Abstract
Introduction
Urinary symptoms constitute the primary reason for female patients to consult their general practitioner. The urinary dipstick test serves as a cornerstone for diagnosing urinary tract infections (UTIs), yet traditional visual interpretation may be subject to variability. Automated devices for dipstick urinalysis are routinely used as alternatives, yet the evidence regarding their accuracy remains limited. Therefore we aimed to compare concordance between visual and automated urinary dipstick interpretation and determine their test characteristics for the prediction of bacteriuria.
Material and methods
We conducted a prospective validation study including urine samples originating from adult patients in general practice that were sent to the Maastricht Medical Centre + for urinary culture. Urinary dipstick tests were performed on each sample, which were interpreted visually and automatically. We calculated Cohen’s κ and percentage agreement and used 2 × 2 tables to calculate test characteristics.
Results
We included 302 urine samples. Visual and automated analysis showed almost perfect agreement (κ = 0.82 and κ = 0.86, respectively) for both nitrite and leukocyte esterase, but moderate agreement for erythrocytes (κ = 0.51). Interpretation of clinically relevant (nitrite and/or leukocyte esterase positive) samples showed almost perfect agreement (κ = 0.88). Urinary dipsticks show similar test characteristics with urinary culture as gold standard, with sensitivities of 0.92 and 0.91 and specificities of 0.37 and 0.41 for visual and automated interpretation respectively.
Conclusion
Automated and visual dipstick analysis show near perfect agreement and perform similarly in predicting bacteriuria. However, automated analysis requires maintenance and occasionally measurement errors can occur.
Keywords: Urinary tract infections, point-of-care testing, general practice, primary health care, diagnostic equipment
Introduction
Urinary tract complaints are the most common cause for patients to visit their general practitioner (GP). In 2022, 148.5 out of every 1000 general practice consultations in the Netherlands took place because of urinary tract infections (UTIs) [1]. Like in Denmark, Dutch general practitioners diagnose a UTI based on patient symptoms combined with the urinary dipstick test, possibly followed by microbiological analysis at a specialised laboratory [2,3]. In the Netherlands, there is the extra option of a dip slide incubation of the patient’s urine, but dip slides are not accepted by every laboratory for further analysis [2]. Nonetheless, the dipstick test determines the presence of leukocyte esterase and nitrite in urine, which indicate the presence of inflammation and bacteria respectively. Incorrectly analysed dipstick results may lead to overprescription of antibiotics [4]. It is estimated that at least one-third of antibiotic prescriptions are inappropriate, leading to adverse effects and facilitating the rise of antibiotic resistance [5,6].
Visual interpretation of urinary dipsticks is potentially subject to a variety of errors [7,8]. Developers sought to diminish the frequency of these errors with automated dipstick testing. Automated testing makes use of reflectance photometry to interpret the colour change on the dipstick [9]. This would take the burden of interpretation away from the clinician, resulting in a more objective evaluation of results [10,11]. Previous research has shown that point-of-care dipstick urinalysers are able to detect nitrite as well as a standard laboratory test (Urisys 2400, Roche), while leukocyte esterase and erythrocyte were detected with lower sensitivity [9]. Furthermore, agreement with the laboratory standard test was similar for automated and visual interpretation of nitrite, leukocyte esterase and erythrocyte results [12].
Although automated dipstick interpretation has demonstrated parity with the laboratory standard (particularly in interpreting urinary nitrite), Dutch, Danish, Norwegian and Swedish standards for GPs currently omit this laboratory test for UTI diagnosis. Instead, when further diagnostics are warranted, GPs rely on submitting urine samples from suspected UTI patients for urinary culture with antibiotic susceptibility testing [2,3,13,14]. Existing studies do not address potential disparities in outcomes within the decision algorithm outlined in the Dutch guidelines when employing visual versus automated urinalysis. This study seeks to explore the concordance between visual and automated urinary dipstick interpretation and determine their test characteristics for the prediction of bacteriuria.
Materials and methods
Study setting
We conducted a prospective validation study, comparing visual and automated interpretation of urinary dipstick tests. We included urinary samples sent in from general practices (both in and out-of-hours) for urinary culture between January and March 2023. The samples underwent regular diagnostics (Gram staining and urinary culture), before we performed the urinary dipstick test. Results of the urinary culture served as gold standard for determining the sensitivity, specificity, positive predictive value and negative predictive value of the dipstick test.
Inclusion criteria
We included non-catheter urinary samples sent in for urinary culture originating from general practice. We excluded samples originating from patients younger than 18 years of age, since UTI manifests itself differently in children [15].
Urinary culture
A 1 µL inoculation loop was used to inoculate a blood agar plate (Becton Dickinson, Franklin Lakes, New Jersey, United States) containing 5% sheep blood and a cystine–lactose–electrolyte deficient (CLED) agar plate (Becton Dickinson, Franklin Lakes, New Jersey, United States). The plates were aerobically incubated (C 150, Binder GmbH, Tuttlingen, Germany) at 35 degrees Celsius for 24 h. If growth had not developed within the first 24 h, the agars were cultured in the same conditions for another 24 h. If growth was observed, pathogen identification took place using matrix assisted laser desorption/ionization time-of-flight analysis (MALDI-TOF) (VITEK MS™, bioMérieux, Marcy-l’Étoile, France).
Urinary dipstick test
We dipped Combur10 Test UX strips (Roche Diagnostics, Basel, Switzerland, 65821903) in homogenised urine for one second until all test fields had made contact with the sample. Subsequently we transferred the test strip to the Roche Urisys 1100 device (Roche Diagnostics, Basel, Switzerland, 05203281), which uses reflectance photometry to interpret the colour change on the dipstick. We chose this device since it performed the best in previous research [9]. After automated urinalysis, the test strip was read visually by SMLC (biomedical researcher) and PH (medical student), the interpreters being blinded for the result of the automated analysis. Nitrite, leukocyte esterase and erythrocyte fields were compared to the legend on the bottle. Possible results were ‘negative’ and ‘positive’ for nitrite; ‘negative’, ‘1+’, ‘2+’ and ‘3+’ for leukocyte esterase; and ‘negative’, ‘1+’, ‘2+’, ‘3+’ and ‘4+’ for erythrocytes. If interpreters’ results differed, consensus was reached before recording of the result. Quality control of the device was performed weekly via Control-Test M strips (Roche Diagnostics, Basel, Switzerland (10)10088825).
Data analysis
Since Dutch guidelines only distinguish between positive and negative leukocyte esterase and erythrocyte dipstick tests, results were dichotomised to negative (when the result was negative) and positive (when the result was 1+ or higher). Agreement of both interpretation methods was determined using percentage agreement and Cohen’s κ. We determined the test characteristics of the dipstick test using 2 × 2 tables, comparing test results to urinary culture results. We conducted the statistical analyses using IBM SPSS Statistics for Windows (Version 28.0. Armonk, NY: IBM Corp).
Results
Patient characteristics
We measured 307 samples, of which we had to exclude five (1.6%) samples due to measurement errors. Therefore, we included 302 patient samples, of which 205 (67.9%) originated from female patients. The median patient age was 69 years (range 19–98). Roughly half (52.6%) of all included samples were tested positive for bacteriuria with urinary culture.
Agreement of visual and automated dipstick interpretation
Table 1 shows the agreement between the dipstick results as rated by visual and automated interpretation. The percentage of test results that were rated the same between tests was >90% for the nitrite and leukocyte tests, but was lower for the erythrocyte test. Cohen’s κ shows almost perfect agreement between visual and automated interpretation for nitrite and leukocyte esterase. Full contingency tables are available in Appendices 1–3.
Table 1.
Agreement of visual and automated interpretation of urinary dipsticks.
| Percent agreement | Cohen’s κ (95% CI) | |
|---|---|---|
| Nitrite | 92.1% | 0.82 (0.81–0.83) |
| Leukocyte esterase | 94.7% | 0.86 (0.80–0.91) |
| Erythrocytes | 83.4% | 0.51 (0.43–0.58) |
CI, confidence interval.
Clinical evaluation
The decision algorithms of the Dutch and Danish UTI guidelines consider nitrite and/or leukocyte esterase positive urine as potentially constituting a UTI [2]. Table 2 shows the agreement of visual and automated urinary dipstick results in determining potential UTIs. The methods show almost perfect agreement.
Table 2.
Agreement of visual and automated interpretation on determining clinically relevant samples.
| Automated visual |
Negative | Positive |
|---|---|---|
| Negative | 64 | 4 |
| Positive | 9 | 225 |
| Cohen’s κ (95% CI) | 0.88 (0.82–0.94) |
CI, confidence interval.
To determine whether automated interpretation would lead to a different diagnosis compared to visual interpretation, we compared results of both methods to the result of the urinary culture. Table 3 shows that both methods have >90% sensitivity, but a specificity <50%. Test characteristics did not differ significantly between methods. Additionally, Appendices 4 and 5 show high specificity and low sensitivity for only the nitrite test for both interpretation methods, and vice versa for only the leukocyte esterase test. Again, there is no significant difference in test characteristics between the interpretation methods.
Table 3.
Test characteristics of visual and automated interpretation with urinary culture as gold standard.
| Culture visual |
Negative | Positive | Culture automated |
Negative | Positive |
|---|---|---|---|---|---|
| Negative | 53 | 15 | Negative | 59 | 14 |
| Positive* | 90 | 144 | Positive* | 84 | 145 |
| Test characteristics visual | 95%CI | Test characteristics automated | 95% CI | ||
| Sensitivity | 0.91 | 0.86–0.95 | Sensitivity | 0.91 | 0.87–0.96 |
| Specificity | 0.37 | 0.29–0.45 | Specificity | 0.41 | 0.33–0.49 |
| PPV | 0.62 | 0.55–0.68 | PPV | 0.63 | 0.57–0.70 |
| NPV | 0.78 | 0.60–0.71 | NPV | 0.81 | 0.72–0.90 |
*Positive for nitrite and/or leukocyte esterase.
CI, confidence interval; NPV, negative predictive value; PPV, positive predictive value.
Discussion
Summary
This study sought to explore the concordance between visual and automated urinary dipstick interpretation and determine their test characteristics for the prediction of bacteriuria. Visual and automated interpretation of urinary dipstick results show almost perfect agreement for the nitrite and leukocyte esterase test. However, the two methods agree only moderately for interpretation of the erythrocyte result. When looking at the clinically relevant outcomes as described in the Dutch and Danish guidelines, the two methods show almost perfect agreement. Test characteristics of the two methods do not differ significantly, with high sensitivity but poor specificity.
Strengths and limitations
This study was a prospective validation study in which we were able to systematically collect many samples in a short time-period. We compared the tests to the current gold standard and the point-of-care analyser has been previously evaluated and validated in routine general practice [9, 12]. Furthermore, the tests were performed on clinically relevant samples, representing the general population of samples that was sent in for urinary culture.
This study also has its limitations. The visual interpreters of the dipstick test were not general practice nurses or general practitioners. Furthermore, we have no records of the individual observers; therefore we were unable to determine inter-observer reliability. Additionally, results obtained from urine samples sent in for culture are not generalisable to all samples in general practice, since Dutch guidelines suggest only ordering urinary culture in case of high-risk patients [2]. High-risk patients include frail elderly patients with therapy failure, which might explain the relatively high age of the patients included in this study. Previous studies showed average ages of 43–54 years for patients consulting their GP with urinary symptoms, considerably lower than the average age in our study [4, 16–18]. Furthermore, since we did not have access to the sample until arrival at the hospital (and therefore were not able to speak to the patients), we were unaware of any symptoms they might have had. Therefore, it was impossible for us to differentiate between asymptomatic bacteriuria and actual urinary tract infections. A patient has asymptomatic bacteriuria when bacteria are present in their urine without causing symptoms typically associated with UTI. Asymptomatic patients do not need to be treated, but since their diagnostic results show bacteria to be present in their urine they often still are. This results in overprescription of antibiotics, leading to antimicrobial resistance [19]. Asymptomatic bacteriuria is more prevalent with age and might therefore be overrepresented in our population [20]. However, since Dutch guidelines suggest urinary culture only in high-risk patients (patients with recurrent UTI, vulnerable elderly, diabetic patients, patients with signs of tissue invasion, such as fever, shivers, signs of sepsis, etc.) we assume that culture positive cases in our study have a high likelihood of being an actual UTI. Additionally, due to our inclusion method we have no records of the time and manner of sample voiding. Storage details at the general practices were also unknown. We suggest repeating this study in daily general practice with records of patient symptoms in order to draw definitive conclusions.
Comparison with existing literature
In this study we show a direct comparison between visual and automated interpretation of the urinary dipstick. We see high agreement between the two methods for the nitrite and leukocyte esterase test, but only moderate agreement for the erythrocyte test. These findings corroborate the results found by Schot et al. who show similar agreements between the Urisys 1100 and the laboratory standard test [9]. The low agreement for erythrocytes seen in our study is mainly caused by samples that were scored negative during visual interpretation, while scoring positive during automated interpretation. We think this has mainly to do with the legend on the tube of the Combur10 Test UX strips, since there it shows ‘1+’ as several green spots on a yellow background, while the automated analysis already scored samples positive when only 1 or 2 spots were visible. This disparity is not as relevant for Dutch GPs, since, like in Denmark, erythrocytes are omitted from the decision algorithm [2,3].
Furthermore, we show that there is no significant difference between visual and automated interpretation of urinary dipstick results for urinary tract infections in terms of clinically relevant samples. This only rings true if the dipstick test is performed as instructed by the manufacturer and is read within 2 min. Though easy to use, the Urisys 1100 needs to be calibrated weekly and cleaned properly after every sample to prevent unreliable results [21]. False positives in the automated dipstick analyses may be due to the discolouration of the dipstick, caused by haematuria, concentrated urine, or smudging from adjacent test fields. On five separate occasions an error occurred during automated interpretation, which we suspect was due to the sample being highly positive for nitrite. In these cases reagent from the nitrite field would spill over to adjacent fields, contaminating the signal in those fields in the process. Visually the samples were still able to be interpreted, but since the machine is calibrated for certain wavelengths in specific fields, it will cause an error when a different wavelength is measured.
Implications for research and/or practice
The test characteristics of the urinary dipstick is similar for both methods when following the decision algorithm of the Dutch and Danish guidelines [2,3]. According to these guidelines, a patient whose urine tests positive for nitrite (when patient symptoms make a UTI plausible), is to be diagnosed with a UTI. A negative nitrite result combines with a positive leukocyte esterase result is cause for further diagnostic tests (urine culture, dip slide, or sediment). This is because previous research has shown the nitrite test to be specific but non-sensitive for UTI, corroborating the results we have found in our study. The inverse is true for the leukocyte esterase test, which is sensitive but non-specific [22–25]. Results of our research underline once again that the urinary dipstick on its own is not accurate enough to predict UTI reliably, but that GPs should consider other factors such as patient symptoms [23, 26–28]. Currently, GPs in Norway and Sweden do not routinely employ the urinary dipstick. For acute cystitis, Norwegian and Swedish GPs base their diagnosis on patient symptoms alone and the urinary dipstick is only used in severe cases [13,14]. However, Bollestad et al. suggested to reconsider the current way of diagnosing UTI, since they showed that symptom severity did not correlate with bacteriuria or symptom duration [29].
Conclusion
Both visual and automated dipstick analyses provide fast results when performed complying with the instructions for use. Automated and visual dipstick analysis perform similarly in predicting bacteriuria when read and interpreted by researchers in a central laboratory. However, automated analysis requires maintenance and occasionally measurement errors can occur.
Ethical approval
The medical ethical board of the Maastricht University Medical Centre + approved this study and decided that it is not subjected to the Dutch Medical Research with Human Subjects Law (ref METC 2022-3363).
Appendices.
Appendix 1. Comparison of visual and automated interpretation of the nitrite test
| Automated visual |
Negative | Positive |
|---|---|---|
| Negative | 192 | 18 |
| Positive | 6 | 86 |
| 95% CI | ||
| Cohen’s κ | 0.82 | (0.81–0.83) |
CI = confidence interval.
Appendix 2. Comparison of visual and automated interpretation of the leukocyte esterase test
| Automated visual |
Negative | Positive |
|---|---|---|
| Negative | 65 | 7 |
| Positive | 9 | 221 |
| 95% CI | ||
| Cohen’s κ | 0.86 | (0.80–0.91) |
CI = confidence interval.
Appendix 3. Comparison of visual and automated interpretation of the erythrocyte test
| Automated visual |
Negative | Positive |
|---|---|---|
| Negative | 37 | 47 |
| Positive | 3 | 215 |
| 95% CI | ||
| Cohen’s κ | 0.51 | (0.43–0.58) |
CI = confidence interval.
Appendix 4. Test characteristics of visual and automated interpretation of the nitrite test
| Culture visual |
Negative | Positive | Culture automated |
Negative | Positive |
|---|---|---|---|---|---|
| Negative | 139 | 71 | Negative | 130 | 68 |
| Positive* | 4 | 88 | Positive* | 13 | 91 |
| Test characteristics visual | 95%CI | Test characteristics automated | 95% CI | ||
| Sensitivity | 0.55 | 0.48–0.63 | Sensitivity | 0.57 | 0.50–0.65 |
| Specificity | 0.97 | 0.95–1.00 | Specificity | 0.91 | 0.86–0.96 |
| PPV | 0.96 | 0.92–1.00 | PPV | 0.88 | 0.81–0.94 |
| NPV | 0.66 | 0.60–0.73 | NPV | 0.66 | 0.59–0.72 |
Based on a prevalence of 0.53.
*Positive for nitrite and/or leukocyte esterase.
CI, confidence interval; NPV, negative predictive value; PPV, positive predictive value.
Appendix 5. Test characteristics of visual and automated interpretation of the combined nitrite and leukocyte esterase test
| Culture visual |
Negative | Positive | Culture automated |
Negative | Positive |
|---|---|---|---|---|---|
| Negative | 53 | 19 | Negative | 59 | 15 |
| Positive* | 90 | 140 | Positive* | 84 | 144 |
| Test characteristics visual | 95%CI | Test characteristics automated | 95% CI | ||
| Sensitivity | 0.88 | 0.83–0.93 | Sensitivity | 0.91 | 0.86–0.95 |
| Specificity | 0.37 | 0.29–0.45 | Specificity | 0.41 | 0.33–0.49 |
| PPV | 0.61 | 0.55–0.67 | PPV | 0.63 | 0.57–0.69 |
| NPV | 0.74 | 0.63–0.84 | NPV | 0.80 | 0.71–0.89 |
Based on a prevalence of 0.53.
*Positive for nitrite and/or leukocyte esterase.
CI, confidence interval; NPV, negative predictive value; PPV, positive predictive value.
Funding Statement
This work was supported by The Netherlands Organisation for Health Research and Development under Grant 10150511910060.
Disclosure statement
No potential conflict of interest was reported by the author(s).
References
- 1.Heins MBJ, Weesie Y, Davids R, et al. Zorg door de huisarts: jaarcijfers 2022 en trendcijfers 2018-2022. 2023. [Google Scholar]
- 2.Bouma MGS, Klinkhamer S, Knottnerus BJ, et al. 2020. NHG-Standaard Urineweginfecties (M05). Nederlands Huisartsen Genootschap . [Google Scholar]
- 3.Medicin DSfA . Urinvejsinfektioner i almen praksis. 2020. [cited 2024 Jun 10]. Available from: https://www.dsam.dk/vejledninger/urinvejsinfektioner/urinvejsinfektioner-i-almen-praksis#diagnostik
- 4.Spek M, Cals JWL, Oudhuis GJ, et al. Workload, diagnostic work-up and treatment of urinary tract infections in adults during out-of-hours primary care: a retrospective cohort study. BMC Fam Pract. 2020;21(1):231. doi: 10.1186/s12875-020-01305-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Butler CCFN, Thomas-Jones E, Llor C, et al. Variations in presentation, management, and patient outcomes of urinary tract infection: a prospective four-country primary care observational cohort study. Br J Gen Pract. 2017;67(665):e830–e841. doi: 10.3399/bjgp17X693641. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Lipsitch M, Samore MH.. Antimicrobial use and antimicrobial resistance: a population perspective. Emerg Infect Dis. 2002;8(4):347–354. doi: 10.3201/eid0804.010312. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Winkens RA, Leffers P, Degenaar CP, et al. The reproducibility of urinalysis using multiple reagent test strips. Eur J Clin Chem Clin Biochem. 1991;29(12):813–818. doi: 10.1515/cclm.1991.29.12.813. [DOI] [PubMed] [Google Scholar]
- 8.Raijmakers M, Oosterhuis W, Kleinveld H.. Preanalytische kwaliteit van urinediagnostiek bij de huisarts. Ned Tijdschr Klin Chem Labgeneesk. 2011;36:77. [Google Scholar]
- 9.Schot MJ, van Delft S, Kooijman-Buiting AM, et al. Analytical performance, agreement and user-friendliness of six point-of-care testing urine analysers for urinary tract infection in general practice. BMJ Open. 2015;5(5):e006857. doi: 10.1136/bmjopen-2014-006857. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Peele JD, Jr., Gadsden RH, Crews R.. Semi-automated vs. visual reading of urinalysis dipsticks. Clin Chem. 1977;23(12):2242–2246. doi: 10.1093/clinchem/23.12.2242. [DOI] [PubMed] [Google Scholar]
- 11.Tighe P. Urine dry reagent strip “error” rates using different reading methods. Accredit Qual Assur. 2000;5(12):488–490. doi: 10.1007/s007690000232. [DOI] [Google Scholar]
- 12.van Delft S, Goedhart A, Spigt M, et al. Prospective, observational study comparing automated and visual point-of-care urinalysis in general practice. BMJ Open. 2016;6(8):e011230. doi: 10.1136/bmjopen-2016-011230. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Folkhälsomyndigheten L. 2023. Behandlingsrekommendationer för vanliga infektioner i öppenvård. [cited 2024 Jun 10]. Available from: https://folkhalsomyndigheten.se/publikationer-och-material/publikationsarkiv/b/behandlingsrekommendationer-for-vanliga-infektioner-i-oppenvard/
- 14.Helsedirektoratet . 2022. Akutt cystitt hos ikke-gravid kvinne 15 - 65 år Antibiotikasenteret for primaermedisin (ASP). [cited 2024 Jun 10]. Available from: https://antibiotikaiallmennpraksis.no/index.php?action=topic&item=f652b4a03e5ba0f653aa
- 15.Leung AKC, Wong AHC, Leung AAM, et al. Urinary tract infection in children. Recent Pat Inflamm Allergy Drug Discov. 2019;13(1):2–18. doi: 10.2174/1872213X13666181228154940. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Fanshawe TR, Judge RK, Mort S, et al. Evidence-based appraisal of two guidelines for the diagnosis of suspected, uncomplicated urinary tract infections in primary care: a diagnostic accuracy validation study. J Antimicrob Chemother. 2023;78(8):2080–2088. doi: 10.1093/jac/dkad212. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Hullegie S, Wootton M, Verheij TJM, et al. Clinicians’ interpretations of point of care urine culture versus laboratory culture results: analysis from the four-country POETIC trial of diagnosis of uncomplicated urinary tract infection in primary care. Fam Pract. 2017;34(4):392–399. doi: 10.1093/fampra/cmx009. [DOI] [PubMed] [Google Scholar]
- 18.Moore M, Trill J, Simpson C, et al. Uva-ursi extract and ibuprofen as alternative treatments for uncomplicated urinary tract infection in women (ATAFUTI): a factorial randomized trial. Clin Microbiol Infect. 2019;25(8):973–980. doi: 10.1016/j.cmi.2019.01.011. [DOI] [PubMed] [Google Scholar]
- 19.Cortes-Penfield NW, Trautner BW, Jump RLP.. Urinary tract infection and asymptomatic bacteriuria in older adults. Infect Dis Clin North Am. 2017;31(4):673–688. doi: 10.1016/j.idc.2017.07.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Nicolle LE. Asymptomatic bacteriuria in the elderly. Infect Dis Clin North Am. 1997;11(3):647–662. doi: 10.1016/s0891-5520(05)70378-0. [DOI] [PubMed] [Google Scholar]
- 21.Urisys® 1100 analyzer operator’s manual version 10 . 2023. [Cited 2024 Jun 10]. Available from: https://elabdoc-prod.roche.com/eLD/api/downloads/13a9ee25-a3ca-ea11-fe90-005056a772fd?countryIsoCode=nl
- 22.Devillé WL, Yzermans JC, van Duijn NP, et al. The urine dipstick test useful to rule out infections. A meta-analysis of the accuracy. BMC Urol. 2004;4(1):4. doi: 10.1186/1471-2490-4-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Little P, Turner S, Rumsby K, et al. Developing clinical rules to predict urinary tract infection in primary care settings: sensitivity and specificity of near patient tests (dipsticks) and clinical scores. Br J Gen Pract. 2006;56(529):606–612. [PMC free article] [PubMed] [Google Scholar]
- 24.Nys S, van Merode T, Bartelds AIM, et al. Urinary tract infections in general practice patients: diagnostic tests versus bacteriological culture. J Antimicrob Chemother. 2006;57(5):955–958. doi: 10.1093/jac/dkl082. [DOI] [PubMed] [Google Scholar]
- 25.Chernaya A, Søborg C, Midttun M.. Validity of the urinary dipstick test in the diagnosis of urinary tract infections in adults. Dan Med J. 2021;69(1) [PubMed] [Google Scholar]
- 26.McIsaac WJ, Moineddin R, Ross S.. Validation of a decision aid to assist physicians in reducing unnecessary antibiotic drug use for acute cystitis. Arch Intern Med. 2007;167(20):2201–2206. doi: 10.1001/archinte.167.20.2201. [DOI] [PubMed] [Google Scholar]
- 27.Bent S, Nallamothu BK, Simel DL, et al. Does this woman have an acute uncomplicated urinary tract infection? JAMA. 2002;287(20):2701–2710. doi: 10.1001/jama.287.20.2701. [DOI] [PubMed] [Google Scholar]
- 28.Knottnerus BJ, Geerlings SE, Moll van Charante EP, et al. Toward a simple diagnostic index for acute uncomplicated urinary tract infections. Ann Fam Med. 2013;11(5):442–451. doi: 10.1370/afm.1513. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Bollestad M, Vik I, Grude N, et al. Predictors of symptom duration and bacteriuria in uncomplicated urinary tract infection. Scand J Prim Health Care. 2018;36(4):446–454. doi: 10.1080/02813432.2018.1499602. [DOI] [PMC free article] [PubMed] [Google Scholar]