[Skip to Navigation]
Sign In
October 2005

Choice of Urine Collection Methods for the Diagnosis of Urinary Tract Infection in Young, Febrile Infants

Author Affiliations

Author Affiliations: Departments of Pediatrics (Drs Schroeder, Newman, and Pantell) and Epidemiology (Dr Newman), University of California, San Francisco; Pediatric Research in the Office Setting, Department of Practice and Research, Center for Child Health Research, American Academy of Pediatrics, Elk Grove Village, Ill (Dr Wasserman and Ms Finch); and Department of Pediatrics, University of Vermont, Burlington (Dr Wasserman). Dr Schroeder is currently at the Department of Pediatrics, Stanford University School of Medicine, Stanford, Calif.

Arch Pediatr Adolesc Med. 2005;159(10):915-922. doi:10.1001/archpedi.159.10.915

Background  The optimal method of urine collection in febrile infants is debatable; catheterization, considered more accurate, is technically difficult and invasive.

Objectives  To determine predictors of urethral catheterization in febrile infants and to compare bag and catheterized urine test performance characteristics.

Design  Prospective analysis of infants enrolled in the Pediatric Research in Office Settings’ Febrile Infant Study.

Setting  A total of 219 practices from within the Pediatric Research in Office Settings’ network, including 44 states, the District of Columbia, and Puerto Rico.

Patients  A total of 3066 infants aged 0 to 3 months with temperatures of 38°C or higher.

Main Outcome Measures  We calculated adjusted odds ratios for predictors of catheterization. Diagnostic test characteristics were compared between bag and catheterization. Urinary tract infection was defined as pure growth of 100 000 CFU/mL or more (bag) and 20 000 CFU/mL or more (catheterization).

Results  Seventy percent of urine samples were obtained by catheterization. Predictors of catheterization included female sex, practitioner older than 40 years, Medicaid, Hispanic ethnicity, nighttime evaluation, and severe dehydration. For leukocyte esterase levels, bag specimens demonstrated no difference in sensitivity but somewhat lower specificity (84% [bag] vs 94% [catheterization], P<.001) and a lower area under the receiver operating characteristic curve for white blood cells (0.71 [bag] vs 0.86 [catheterization], P = .01). Infection rates were similar in bag and catheterized specimens (8.5% vs 10.8%). Ambiguous cultures were more common in bag specimens (7.4% vs 2.7%, P<.001), but 21 catheterized specimens are needed to avoid each ambiguous bag result.

Conclusions  Most practitioners obtain urine from febrile infants via catheterization, but choice of method is not related to the risk of urinary tract infection. Although both urine cultures and urinalyses are more accurate in catheterized specimens, the magnitude of difference is small but should be factored into clinical decision making.

Urinary tract infections (UTIs) are the most common cause of serious bacterial infections in febrile infants younger than 90 days.1-7 Diagnosis requires collection of urine generally by 1 of 4 methods: sterile urine bag, urethral catheterization (CATH), suprapubic aspiration (SPA), or clean-catch (CC). Both CATH and SPA are thought to yield the most reliable results by minimizing false-positive results, but these methods are invasive and painful. The bag method is a noninvasive and easy alternative but has been criticized as having high false-positive rates,8-12 prompting the American Academy of Pediatrics (AAP) to discourage its use for urine cultures in infants.8 Choosing the appropriate urine collection method presents the practitioner with a difficult decision, and little is known about how this decision is made.

Once urine is collected, urinalysis (UA) is often performed as a screen for UTIs. Although substantial data exist on the diagnostic test characteristics of UA,13 few studies have been performed in the 0- to 3-month-old age group, and we know of no studies that compare the performance of the UA across urine collection methods.

A prior publication7 from the Pediatric Research in Office Settings’ (PROS) Febrile Infant Study reported that urine was tested in only 54% of febrile infants aged 0 to 93 days. Urine testing was more likely in infants with a higher temperature, ill appearance, younger age, and lack of a fever source. This report will address the following questions: (1) Which urine collection techniques do practitioners use to diagnose UTIs in young, febrile infants? (2) What are the infant and practitioner characteristics that predict the choice of urine collection method used? (3) Do the diagnostic test characteristics of UA differ between bag and CATH methods? (4) Do the rates of positive culture results and ambiguous urine culture results differ between methods?


Data are presented from infants prospectively enrolled in the PROS Febrile Infant Study. The methods of the PROS Febrile Infant Study have been previously described7,14 and are summarized briefly herein.

Pros febrile infant study

The PROS is the practice-based research network of the AAP. The Febrile Infant Study involved 573 practitioners from 219 practices from within the PROS network. Practitioners represented 44 states, the District of Columbia, and Puerto Rico.


Inclusion criteria for this study were (1) age of 93 days or younger; (2) axillary, rectal, or tympanic temperature of 38°C or higher in the office or in the previous 24 hours at home; and (3) initial examination by a PROS practitioner. A total of 3066 eligible patients were enrolled in the study between February 28, 1995, and April 25, 1998. However, not all of these patients underwent UAs and urine cultures, and the sample size varies for each analysis and is specified in the results. To analyze predictors of CATH, we included all patients who underwent UA and/or urine culture. To analyze the test performance characteristics of the UA, we included patients who underwent both a UA and urine culture. To analyze urine culture results, we included all patients who had urine cultures. Although some infants had more than 1 urine culture, we limited all of our analyses to tests performed on the day of initial presentation. Furthermore, because so few patients had urine collected by CC or SPA, we chose to focus the analyses on those patients who had specimens obtained by bag or CATH.

Because tests and treatments were ordered at the discretion of the individual practitioners according to their customary clinical care, informed consent was not required. The Committee on Human Research from the University of California, San Francisco, approved the study.

Demographic and clinical characteristics

Multiple aspects of the infants’ clinical histories and physical examination findings were recorded, predominantly by checking boxes on the study form. The temperature variable reflected the higher of the temperatures taken in the office or at home. For axillary temperatures, 0.5°C was added to the reported temperature. Categorical variables such as hydration status were assumed to be normal if abnormalities were not documented. Infant and practitioner demographic characteristics were documented. The study protocol required that the practitioner’s diagnostic impression and assessment of severity of illness be recorded before laboratory test results were available.

Laboratory data

All practices were supplied with urine dipsticks (Ames-Multistix; Miles Inc, Elkhart, Ind) in an attempt to facilitate comparisons of results. All other laboratory testing, including urine microscopy and culture, was performed in the laboratories normally used by the practitioners. All laboratory tests were ordered in accordance with the practitioners’ judgment.

For UAs, practitioners recorded the method of collection, urine dipstick results (leukocyte esterase [LE], nitrite, and blood levels), and urine microscopy results (white blood cell count, red blood cell count, and whether bacteria were seen). The choice of urine collection method was at the practitioners’ discretion. A UA was considered to be complete if one of the following was recorded: LE level, nitrite level, blood level, urine white blood cell count, urine red blood cell count, or bacterial presence. For urine cultures, practitioners recorded the date that the culture was performed, whether it was positive or negative, the method of collection, and the types and levels of organism(s).

The diagnosis of UTI was based on urine culture results, independent of UA results, and was defined a priori. The definition of a positive urine culture required (1) classification of the organism as pathogenic or sometimes pathogenic by a pediatric infectious disease consultant who was blinded to data on individual subjects and (2) a minimum number of colony-forming units per milliliter, depending on the urine collection method. For SPA specimens, at least 100 CFU/mL of 1 or more pathogenic organisms was required. For CATH specimens, 20 000 CFU/mL of a single pathogenic organism was required. For bag and CC specimens, at least 100 000 CFU/mL of a single pathogenic organism was required. Results were occasionally reported as greater than or less than, predominantly in the form of “greater than 100 000 CFU/mL” or “less than 10 000 CFU/mL.” To facilitate tabulations and statistical analysis, if the culture result was reported as less than a certain value, the value was divided in half. If the culture was reported as greater than a certain value, the value was doubled.

We defined ambiguous cultures as having either (1) the minimum number of colonies to qualify as a UTI for one organism but also having a second organism present or (2) having 20% to 99% of the minimum number of colonies of pure growth. Examples of ambiguous cultures include a CATH urine culture that had 20 000 CFU/mL of Escherichia coli but also contained a second organism or a bag urine culture that contained 50 000 CFU/mL of E coli (pure growth).

Statistical analysis

To analyze predictors of urine collection methods (Table 1), we used a multivariate logistic regression model (Stata version 7; Stata Corp, College Station, Tex) to calculate adjusted odds ratios for CATH, adjusting for clustering by physician. For this model, we identified 11 clinically and biologically plausible predictor variables (Table 2), then used a backward stepwise elimination model with a P to remove of <.06. An infant was considered catheterized if either the UA or urine culture was obtained by CATH. The catheterization variable was designated as missing if urine was obtained by CC or SPA or if the method was not documented. Goodness of fit of the multivariate model was assessed using the Hosmer-Lemeshow method with 10 groups.15 Discrimination was assessed using the c-statistic, equal to the area under the receiver operating characteristic curve.15

Table 1. 
Urinalysis and Urine Culture Collection Methods for Day of Presentation
Urinalysis and Urine Culture Collection Methods for Day of Presentation
Table 2. 
Unadjusted Associations Between Infant or Practitioner Characteristics and CATH in Patients Who Had Urinalyses and/or Urine Cultures Obtained by Either Bag or CATH on the Day of Presentation
Unadjusted Associations Between Infant or Practitioner Characteristics and CATH in Patients Who Had Urinalyses and/or Urine Cultures Obtained by Either Bag or CATH on the Day of Presentation

To compute characteristics of the UA as a diagnostic test for UTI, the gold standard was the urine culture result, where ambiguous cultures were considered negative. We included cases where the UA method was different from the urine culture method. If a method was noted for one test but absent for the other, we assumed that the same method was used for both.

To determine whether the urine culture method was an independent predictor of a positive culture, a multivariate logistic regression model was used, with UTI as an outcome and bag vs CATH as a predictor. Variables determined from our previous study to increase or decrease the risk of UTI (sex, circumcision, maximum temperature, ill family members, fever duration, inconsolability, Hispanic ethnicity, and respiratory distress)7 were included as covariates.


Urine collection methods

At the initial visit, a UA was performed on 1639 (53%) of the 3066 infants, and a urine culture was obtained from 1605 (52%) of the infants (Table 1). A total of 1763 patients had either a UA or urine culture, whereas 1482 patients had both a UA and urine culture at presentation. Most urine specimens were obtained by CATH (63% of UAs and 69% of urine cultures). Bag specimens composed 31% of UAs and 24% of urine cultures.

Predictors of catheterization

A total of 1646 patients had urine specimens obtained on the day of presentation by either bag or CATH; 70% were catheterized either for UA or urine culture, and 30% had bag urine testing only. The highest frequencies of catheterization were found in the following subgroups (Table 2): patients with severe dehydration (88%), patients who were seen by younger (<40 years) practitioners (82%), female infants (77%), Medicaid patients (77%), and patients who were seen after hours (76%). Circumcision, height of fever, and general appearance did not predict catheterization. The final multivariate logistic model for predictors of catheterization (Table 3) demonstrated good fit (Hosmer-Lemeshow χ29 = 3.8; P = .70). Discrimination of the model was fair (C = 0.69).

Table 3. 
Multivariate Model for Predictors of CATH
Multivariate Model for Predictors of CATH

Ua test characteristics

Of the 1482 patients who had both UAs and urine cultures, 1384 had the samples obtained by bag or CATH. Overall, LE had higher sensitivity, whereas nitrites demonstrated better specificity (Table 4). Sensitivity and specificity were higher in CATH specimens compared with bag specimens for both LE and nitrites, but the only statistically significant difference was the comparison of specificity of LE (84% [bag] vs 94% [CATH], P<.001). These calculations changed very little (differences of <2%) if the UTI threshold colony count for CATH specimens was reduced to 10 000 CFU/mL or if the threshold for bag specimens was reduced to 50 000 CFU/mL.

Table 4. 
Sensitivity and Specificity of Leukocyte Esterase and Nitrites as Diagnostic Tests for Urinary Tract Infections Stratified by Method of Collection
Sensitivity and Specificity of Leukocyte Esterase and Nitrites as Diagnostic Tests for Urinary Tract Infections Stratified by Method of Collection

To further examine the difference in specificities of LE between methods, we analyzed the 54 patients who had bag UAs with false-positive results for LE. Of the patients who were also tested for nitrites, only 4 (8%) of 51 had positive results. Of the patients who were also tested for urine white blood cell counts, 9 (19%) of 47 had more than 10 white blood cells per high-power field. Ambiguous cultures occurred in 13 (24%) of 54 of these patients. If patients who had specimens with positive LE test results and positive nitrite test results, more than 10 white blood cells per high-power field, or ambiguous culture results are considered to be positive for UTI, the difference between methods in specificity for LE is still significant (89% [bag] vs 95% [CATH], P<.001).

The CATH specimens performed better for urine white blood cell counts as well (Table 5). When compared with bag specimens, the area under the receiver operating characteristic curve (C statistic) for urine white blood cell counts and UTI (Figure) is higher in patients with CATH specimens (0.86 vs 0.71, P = .01) Again, areas under the curve did not change significantly (<.01) when thresholds for UTI were changed to 10 000 CFU/mL for CATH specimens and/or 50 000 CFU/mL for bag specimens.

Receiver operating characteristic (ROC) curve for urine white blood cell counts and urinary tract infection stratified by bag and urethral catheterization (CATH).

Receiver operating characteristic (ROC) curve for urine white blood cell counts and urinary tract infection stratified by bag and urethral catheterization (CATH).

Table 5. 
Likelihood Ratios for Urine White Blood Cell Counts as a Diagnostic Test for Urinary Tract Infections Stratified by Urine Culture Collection Method
Likelihood Ratios for Urine White Blood Cell Counts as a Diagnostic Test for Urinary Tract Infections Stratified by Urine Culture Collection Method

Urine culture results

Urine cultures were obtained by CATH or bag in 1490 patients. Compared with CATH, bag urine cultures were more likely to have 2 organisms, to have nonpathogenic bacteria, and to have an ambiguous result (Table 6). The relative risk of an ambiguous culture for specimens obtained by bag was 2.7 (95% confidence interval [CI], 1.7-4.5); however, the absolute risk was small (7.4% [bag] vs 2.7% [CATH]). Twenty-one cultures (95% CI, 13-53) would have to be obtained by CATH to avoid 1 ambiguous culture obtained by bag. The unadjusted UTI rates were similar between bag and CATH methods (8.5% vs 10.8%, P = .19). The adjusted odds ratio for a UTI by bag when compared with CATH was 0.9 (95% CI, 0.6-1.3; P = .51).

Table 6. 
Urine Culture Results Stratified by Method of Collection
Urine Culture Results Stratified by Method of Collection


Most urine specimens in this study were obtained by catheterization. The strongest predictors of CATH were practitioner age and infant sex. Surprisingly, neither height of fever nor ill appearance was associated with increased catheterization. This suggests that practitioners do not base decisions to catheterize on the pretest probability of UTI. Although female infants are at high risk for UTI, uncircumcised males are at an even higher risk7,16-19 but were significantly less likely than females to be catheterized. This discrepancy may be due to perceived technical difficulties or other unmeasured factors. The association between young practitioner age and increased catheterization may reflect evolving training practices, the fact that younger practitioners tend to adhere more closely to guidelines than older practitioners,20,21 or a propensity by younger practitioners to do or order more procedures.

Regarding the test performance characteristics of the UA, our findings for LE and nitrites were consistent with prior studies,13,22 specifically in regard to the low sensitivity and high specificity of nitrites. The differences in sensitivities between methods for both LE and nitrites were not statistically significant, but the CIs were wide, reflecting the lower sample size of patients with UTIs. The extremely low sensitivity of the nitrite test for both bag and CATH methods (25% vs 43%) demonstrates that practitioners should not rely on this test to rule out UTIs. For LE, the sensitivities were much higher (76% vs 86%, P = .19) for bag and CATH specimens. Should future studies document a statistically significant difference in sensitivity between methods of 10%, this could potentially influence management decisions, particularly for those practitioners who use the UA results to guide the decision to obtain a culture, hospitalize, or provide immediate antibiotic treatment while awaiting culture results.

The decreased specificity of LE in bag UAs was statistically significant, indicating that there are increased false-positive LE test results on bag UAs. This finding could be due to inflammatory processes in the perineal or periurethral area unrelated to UTIs that are detected by bag but not CATH specimens. Alternatively, there may have been false-negative bag cultures that led to the appearance of false-positive UA results. Bag cultures are generally thought to be highly sensitive,8 but sensitivity depends on the definition of UTI. In this study, we used fairly strict criteria for a diagnosis of UTI by bag cultures. Furthermore, perineal cleansing techniques likely varied between practices, and if cleansing materials with antimicrobial properties interfered with the bag urine culture, false-negative cultures may have resulted. However, the difference in specificities between methods persisted even after patients with increased likelihood of a false-negative urine culture (positive nitrites or >10 white blood cells per high-power field on UA or an ambiguous culture result) were considered to have a UTI.

To our knowledge, this is the first study to find that the performance of the UA varies by method. Further studies that compare urinalyses obtained simultaneously in the same patient by both methods would help to substantiate or refute our findings.

Bag urine cultures were more likely to have more than 1 organism, to yield nonpathogenic bacteria, and to have bacterial growth in insufficient quantity to qualify as a UTI. In prior studies,10,11 investigators have classified cultures of this sort as false positive and have reported poor specificity of urine bag cultures. Such classification is misleading, however, because it assumes that practitioners will interpret results of this nature as true UTIs. In our analysis, we attempted to define ambiguous cultures (ie, culture results that are most likely to cause diagnostic uncertainty). Although ambiguous culture results were nearly 3 times as likely in bag specimens, the risk in both groups and the absolute difference between the risks were low (7% vs 3%), translating into a high number (n=21) needed to catheterize to avoid 1 ambiguous bag specimen.

The overall rate of UTI for urine cultures obtained on the day of presentation was slightly higher in catheterized infants than in infants who had urine obtained by bag. These findings are not attributable to preferential catheterization of high-risk infants and challenge the notion that false-positive results are more common in bag urine cultures. If bag cultures yield high rates of false-positive results, then the method of collection should be an independent predictor of UTI after adjustment for other UTI risk factors. Our analysis demonstrates that this was not the case. Theoretically, however, false-positive bag cultures may have been balanced by false-negative bag cultures.

Limited data exist addressing the comparison between bag and CATH cultures. The AAP’s practice parameter on UTI8 cites 3 references to support the high false-positive rate of bag urine cultures23-25 and recommends obtaining urine by CATH or SPA. However, none of the studies cited analyzes urine cultures obtained by CATH. The study by Sorensen et al25 is on incontinence and school-aged girls and has no mention of specific urine collection methods. In the studies by Shannon et al23 and Leong and Tan,24 SPA specimens were obtained on infants and children after prior bag urine cultures yielded bacterial growth. In both studies, SPA yielded a significant number of sterile cultures, indicating that a proportion of the bag cultures were false positive. However, it is likely that many UTIs resolve spontaneously,7 which may explain part of this discrepancy. Furthermore, the study by Leong and Tan has no mention of perineal cleansing before bag placement and states specifically that the “child’s external genitalia is not swabbed,”24 which is not the standard of practice today.

The consequences of false-positive urine cultures are not benign because they may lead to unnecessary antibiotics, hospitalization, and diagnostic imaging.9 On the other hand, CATH specimens are painful, technically difficult, and a common source of parental anxiety.26,27 Parents have been shown to differ from physicians in placing greater emphasis on the pain and discomfort of tests and when given scenarios ranked CATH as substantially more undesirable than bag urine and closer in value to lumbar puncture and venipuncture.28 In addition, CATH may risk introducing bacteria into the urinary tract29 and has been shown to be a risk factor for septicemia in neonates.30 Although catheterized specimens appear to perform better than bag specimens, the magnitude of the difference is small, and the technique undoubtedly has limitations.

This study is limited in that all infants were enrolled by PROS practitioners who performed the initial clinical examination. Infants initially seen by emergency department physicians were not included in the study, and the performance of urine testing might differ in these infants. Furthermore, positive, negative, and ambiguous urine cultures were defined by the study investigators. Although these definitions were meant to reflect standards available in the literature, they may not represent standards followed by all practitioners. Finally, not all of the 3066 infants in this study underwent urine testing. Although the issue of which infants did or did not get tested was addressed in a prior publication,7 this selective testing may limit the generalizability of our findings. In other words, if urine testing was mandatory as opposed to discretionary, the predictors of catheterization and the UA test characteristics may have differed.

We believe that if bag urine cultures are performed appropriately and interpreted cautiously, errors can be minimized. The poorer specificity of bag specimens is of greater concern for practitioners who manage UTIs aggressively (ie, routine hospitalization and imaging). Ultimately, the choice of urine collection method should incorporate a number of factors, including patient age, parental preference, need for immediate diagnosis and/or antibiotic treatment, and plans for future imaging. The diagnosis of UTI should be made only after careful consideration of pretest probability, UA results, method of collection, and urine culture results.

Correspondence: Robert H. Pantell, MD, University of California, San Francisco, Campus Box 0503, LH 245, San Francisco, CA 94143-0503 (pantell@itsa.ucsf.edu).

Disclaimer: Dr Pantell had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Previous Presentation: This work was presented in part at the Pediatric Academic Societies Annual Meeting; May 4, 2004; San Francisco, Calif.

Back to top
Article Information

Accepted for Publication: March 28, 2005.

Funding/Support: This work was supported by a grant from the Agency for Healthcare Research and Quality (Rockville, Md) (R01 HS06485), which approved the original design, with additional support from the Health Resources and Services Administration Maternal and Child Health Bureau (Rockville) (MCJ-177022) and the AAP (Elk Grove Village), which developed and maintain the PROS network of practitioners. For this project, Dr Schroeder was supported through the University of California, San Francisco by a T32 National Institute of Child Health and Human Development Institutional Training Grant for Pediatricians (HD044331) and by a faculty development fellowship from the Health Resources and Services Administration Bureau of Health Professions (Rockville) (D5HP-00054).

Acknowledgment: We especially appreciate the help of the PROS practitioners. The pediatric practices or individual practitioners who enrolled subjects in this study are listed here by AAP Chapter: Alabama: Drs Heilpern and Reynolds, PC (Birmingham), Growing Up Pediatrics, PC (Birmingham), and University of Alabama (Birmingham); Alaska: Anchorage Neighborhood Health Center (Anchorage), Anchorage Pediatric Group (Anchorage), La Touche Pediatrics (Anchorage), and Eielson Clinic (Eielson); Arizona: Mesa Pediatrics Professional Associates (Mesa), Orange Grove Pediatrics (Tucson), Tanque Verde Pediatrics (Tucson), and Cigna HealthCare (Tucson); California 1: Palo Alto Medical Clinic (Palo Alto), University of California, San Francisco, Laurel Heights (San Francisco), Palo Alto Medical Foundation (Los Altos), Palo Alto Medical Clinic (Fremont), Shasta Community Health Center (Redding), Healthy Trails Pediatric Medical Group (Freedom), Anita Tolentino-Macaraeg, MD (Hollister), Eureka Pediatrics (Eureka), Drs Cantor, Giedt, Kamachi & Brennan (Salinas), and Marin Community Clinic (Greenbrae); California 2: Clinic Sierra Vista (Lamont), Rose City Pediatrics Medical Group (Pasadena), Touraj Shafai, MD (Riverside), and Facey Medical Group (Northridge); California 4: Edinger Medical Group, Inc (Fountain Valley) and Southern Orange County Pediatric Associates (Lake Forest); Colorado: Rocky Mountain Youth Medical Providers, PC (Thornton); Connecticut: St Francis Pediatric Primary Care Center (Hartford), Hemant Panchal, MD (Enfield), and Uwe Koepke, MD (Danbury); Delaware: Pediatric Associates (Newark); District of Columbia: George Washington University Health Plan (Washington); Florida: Sawgrass Pediatrics, PA (Coral Springs), Jonathan Rubin, MD, PA (Margate), MacKoul Pediatric (Cape Coral), SW Florida Pediatric Network (Fort Meyers), Atlantic Coast Pediatrics (Merritt Island), Orlando Health Care Group (Orlando), Sacred Heart Pediatric Care Center (Pensacola), Giangreco & Scarano Pediatrics (Bradenton), Emilio Del Valle, MD (Fort Myers), and Arnold Palmer Women & Children's Hospital (Orlando); Georgia: The Pediatric Center (Stone Mountain), Children's Hospital at Memorial (Savannah); Hawaii: Jeffrey Lim, MD (Honolulu), Melinda Ashton, MD (Honolulu), and University of Hawaii (Honolulu); Illinois: Southwest Pediatrics SC (Orland Park), SIU Physicians & Surgeons–Auburn (Auburn), LaGrange Pediatrics (Western Springs), Sidney Smith, MD (Carbondale), and Signature Medical Associates (Elgin); Indiana: Georgetown Pediatrics (Indianapolis), Pediatric Advocates (Peru), Southern Indiana Pediatrics (Bedford), Bloomington Pediatric Associates, PC (Bloomington), Lynn Ryan, MD (Lawrenceburg), Marshall County Pediatrics (Plymouth), Jeffersonville Pediatrics (Jeffersonville), Children's Health Care (Batesville), Northpoint Pediatrics-Fischers (Fischers), and Southern Indiana Pediatrics, LLC (Bloomington); Iowa: David Kelly, MD (Marshalltown), West Des Moines Family Physicians (West Des Moines); Kansas: Bethel Pediatrics (Newton), and Ashley Clinic (Chanute); Kentucky: Pediatric & Adolescent Medicine (Lexington); Maine: John Salvato, MD (Waterville) and Intermed Pediatrics (Portland); Maryland: Clinical Associates, PA (Towson), Drs Andorsky, Finkelstein and Cardin (Owings Mills), Children's Medical Group (Cumberland), Steven Caplan, MD (Baltimore), Shore Pediatrics (Easton), O'Donovan & Ahluwalia, MD, PA (Baltimore), Drs Wiczer, Korengold, Mayol (Bethesda), D'Albora & Osha, MD, PA (Rockville), Shady Side Medical Associates (Shady Side), The Children's Doctors (Westminster), Drs Coleman, Sachs & Thillairajah (Rockville), and Potomac Physicians (Severna Park); Massachusetts: Framingham Pediatric, PC (Framingham), Garden City Pediatrics (Beverly), Burlington Pediatrics (Burlington), Riverbend Medical Group (Chicopee), Holyoke Pediatric Associates (Holyoke), John Mulqueen, MD (Gardner), Pediatric Associates of Norwood (Norwood), Cape Cod Pediatrics (Forestdale), and Winthrop Community Health Center (Winthrop); Michigan: Botsford Pediatrics (Farmington), H. M. Hildebrandt, MD (Ypsilanti), Essexville Medical Clinic (Bay City), Downriver Pediatric Associates, PC (Lincoln Park), Child Health Associates (Ann Arbor), Pediatric & Family Care of Rochester Hills, PC (Rochester Hills), Drs Lee & Kim Associates (Warren), and Orchard Pediatrics (West Bloomfield); Minnesota: South Lake Clinic (Minnetonka); Missouri: Pediatric Associates of SW Missouri (Joplin), Children's Clinic (Springfield), and Doctor’s Clinic (Carruthersville); Montana: Stevensville Pediatrics (Stevensville); New Hampshire: Exeter Pediatric Associates (Exeter), Lahey-Hitchcock Clinic-Concord (Concord), Dartmouth-Hitchcock Clinic (Lebanon), Laconia Clinic (Laconia), Pediatric & Adolescent Medicine (Kingston), and Lahey-Hitchcock Clinic-Keene (Keene); New Jersey: Kids Care Pediatrics (Egg Harbor Township), Salem Road Pediatrics (Burlington), and Coventry Family Practice (Phillipsburg); New Mexico: Albuquerque Pediatric Associates, Ltd (Albuquerque), and University of New Mexico Hospital (Albuquerque); Nevada: Physician’s Center West (Fallon), and Job’s Peak Primary Care Specialists (Gardnerville); New York 1: Panorama Pediatric Group (Rochester), Elmwood Pediatric Group (Rochester), Albany Medical College Pediatric Group (Albany), Southern Tier Health Associates (Wayland), Parkway Pediatrics (Rochester), Lewis Pediatrics (Rochester), Gayle Buckley, MD (Ballston Lake), Genesee Health Service (Rochester), Pine Street Pediatric Associates, PC (Kingston), North Country Children's Clinic (Watertown), and Springville Pediatrics (Springville); New York 2: Women & Children’s Health Center (Long Island City), Gary Mirkin, MD (Great Neck), Southampton Pediatric Associates (Southampton), and Sonia Vinas, MD (Brooklyn); New York 3: Saint Vincent's Pediatric Associates (New York); North Carolina: Novant Health, EastoverPediatrics (Charlotte), Triangle Pediatric Center (Cary), and Peace Haven Family Health Center (Winston-Salem); North Dakota: Medical Arts Clinic (Minot), Altru Clinic (Grand Forks), and Dakota Clinic, Ltd - Jamestown (Jamestown); Ohio: Bryan Medical Group (Bryan), Cleveland Pediatric & Adolescent Clinic (Cleveland), South Dayton Pediatrics, Inc (Dayton), Oxford Pediatrics & Adolescents (Oxford), John DiTraglia (Portsmouth), Family Health Center (Idaho), Oberlin Clinic (Oberlin), Children's Hospital Physicians Associates (Twinsburg), North Central Ohio Family Care Center, Inc (Galion), and Drs Harris & Rhodes (Lancaster); Oklahoma: Pediatric & Adolescent Care, LLP (Tulsa), Medical Care Associates of Tulsa (Tulsa), and OU Pediatric Clinic (Tulsa); Oregon: Eugene Clinic (Eugene) and NBMC (Coos Bay); Pennsylvania: Pennridge Pediatric Associates (Sellersville), Praful Bhatt, MD (Lock Haven), Reading Pediatrics, Inc (Wyomissing), Children's Health Care (Allentown), Erdenheim Pediatrics (Flourtown), Yoon-Taek Chun, MD (East Stroudsburg), Pediatric Associates of Plymouth (Plymouth Meeting), Plum Pediatrics (Pittsburgh), Einstein Community Health Associates (Philadelphia), Cevallos and Moise Pediatric Associates, PC (Quakertown), Pediatric Group Services (Philadelphia), VNA KidsCare (Bethlehem), and Laurel Health Center (Blossburg); Puerto Rico: Edna Zayas, MD (San Juan) and Pediatric Ambulatory Clinic (San Juan); Rhode Island: Marvin Wasser, MD (Cranston); South Carolina: Anderson Pediatric Group (Anderson), Grand Strand Pediatrics & Adolescent Medicine (Surfside Beach), and Barnwell Pediatrics, PA (Barnwell); Tennessee: Johnson City Pediatrics, PC (Johnson City); Texas: The Pediatric Clinic (Greenville), Winnsboro Pediatrics (Winnsboro), Pediatrics (Sherman), Sarah Helfand, MD (Dallas), UT Health Center at Tyler Pediatrics (Tyler), Pediatric Clinic (Mineral Wells), White Rock Pediatrics, PA (Dallas), UNTHSC at Ft Worth Pediatric Clinic (Ft Worth), and Family Medical Center (Big Spring); Utah: Utah Valley Pediatrics, LC (American Fork), Mountain View Pediatrics (Sandy), Granger Medical Center (Salt Lake City), Willow Creek Pediatrics (Salt Lake City), John Weipert, MD (American Fork), and University of Utah Health Sciences Center (Salt Lake City); Vermont: University Pediatrics, UHC Campus (Burlington), Pediatric Medicine (South Burlington), Timber Lane Pediatrics (South Burlington), Hagan & Rinehart Pediatrics (South Burlington), Brattleboro Pediatrics (Brattleboro), University Pediatrics–Williston Office (Williston), Rebecca Collman, MD (Colchester), Mousetrap Pediatrics (Milton), Green Mountain Pediatrics, PC (Bennington), and St Johnsbury Pediatrics (St Johnsbury); Virginia: Pediatric Association of Richmond, Inc (Richmond), Alexandria Lakeridge Pediatrics (Alexandria), Drs Casey, Goldman, Lischwe, Garrett & Kim (Arlington), Fishing Bay Family Practice (Deltaville), Stafford Pediatrics, PC (Stafford), and Pediatric Clinic (Arlington); Washington: Valley Children's Clinic (Renton), Rockwood Clinic (Spokane), Yakima Valley Farm Workers Clinic (Toppenish), Paulouse Pediatrics (Pullman), and Columbia Health Center (Seattle); West Military: Wilford Hall Medical Center (Lackland Air Force Base); West Virginia: Grant Memorial Pediatrics (Petersburg), and Tess Alejo, MD (Martinsburg); Wisconsin: Beloit Clinic SC (Beloit), Lutheran Hospital (La Crosse), Waukesha Pediatric Associates (Waukesha), CHW Pediatric Clinic (Milwaukee), and Aurora Health Center, Waukesha (Waukesha); Wyoming: Jackson Pediatrics (Jackson) and Cheyenne Children’s Clinic (Cheyenne).

Bachur  RGHarper  MB Predictive model for serious bacterial infections among infants younger than 3 months of age.  Pediatrics 2001;108311- 316PubMedGoogle ScholarCrossref
Baker  MDBell  LMAvner  JR Outpatient management without antibiotics of fever in selected infants.  N Engl J Med 1993;3291437- 1441PubMedGoogle ScholarCrossref
Bonadio  WA Evaluation and management of serious bacterial infections in the febrile young infant.  Pediatr Infect Dis J 1990;9905- 912PubMedGoogle ScholarCrossref
Byington  CLRittichier  KKBassett  KE  et al.  Serious bacterial infections in febrile infants younger than 90 days of age: the importance of ampicillin-resistant pathogens.  Pediatrics 2003;111964- 968PubMedGoogle ScholarCrossref
Crain  EFGershel  JC Urinary tract infections in febrile infants younger than 8 weeks of age.  Pediatrics 1990;86363- 367PubMedGoogle Scholar
Jaskiewicz  JAMcCarthy  CARichardson  AC  et al. Febrile Infant Collaborative Study Group, Febrile infants at low risk for serious bacterial infection: an appraisal of the Rochester criteria and implications for management.  Pediatrics 1994;94390- 396PubMedGoogle Scholar
Newman  TBBernzweig  JATakayama  JIFinch  SAWasserman  RCPantell  RH Urine testing and urinary tract infections in febrile infants seen in office settings: the Pediatric Research in Office Settings' Febrile Infant Study.  Arch Pediatr Adolesc Med 2002;15644- 54PubMedGoogle ScholarCrossref
American Academy of Pediatrics, Committee on Quality Improvement, Subcommittee on Urinary Tract Infection, Practice parameter: the diagnosis, treatment, and evaluation of the initial urinary tract infection in febrile infants and young children.  Pediatrics 1999;103843- 852PubMedGoogle Scholar
Al-Orifi  FMcGillivray  DTange  SKramer  MS Urine culture from bag specimens in young children: are the risks too high?  J Pediatr 2000;137221- 226PubMedGoogle ScholarCrossref
Grisaru-Soen  GGoldman  RBarzilai  ALotan  DKeller  N False-positive urine cultures using bag collection.  Clin Pediatr (Phila) 2000;39499- 500PubMedGoogle ScholarCrossref
Li  PSMa  LCWong  SN Is bag urine culture useful in monitoring urinary tract infection in infants?  J Paediatr Child Health 2002;38377- 381PubMedGoogle ScholarCrossref
Liao  JCChurchill  BM Pediatric urine testing.  Pediatr Clin North Am 2001;481425- 1440, vii-viiiPubMedGoogle ScholarCrossref
Gorelick  MHShaw  KN Screening tests for urinary tract infection in children: a meta-analysis.  Pediatrics[serial online] 1999;104e54.Available at:http://pediatrics.aappublications.org/cgi/content/full/104/5/e54Accessed April 15, 2004PubMedGoogle Scholar
Pantell  RHNewman  TBBernzweig  J  et al.  Management and outcomes of care of fever in early infancy.  JAMA 2004;2911203- 1212PubMedGoogle ScholarCrossref
Stata Corp, Stata Statistical Software: Release 7.0.  College Station, Tex Stata Corp2001;229- 231
Herzog  LW Urinary tract infections and circumcision: a case-control study.  AJDC 1989;143348- 350PubMedGoogle ScholarCrossref
Wiswell  TEHachey  WE Urinary tract infections and the uncircumcised state: an update.  Clin Pediatr (Phila) 1993;32130- 134PubMedGoogle ScholarCrossref
Wiswell  TERoscelli  JD Corroborative evidence for the decreased incidence of urinary tract infections in circumcised male infants.  Pediatrics 1986;7896- 99PubMedGoogle Scholar
Wiswell  TESmith  FRBass  JW Decreased incidence of urinary tract infections in circumcised male infants.  Pediatrics 1985;75901- 903PubMedGoogle Scholar
Christakis  DARivara  FP Pediatricians' awareness of and attitudes about four clinical practice guidelines.  Pediatrics 1998;101825- 830PubMedGoogle ScholarCrossref
Young  PC The management of febrile infants by primary-care pediatricians in Utah: comparison with published practice guidelines.  Pediatrics 1995;95623- 627PubMedGoogle Scholar
Hoberman  AWald  ER Urinary tract infections in young febrile children.  Pediatr Infect Dis J 1997;1611- 17PubMedGoogle ScholarCrossref
Shannon  FSepp  ERose  G The diagnosis of bacteriuria by bladder puncture in infancy and childhood.  Aust Paediatr J 1969;597- 100Google Scholar
Leong  YYTan  KW Bladder aspiration for diagnosis of urinary tract infection in infants and young children.  J Singapore Paediatr Soc 1976;1843- 47PubMedGoogle Scholar
Sorensen  KLose  GNathan  E Urinary tract infections and diurnal incontinence in girls.  Eur J Pediatr 1988;148146- 147PubMedGoogle ScholarCrossref
Kleiber  CMcCarthy  AM Parent behavior and child distress during urethral catheterization.  J Soc Pediatr Nurs 1999;495- 104PubMedGoogle ScholarCrossref
Bauchner  HVinci  RBak  SPearson  CCorwin  MJ Parents and procedures: a randomized controlled trial.  Pediatrics 1996;98861- 867PubMedGoogle Scholar
Kramer  MSEtezadi-Amoli  JCiampi  A  et al.  Parents' versus physicians' values for clinical outcomes in young febrile children.  Pediatrics 1994;93697- 702PubMedGoogle Scholar
Lohr  JADowns  SMDudley  SDonowitz  LG Hospital-acquired urinary tract infections in the pediatric patient: a prospective study.  Pediatr Infect Dis J 1994;138- 12PubMedGoogle ScholarCrossref
Fok  TFLee  CHWong  EM  et al.  Risk factors for Enterobacter septicemia in a neonatal unit: case-control study.  Clin Infect Dis 1998;271204- 1209PubMedGoogle ScholarCrossref