.1177/0956462419866052. [DOI] [PubMed] [Google Scholar]
- 41.Springer Y. P., Samuel M. C., Bolan G. Socioeconomic gradients in sexually transmitted diseases: a geographic information system-based analysis of poverty, race/ethnicity, and gonorrhea rates in California, 2004-2006. American Journal of Public Health . 2010;100(6):1060–1067. doi: 10.2105/ajph.2009.172965. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Townes A., Rosenberg M., Guerra-Reyes L., Murray M., Herbenick D. Inequitable experiences between Black and White women discussing sexual health with healthcare providers: findings from a US probability sample. The Journal of Sexual Medicine . 2020;17(8):1520–1528. doi: 10.1016/j.jsxm.2020.04.391. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplementary file 1: the univariate analysis for UTI diagnosis and ≥10,000 CFU/mL bacteriuria vs. <10,000 CFU/mL bacteriuria on urine culture. Supplementary file 2: the univariate analysis for the empiric treatment for gonorrhea and chlamydia in those tested for the diseases.
Data Availability Statement
The dataset is not available for dissemination.