ABSTRACT
Purpose
While urinary tract infections (UTIs) are thought to be common among women, as many as 65% of UTI diagnoses may be inaccurate. To identify strategies to improve antibiotic stewardship, we sought to determine the clinical and laboratory factors associated with overdiagnosis and overtreatment of UTIs.
Methods
Electronic health records identified patients bearing an isolated diagnostic code for UTI within a single healthcare system during July 2019. Demographic, clinical, and microbial data were collected by manual chart review. Regression analyses were utilized to determine factors associated with guideline non‐concordant UTI diagnosis and treatment utilizing R statistical software (version 4.3.1).
Results
In patients diagnosed with UTI, 64% were treated with antibiotics despite only 28% having symptoms consistent with UTI diagnostic criteria. Of patients diagnosed with a UTI who were treated in an emergency room (ER) setting, 95% were given antibiotics compared with only 55% of patients in an outpatient setting. Even without any urinary symptoms, 95% of patients in the ER and 27% of patients in outpatient settings were treated with antibiotics. Patients who presented to the ER for non‐localizing symptoms, such as mental status changes (MSC), were more likely to be diagnosed with UTI compared with those in an outpatient setting.
Conclusion
The results of this study demonstrate that patients were frequently diagnosed with and treated for a UTI despite not meeting diagnostic criteria. This pattern of overdiagnosis leads to overtreatment, particularly in acute care settings, contributing to worsening antibiotic resistance in conjunction with incomplete evaluation of patients' primary complaints.
Keywords: antibiotic stewardship, cystitis, diagnostic stewardship, overtreatment, recurrent UTI, urinary tract infection
1. Introduction
With nearly half of women experiencing a urinary tract infection (UTI) in their lifetime [1], these infections are one of the leading reasons for antibiotic prescriptions in the United States and worldwide [2]. Despite their clinical ubiquity, physicians exhibit significant variation in their approaches to the workup and treatment of uncomplicated UTIs that may not always follow clinical practice guidelines [3, 4]. Use of unvalidated approaches to the management of UTIs may lead to inaccurate diagnoses and inappropriate antibiotic treatments; prior studies demonstrate that 50%–70% of antibiotic prescriptions for UTI may be inappropriate [5], which contributes to worsening global rates of antibiotic resistance.
Current treatment guidelines from the American Urological Society (AUA), Infectious Disease Society of America (IDSA), and American Geriatrics Society (AGS) agree that UTI diagnosis should be based on clinical symptomology of urinary symptoms (e.g., dysuria, hematuria, urinary frequency, new incontinence in the absence of vaginal symptoms) associated with culture‐proven bacteriuria with a bacterial uropathogen [6, 7, 8]. Rates of asymptomatic bacteriuria range from 1% to 5% in premenopausal women to 16% in older women in the community and as high as 50% for older women in long‐term care facilities [6], demonstrating why appropriate clinical symptoms are needed to diagnose a UTI.
Prior studies, however, have shown that among patients diagnosed with a UTI in the emergency room (ER), only 32% of adults and 17% of the oldest adults have urinary symptoms [9], demonstrating poor adherence to current clinical practice guidelines. Patients who have been inappropriately treated with antibiotics for a UTI (e.g., treating those with asymptomatic bacteriuria) have worse outcomes, which may include subsequent infections with multidrug‐resistant organisms or substantial medication side effects [10, 11, 12]. Patients who have an inappropriate initial antibiotic prescription are also subjected to additional drug costs and are at risk of antibiotic switching which can cost up to $2000 more per patient [13, 14].
Overdiagnosis and overtreatment of UTIs produce worse patient outcomes, increased healthcare costs, and higher societal antibiotic burden. The goals of our study were to evaluate the factors associated with inappropriate diagnosis and treatment of cystitis in the outpatient setting in a set of patients diagnosed with UTI without signs of bacteremia.
2. Materials & Methods
All consecutive patients bearing an isolated diagnostic code for UTI/cystitis within a single healthcare system during July 2019 were identified by the ICD‐10 diagnostic code (Appendix SA). The healthcare system utilized was a large tertiary medical center in a major metropolitan city with high patient volumes. July 2019 was selected to avoid any alterations in treatment delivery resulting from the coronavirus pandemic. Patients demonstrating clinical signs or hemodynamic features of bacteremia, positive blood cultures, medical comorbidities associated with complicated UTI (e.g., indwelling catheter, recent urologic surgery, pregnancy, immunodeficiency, or transplant), or those admitted for inpatient care were excluded, resulting in a cohort of 909 patients.
2.1. Data Collection
Manual chart review was utilized to extract demographic (e.g., age, gender), clinical (e.g., symptoms), and microbial (e.g., features of microscopic urinalysis, urine microbiological testing) data from patients who met inclusion criteria. We also catalogued antibiotic treatments used, alternative diagnoses explored, and urine culture and susceptibility results. Symptoms were designated as “acute” if arising < 7 days before the encounter and “chronic” when present for > 28 days before presentation. The presence of “UTI symptoms,” specified by the American Urological Association and expert consensus, was defined as either dysuria or hematuria with at least one additional urinary symptom, including urinary urgency, urinary frequency, worsening incontinence, or suprapubic pressure in the absence of any vaginal symptoms [7, 15]. “Mental Status Changes” (MSC) captured symptomatology such as altered mental status, confusion, and lethargy. Escherichia coli, Staphylococcus saprophyticus, or Enterococcus faecalis were defined as the most common “uropathogenic” bacteria, while all other species were designated as atypical bacteria. Risk factors for complicated UTIs such as anatomic abnormalities of the urinary tract, systemic disease, immunodeficiency, pregnancy, or history of multidrug resistance bacteria were not collected. Clinical definitions are defined in Appendix SB.
2.2. Statistical Analysis
Categorical variables were reported as number and frequencies (%) and continuous variables as means with standard deviation (SD). We analyzed demographic, symptomatic, and microbial measures using a simple, linear or logistic regression analysis to determine the effects on diagnosis and treatment of UTIs. We analyzed the adjusted effect of these predictor variables in the diagnosis and treatment of UTIs with multivariable logistic regression or Firth logistic regression as appropriate. p < 0.05 were considered significant. All statistical analyses were performed using R statistical software (version 4.3.1).
3. Results
Of 909 patients, 85% (773/909) were female, with a mean age of 57 years of age. Of all patients, 64% (579/909) were prescribed antibiotics, although only 28% of patients exhibited symptoms consistent with UTI and only 40% and 41% of patients had positive bacteria on urinalysis or positive urine cultures, respectively (Figure 1). Only 34% of prescribed antibiotics were first‐line (Appendix SC). Of all patients, 22% were seen in an ER setting, while the remaining 78% of patients presented to outpatient medical offices. A small subset of patients (3%) had chronic UTI symptoms, while 25% had acute UTI symptoms. Patients who presented to the ER had higher levels of white blood cells (WBCs), red blood cells (RBCs), urinary squamous cells, and bacteria seen on urinalysis than patients who presented to office‐based clinical settings (Table 1).
Figure 1.
Bar chart demonstrating the varying patient features for different patient subsets. UTI symptoms were defined as dysuria or hematuria with additional urinary symptoms (e.g., urinary frequency, urgency, or new or worsening incontinence) and no vaginal symptoms. Positive culture rates are reported as the percentage of patients with a positive culture of all those patients who were tested with urinary cultures. Mental status changes (MSC) noted any patient with symptoms of confusion or delirium.
Table 1.
Characteristics of patients diagnosed with UTI in ER and office settings.
| ER, n = 209 | Office, n = 740 | p | |
|---|---|---|---|
| Age (mean (SD)) | 50.79 (26.06) | 59.23 (22.07) | < 0.001 |
| UTI symptoms (n (%)) | 55 (27.8) | 195 (27.6) | > 0.999 |
| Vag symptoms (n (%)) | 12 (5.7) | 39 (5.3) | 0.926 |
| MSC (n (%)) | 14 (6.7) | 17 (2.3) | 0.003 |
| Chronic (n (%)) | 44 (22.2) | 323 (45.9) | < 0.001 |
| Urine culture performed (n (%)) | 176 (88.4) | 574 (80.5) | 0.013 |
| Positive urine culture (n (%)) | 83 (41.7) | 290 (40.7) | 0.868 |
| Antibiotics prescribed (n (%)) | 189 (95) | 390 (55) | < 0.001 |
| Microscopic urinalysis | |||
| WBC (mean (SD)) | 62.7 (71.54) | 48.6 (74.29) | 0.015 |
| RBC (mean (SD)) | 25.5 (56.8) | 16.2 (44.9) | 0.013 |
| Squam (mean (SD)) | 22.97 (52.06) | 12.15 (34.73) | < 0.001 |
| UABac (n (%)) | 118 (56.5) | 259 (35) | < 0.001 |
Note: Statistically significant values are shown in bold.
On simple logistic regression (Table 2), presentation to the ER was strongly associated with antibiotic treatment (OR: 22.09, 95% CI: 11.12–48.43). Antibiotic treatment was also associated with positive urine culture, the presence of vaginal symptoms (vaginal itching, burning, or discharge), the presence of UTI symptoms, uropathogenic species on urine culture, and the presence of WBC on microscopic urinalysis. Patient age was notably a negative predictor of antibiotic treatment (OR: 0.99, 95% CI: 0.98–0.99) (Table 2). The presence of RBCs, bacteria, and squamous cells on urinalysis were each associated with antibiotic treatment when assessed individually but did not remain statistically significant in multivariable analysis. Mental status changes were also predictive of antibiotic treatment when assessed individually but once again lost statistical significance when assessed in multivariable analysis. Additionally, chronic symptoms were negatively predictive of treatment when assessed individually but were no longer statistically significant in multivariable analysis.
Table 2.
Patient and laboratory features associated with UTI treatment on logistic regression.
| Simple logistic regression | Multivariable logistic regression | |
|---|---|---|
| Feature | OR (95% CI) | OR (95% CI) |
| Chronic symptoms (> 28 days) | 0.60 (0.46–0.79) | 0.68 (0.45–1.03) |
| ER visit | 15.34 (8.40–31.45) | 22.0 (11.1–48.4) |
| Patient age | 0.98 (0.98–0.99) | 0.98 (0.97–0.99) |
| Male | 0.63 (0.44–0.91) | 0.65 (0.36–1.15) |
| Vaginal symptoms | 1.25 (0.70–2.37) | 2.50 (1.14–5.50) |
| Mental status changes | 3.05 (1.25–9.08) | 1.63 (0.47–6.68) |
| Positive urine culture | 10.87 (7.52–16.11) | 8.44 (4.62–15.9) |
| UTI symptoms present | 8.15 (5.31–13.06) | 10.7 (6.34–18.6) |
| Urine culture performed | 3.96 (2.78–5.68) | 1.05 (0.63–1.74) |
| Uropathogenic species | 11.47 (7.05–19.92) | 2.29 (1.10–4.82) |
| Microscopic Urinalysis | ||
| WBC | 1.01 (1.01–1.01) | 1.00 (1.00–1.01) |
| RBC | 1.01 (1.00–1.01) | 0.99 (0.99–1.00) |
| Bacteria | 3.11 (2.30–4.22) | 0.85 (0.53–1.35) |
| Squam | 1.01 (1.00–1.02) | 1.00 (0.99–1.01) |
Note: Statistically significant values are shown in bold.
Ninty‐five percent (189/209) of patients presenting to the ER were treated for UTI with antibiotics compared with 55% (390/740) of patients who presented to office‐based settings (p < 0.001) (Table 1). Both groups of patients had similar rates of classic UTI symptoms (ER: 28% vs. Office: 28%, p > 0.999), however classic UTI symptomatology only predicted the likelihood of treatment in the office‐based patients (OR: 10.56, 95% CI: 5.16–23.20); it did not predict treatment for patients in the ER (OR: 0.33, 95% CI: 0.06–1.85) (Table 3).
Table 3.
Patient and laboratory features associated with antibiotic treatment for UTI on logistic regression for ER and office‐based visits.
| Office | ER | |
|---|---|---|
| Feature | OR (95% CI) | OR (95% CI) |
| Chronic | 0.59 (0.32–1.07) | 0.06 (0.002–0.7) |
| Patient age | 0.98 (0.97–0.99) | 1.02 (0.99–1.07) |
| Male | 0.78 (0.30–2.02) | 25.4 (1.34–2.00) |
| Vaginal symptoms | 1.95 (0.75–5.16) | 2.70 (0.00–105) |
| MSC | 1.13 (0.22–7.88) | 0.28 (0.001–69.7) |
| UCx positive | 9.19 (4.40–‐20.6) | 39.5 (3.22–611) |
| UTI symptoms | 9.50 (4.75–20.3) | 0.33 (0.06–1.84) |
| UCx taken | 1.44 (0.72–2.91) | 0.05 (0.001–0.82) |
| Uropathogenic species | 0.50 (0.13–2.09) | 0.03 (0.001–8.5) |
| Microscopic urinalysis | ||
| WBC | 1.01 (1.00–1.01) | 1.02 (1.001–1.04) |
| RBC | 1.00 (0.99–1.00) | 0.997 (0.98–1.01) |
| Bacteria | 1.41 (0.74–2.72) | 0.51 (0.06–2.19) |
| Squamous cells | 0.99 (0.99–1.01) | 1.002 (0.99–3.06) |
Note: Statistically significant values are shown in bold.
Symptomatically, patients who presented to the ER were more likely to present with mental status changes than individuals presenting to outpatient settings (ER: 7%, 14/209 vs. Office 2%, 17/740, p = 0.003) (Table 1). The characteristics of patients presenting with mental status changes did not significantly differ between ER and outpatient settings for parameters including age, lack of vaginal symptoms, rates of cultures taken, and rates of positive culture (Table 4).
Table 4.
Characteristics of patients with mental status changes diagnosed with UTI in ER and office.
| ER, n = 14 | Office, n = 17 | p | |
|---|---|---|---|
| Patient age (mean (SD)) | 82.43 (16.16) | 79.88 (21.14) | 0.714 |
| UTI symptoms (n (%)) | 1 (7.1) | 2 (11.8) | > 0.999 |
| No vaginal symptoms (n (%)) | 14 (100.0) | 17 (100.0) | NA |
| Chronic symptoms (n (%)) | 14 (100.0) | 17 (100.0) | NA |
| Urine culture taken (n (%)) | 14 (100.0) | 15 (88.2) | 0.554 |
| Positive urine culture (n (%)) | 11 (78.6) | 12 (70.6) | 0.926 |
| Antibiotics given (n (%)) | 14 (100.0) | 12 (70.6) | 0.085 |
For the subset of patients who did not endorse any urinary symptoms (e.g., no dysuria, hematuria, urinary frequency, urinary urgency), comprising 39% of all UTI‐coded encounters (n = 373/949), 95% (60/63) of those who presented to the ER received antibiotic treatment for a UTI. This was a significantly higher treatment percentage than the 27% (84/310) of their counterparts who were evaluated in outpatient settings (p < 0.001). Patients with no urinary symptoms in the ER were more likely to present with mental status changes (18%, 11/63) than patients seen in an outpatient office (4%, 13/310, p < 0.001) and were more likely to receive a urinary culture as evaluation (ER: 92%, 58/63 vs. Office: 71%, 221/310, p = 0.001). The rate of culture positivity trended toward higher values in the ER, but this difference was not statistically significant (ER: 43%, 27/63 vs. Office: 30% 93/310, p = 0.068). This group of patients with no reported urinary symptoms were presenting for a variety of non‐urogynecologic reasons. In the ER, presentation varied from acute chest pain to mechanical fall to constipation. Patients seen in an outpatient office often had no specific complaints and received urine testing as part of their laboratory workup for an annual physical exam. Other asymptomatic outpatients had urine testing as part of follow‐up lab work after transplant, before chemotherapy treatments, or as part of autoimmune diseases monitoring. Patients in the office setting were more likely to have a chronic course of symptomology (Table 5).
Table 5.
Characteristics of patients with no urinary symptoms diagnosed with UTI in ER and office.
| ER, n = 63 | Office, n = 310 | p | |
|---|---|---|---|
| Patient age (mean (SD)) | 54.41 (28.82) | 63.42 (22.05) | 0.030 |
| Vaginal symptoms (n (%)) | 3 (4.8) | 8 (2.6) | 0.600 |
| MSC (n (%)) | 11 (17.5) | 13 (4.2) | < 0.001 |
| Chronic symptoms (n (%)) | 11 (17.7) | 149 (48.4) | < 0.001 |
| Urine culture performed (n (%)) | 58 (92.1) | 221 (71.3) | 0.001 |
| Positive urine culture (n (%)) | 27 (42.9) | 93 (30.1) | 0.068 |
| Antibiotics prescribed (n (%)) | 60 (95.2) | 84 (27.1) | < 0.001 |
Note: Statistically significant values were bolded.
4. Discussion
In this study, we performed a detailed manual review of UTI encounters at a large academic medical center to determine factors contributing to the significant overprescription of antibiotics for UTI. Overall, profound over screening and overtreatment were present in all practice settings, frequently justified by rationales discordant with evidence‐based and guideline‐concurrent care. Our analysis revealed that an emergent care setting as well as the presence of mental status changes were both associated with much higher rates of inappropriate diagnosis and antibiotic treatment. Asymptomatic and culture‐negative individuals were frequently treated with antibiotics, either because of a urinalysis or urine dipstick positive for leukocytes or because of the attribution of a wide range of non‐urinary symptoms to UTI. These findings stress the urgent need for improvements in both diagnostic and antibiotic stewardship.
Patients have urinary testing despite not having urinary symptoms. A total of 92% and 71% of asymptomatic ER and office visit patients respectively went on to receive urine culture testing despite being asymptomatic from a urinary standpoint. For many encounters, particularly ER visits, the decision to test a patient with no urinary symptoms was often based on an overarching justification that UTI can be responsible for a wide range of nonspecific signs and symptoms. This indicates that poor understanding of the defining characteristics of UTI leads to poor diagnostic stewardship, with culture and laboratory urinalysis being overused for patients who are not experiencing acute‐onset urinary symptoms. In addition, markers of poor urinalysis collection quality (e.g., elevated squamous cells) were also higher in the ER, suggesting that poor collection methods or specimen processing may also contribute to overdiagnosis. Emergency medicine guidelines support the fact that patients may not even need urinary testing because the basis of a UTI diagnosis should be based on clinical symptomology [16]. Urine dipsticks can be a useful adjunct to a clinical exam to help exclude the presence of a UTI, but it should not be used alone to rule in infections even with high pre‐test probability [17].
Over testing for UTIs is leading to overtreatment in asymptomatic patients. The presence of urinary symptoms occurring with a positive culture in the absence of vaginal symptoms are key features of appropriate UTI diagnosis [16], however, our data showed that of the patients in the ER who were diagnosed with a UTI, only 28% had symptoms consistent with UTI. This aligns well with previous studies examining rates of cystitis symptoms in those diagnosed with a UTI in the ER [18, 19]. In addition to being more likely to be tested for a UTI, patients treated in the ER were also more likely to receive antibiotics for a UTI than patients who sought care in an outpatient setting (ER: 95%, 189/209 vs. Office: 55%, 390/740). While it might be hypothesized that ER patients would have a more severe presentation or be more frail, our data showed the ER patients tended to be younger overall with similar clinical features including similar rates of urinary symptoms, vaginal symptoms, and culture positivity. Together, these findings may indicate that over‐testing of patients in the ER for UTI is driving much of the inappropriate treatment of both asymptomatic bacteriuria and culture‐negative subjects with nonspecific symptoms [9, 20, 21].
Urinary testing should be reserved for patients with symptoms consistent with a UTI. It is important to note that more than half of urine cultures for both patients evaluated in the ER and outpatient‐based clinics were negative. In most of the encounters in which culture‐negative patients were treated with antimicrobials for UTI, the rationale to provide antibiotics was either poorly documented in the clinical record or based on point‐of‐care urine dipstick testing. While the absence of pyuria on urinalysis has a high negative predictive value and can help rule out the diagnosis, urinalysis when utilized independent of UTI symptoms should not be used to guide treatment decisions even if positive for pyuria as it has a poor positive predictive value for UTI [6, 10, 17]. Given that pyuria is common among women of all ages, especially elderly women [6], performing an initial screening urinalysis in the absence of urinary symptoms often results in overtreatment as it may be uncomfortable for providers to withhold antibiotic treatment in the face of positive laboratory results. These principles underlie clinical guidance by the US Preventative Services Task Force and Infectious Disease Society of America both recommending against screening urinalysis in nonpregnant adults [6, 22]. This is why we encourage not only antibiotic stewardship but also diagnostic stewardship appropriately restricting laboratory testing to patients in whom there is significant clinical suspicion for UTI.
Mental status changes are poorly understood to be a symptom of UTI. Despite AGS and IDSA guidelines recommending against testing and treatment for patients with only mental status changes [4, 6, 20, 23], all of the patients (100%, n = 14) who presented with mental status changes to the ER were treated with antibiotics despite only one (7%) having urinary symptoms. Mental status changes are often misunderstood to be a definitive feature of UTIs [3, 4, 23], but in the absence of localizing urinary symptoms or systemic signs of infection, such as fever or hemodynamic instability, mental status changes correlate poorly to infectious processes [10]. In addition to the lack of clinical benefit from antibiotic treatment, patients with mental status changes may suffer harms from treatment with adverse drug reactions and reinfection with drug‐resistant organisms [11, 12]. These harms are so great that the AGS and IDSA recommend against even testing for UTI in patients with only mental status changes due to the risk for inappropriate treatment. When evaluating a patient presenting with delirium‐like symptoms or mental status changes, providers should explore non‐urinary etiologies of MSC and observe the patient while encouraging or facilitating fluid intake without prescribing antibiotics [10]. Our study showed a lack of adherence to these guidelines in both the ER and office settings with 100% of ER patients and 71% of office visit patients with MSC receiving antibiotic treatment despite only 7% of ER patients and 12% of office patients experiencing urinary symptoms. This discordance between the symptomatology of these patients and their diagnosis and treatment highlights the need for further provider education and understanding of guidelines for the treatment of MSC in the absence of urinary symptoms.
Additionally, it is important to note that our manual chart review revealed that ~75% of patients who were diagnosed with a UTI by diagnostic code did not meet the definition or clinical criteria for a UTI diagnosis. This suggests that we must reconsider the data that exist on UTI from claims‐based databases that underlie many of our currently held beliefs about this condition. Genitourinary symptoms, particularly when uncomfortable or painful, appear to be frequently misdiagnosed, which complicates the clinical care of patients with these symptoms, particularly women. Rates of pelvic pain among women are estimated at 14%–40% [24, 25], and at least half of these patients have genitourinary symptoms [26]. Only a third of these women end up getting care [24, 27] and 50% never even receive a diagnosis [26, 27]. If most genitourinary complaints in women are written off as a UTI, it both prevents us from understanding how best to manage actual acute cystitis and recurrent UTIs in women who have them and does a tremendous disservice to the patients with other genitourinary complaints who are not getting treatment for their true, underlying disorder.
Suboptimal treatment of women with genitourinary symptoms leads to additional healthcare costs, worse patient outcomes, and increasing rates of antimicrobial resistance [10, 13]. Although treatment is reasonable for patients with symptomatic UTIs to relieve symptoms, shorten duration of symptoms, and prevent progression, our data suggest that the majority of UTIs are inappropriately diagnosed and overtreated in both the ER and office‐based settings; treatment rates of 95% in the ER and 55% in outpatient settings far exceed the rates of urinary symptoms of 28% in both settings. It is of note that treatment rates were significantly higher (p < 0.01) in the ER versus office‐based settings despite similar levels of urinary symptoms and positive cultures between the groups. The equitable care of women requires appropriate consideration of the complete differential diagnosis for nonspecific genitourinary and generalized symptoms; providers should not assume any genitourinary complaint is indicative of a UTI. Consideration of a complete differential is especially important in elderly women with non‐localizing symptoms. Given the challenges providers face in delivering care in the ER setting, this discrepancy in the appropriateness of UTI care further supports the current movement to transition to a “medical home model” that can provide better comprehensive and accessible care that may prevent much of this overtreatment [28].
These patient encounters were from a large, tertiary medical center in a major metropolitan city with high patient volumes. Although our data were collected from multiple ERs and > 200 medical practice offices, analysis of a single healthcare system may limit the generalizability of our findings. Therefore, our results may not be reproducible in community‐based or smaller‐volume practice settings. As this was a retrospective cohort study, it is difficult to establish causality between any patient factors and UTI diagnosis and treatment patterns. Data were collected via manual chart review and therefore relies on physician documentation, which can be unreliable. Rates of asymptomatic UTIs may be overestimated if patient symptoms were not properly documented in the medical record. The performance of a manual chart review can provide insight on provider rationale, but additional research is needed to identify why UTI treatment rates and patterns vary in the ER and office setting. Patients with medical comorbidities associated with complicated UTIs were included in this analysis, and thus these results are unlikely to be able to be extrapolated to that population. Providers may be empirically treating patients deemed higher risk for complicated UTIs, but without the comparison of those populations, further research is necessary to determine if results are similar in these risk‐stratified populations.
5. Conclusions
Our data continue to support evidence that UTIs are often diagnosed and treated in a manner inconsistent with guidelines meant to direct best clinical practices. Laboratory tests are frequently used in the absence of urinary symptoms demonstrating inappropriate testing. Over‐testing is contributing to the overtreatment of patients without UTI symptoms and often without microbiological evidence of infection. Given the potential harms of overtreatment such as antibiotic resistance, recurrent infections, and high healthcare costs, our study demonstrates the need for advances in adherence to and understanding of UTI diagnostic criteria and treatment guidelines as well as improved diagnostic and antimicrobial stewardship.
Ethics Statement
IRB approval: IRB#21‐001403.
Conflicts of Interest
A.L. Ackerman receives grant funding from Medtronic, Inc. and MicrogenDx and is an advisor for Abbvie, GlaxoSmithKline, and Watershed Medical.
Supporting information
Supporting information.
Supporting information.
Supporting information.
Acknowledgments
ALA is supported by NIDDK K08DK118176 and Department of Defense PRMRP W81XWH2110644.
Data Availability Statement
Underlying data are available upon request from the corresponding author.
References
- 1. Kunin C., Urinary Tract Infections: Detection, Prevention, and Management (Baltimore, Md: Williams & (Yang, 2022) Wilkins, 1997). [Google Scholar]
- 2. Yang X., Chen H., Zheng Y., Qu S., Wang H., and Yi F., “Disease Burden and Long‐Term Trends of Urinary Tract Infections: A Worldwide Report,” Frontiers in Public Health 10 (2022): 888205, 10.3389/fpubh.2022.888205. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Kistler C. E., Beeber A. S., Zimmerman S., et al., “Nursing Home Clinicians' Decision to Prescribe Antibiotics for a Suspected Urinary Tract Infection: Findings From a Discrete Choice Experiment,” Journal of the American Medical Directors Association 21, no. 5 (May 2020): 675–682.e1, 10.1016/j.jamda.2019.12.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Millard J. W., Howell C. K., and Palasik B. N., “Asymptomatic but Altered? When Urinalysis Holds the Key,” The Senior Care Pharmacist 36, no. 8 (August 2021): 381–386, 10.4140/TCP.n.2021.381. [DOI] [PubMed] [Google Scholar]
- 5. Clark A. W., Durkin M. J., Olsen M. A., et al., “Rural‐Urban Differences in Antibiotic Prescribing for Uncomplicated Urinary Tract Infection,” Infection Control & Hospital Epidemiology 42, no. 12 (2021): 1437–1444, 10.1017/ice.2021.21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Nicolle Lindsay E., Gupta Kalpana, Bradley Suzanne F., et al., “Clinical Practice Guideline for the Management of Asymptomatic Bacteriuria: 2019 Update by the Infectious Diseases Society of America,” Clinical Infectious Diseases 68, no. Issue 10, (May 2019): e83–e110, 10.1093/cid/ciy1121. [DOI] [PubMed] [Google Scholar]
- 7. Anger J. T., Bixler B. R., Holmes R. S., Lee U. J., Santiago‐Lastra Y., and Selph S. S., “Updates to Recurrent Uncomplicated Urinary Tract Infections in Women: AUA/CUA/Sufu Guideline,” Journal of Urology 208, no. 3 (2022): 536–541, 10.1097/ju.0000000000002860. [DOI] [PubMed] [Google Scholar]
- 8. Gupta K., Hooton T. M., Naber K. G., et al., “International Clinical Practice Guidelines for the Treatment of Acute Uncomplicated Cystitis and Pyelonephritis in Women: A 2010 Update by the Infectious Diseases Society of America and the European Society for Microbiology and Infectious Diseases,” Clinical Infectious Diseases 52, no. 5 (2011): e103–e120. [DOI] [PubMed] [Google Scholar]
- 9. Rowe T. A. and Juthani‐Mehta M., “Diagnosis and Management of Urinary Tract Infection in Older Adults,” Infectious Disease Clinics of North America 28, no. 1 (March 2014): 75–89, 10.1016/j.idc.2013.10.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Goebel M. C., Trautner B. W., and Grigoryan L., “The Five Ds of Outpatient Antibiotic Stewardship for Urinary Tract Infections,” Clinical Microbiology Reviews 34, no. 4 (2021): e00003–e00020, 10.1128/CMR.00003-20. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Nicolle L. E., Bjornson J., Harding G. K. M., and MacDonell J. A., “Bacteriuria in Elderly Institutionalized Men,” New England Journal of Medicine 309 (1983): 1420–1425. [DOI] [PubMed] [Google Scholar]
- 12. Nicolle L. E., Mayhew W. J., and Bryan L., “Prospective Randomized Comparison of Therapy and no Therapy for Asymptomatic Bacteriuria in Institutionalized Elderly Women,” American Journal of Medicine 83, no. 1 (1987): 27–33. [DOI] [PubMed] [Google Scholar]
- 13. Shafrin J., Marijam A., Joshi A. V., et al., “Impact of Suboptimal or Inappropriate Treatment on Healthcare Resource Use and Cost Among Patients With Uncomplicated Urinary Tract Infection: An Analysis of Integrated Delivery Network Electronic Health Records,” Antimicrobial Resistance & Infection Control 11, no. 1 (2022): 133, 10.1186/s13756-022-01170-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Moon R. C., Marijam A., Mitrani‐Gold F. S., et al., “Treatment Patterns, Healthcare Resource Use, and Costs Associated With Uncomplicated Urinary Tract Infection Among Female Patients in the United States,” PLoS One 17, no. 11 (2022): e0277713, 10.1371/journal.pone.0277713. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Bilsen M. P., Conroy S. P., Schneeberger C., et al., “A Reference Standard for Urinary Tract Infection Research: A Multidisciplinary Delphi Consensus Study,” Lancet Infectious Diseases 24 (2024): e513–e521, 10.1016/s1473-3099(23)00778-8. [DOI] [PubMed] [Google Scholar]
- 16. Long B. and Koyfman A., “The Emergency Department Diagnosis and Management of Urinary Tract Infection,” Emergency Medicine Clinics of North America 36, no. 4 (November 2018): 685–710, 10.1016/j.emc.2018.06.003. [DOI] [PubMed] [Google Scholar]
- 17. Devillé W. L., Yzermans J. C., van Duijn N. P., Bezemer P. D., van der Windt D. A., and Bouter L. M., “The Urine Dipstick Test Useful to Rule out Infections. A Meta‐Analysis of the Accuracy,” BMC Urology 4 (2004): 4, 10.1186/1471-2490-4-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Dubbs S. B. and Sommerkamp S. K., “Evaluation and Management of Urinary Tract Infection in the Emergency Department,” Emergency Medicine Clinics of North America 37, no. 4 (2019): 707–723, 10.1016/j.emc.2019.07.007. [DOI] [PubMed] [Google Scholar]
- 19. Caterino J. M., Ting S. A., Sisbarro S. G., Espinola J. A., and Camargo C. A., “Age, Nursing Home Residence, and Presentation of Urinary Tract Infection in U.S. Emergency Departments, 2001–2008,” Academic Emergency Medicine 19, no. 10 (2012): 1173–1180, 10.1111/j.1553-2712.2012.01452. [DOI] [PubMed] [Google Scholar]
- 20. Kwok M., McGeorge S., Mayer‐Coverdale J., et al., “Guideline of Guidelines: Management of Recurrent Urinary Tract Infections in Women,” BJU International 130, no. Suppl 3 (November 2022): 11–22, 10.1111/bju.15756. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Trautner B. W., Grigoryan L., Petersen N. J., et al., “Effectiveness of an Antimicrobial Stewardship Approach for Urinary Catheter‐Associated Asymptomatic Bacteriuria,” JAMA Internal Medicine 175, no. 7 (July 2015): 1120–1127, 10.1001/jamainternmed.2015.1878. [DOI] [PubMed] [Google Scholar]
- 22. U.S. Preventive Services Task Force , “Screening for Asymptomatic Bacteriuria in Adults: U.S. Preventive Services Task Force Reaffirmation Recommendation Statement,” Annals of Internal Medicine 149, no. 1 (July 2008): 43–47, 10.7326/0003-4819-149-1-200807010-00009. [DOI] [PubMed] [Google Scholar]
- 23. Caterino J. M., Kline D. M., Leininger R., et al., “Nonspecific Symptoms Lack Diagnostic Accuracy for Infection in Older Patients in the Emergency Department,” Journal of the American Geriatrics Society 67, no. 3 (March 2019): 484–492, 10.1111/jgs.15679. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Mathias S., “Chronic Pelvic Pain: Prevalence, Health‐Related Quality of Life, and Economic Correlates,” Obstetrics & Gynecology 87, no. 3 (March 1996): 321–327, 10.1016/0029-7844(95)00458-0. [DOI] [PubMed] [Google Scholar]
- 25. Jamieson D. and Steege J., “The Prevalence of Dysmenorrhea, Dyspareunia, Pelvic Pain, and Irritable Bowel Syndrome in Primary Care Practices,” Obstetrics & Gynecology 87, no. 1 (January 1996): 55–58, 10.1016/0029-7844(95)00360-6. [DOI] [PubMed] [Google Scholar]
- 26. Grace V. M. and Zondervan K. T., “Chronic Pelvic Pain in New Zealand: Prevalence, Pain Severity, Diagnoses and Use of the Health Services,” Australian and New Zealand Journal of Public Health 28, no. 4 (August 2004): 369–375, 10.1111/j.1467-842x.2004.tb00446.x. [DOI] [PubMed] [Google Scholar]
- 27. Zondervan K. T., Yudkin P. L., Vessey M. P., et al., “Chronic Pelvic Pain in the Community‐‐Symptoms, Investigations, and Diagnoses,” American Journal of Obstetrics and Gynecology 184, no. 6 (May 2001): 1149–1155, 10.1067/mob.2001.112904. [DOI] [PubMed] [Google Scholar]
- 28. Yee H. F. Jr., “The Patient‐Centered Medical Home Neighbor: A Subspecialty Physician's View,” Annals of Internal Medicine 154, no. 1 (January 2011): 63–64, 10.7326/0003-4819-154-1-201101040-00011. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supporting information.
Supporting information.
Supporting information.
Data Availability Statement
Underlying data are available upon request from the corresponding author.