Abstract
BACKGROUND. Previous studies documented considerable variability in the treatment of febrile infants, despite the existence of practice guidelines for this condition. None of those studies documented the extent to which this variability is accounted for by differences in clinical severity.
OBJECTIVE. To quantify the individual effects of the patient's clinical presentation, demographic, provider, and practice characteristics, and regional variables on practice variability in the evaluation and treatment of febrile infants.
METHODS. With data collected through the Pediatric Research in Office Settings network, we analyzed data on the treatment of 2712 febrile infants examined by 484 pediatricians located in 194 practices. We analyzed hospitalization, lumbar puncture, urinalysis and/or urine culture, blood work, and initial antibiotic administration. We obtained a summary score for evaluation and treatment intensity (ranging from no tests or treatments to comprehensive testing, hospitalization, and antibiotic therapy) by performing principal-components analysis with these 5 variables. This summary score was regressed with respect to patients' clinical presentation, demographic and practice/practitioner features, and geographic region. Provider fixed effects were also included in the model.
RESULTS. Although the overall model explained 46.5% of the variance, the clinical characteristics of the patient alone explained 29.7% of the overall variance. Practice site fixed effects explained nearly 15% of the overall variance. Provider and practitioner characteristics and geographic region had minimal explanatory power.
CONCLUSIONS. Our results show that measures of the patient's clinical presentation account for nearly one third of the variability that our model explains. This suggests that differences in clinical presentation and severity of illness underlie much of the observed practice variability among pediatricians evaluating and treating febrile infants. These findings demonstrate that the management of this common and potentially serious condition depends more on the clinical presentation of the patient than on the characteristics of the provider/practice and the residential region.
In 1993 and again in 2000, Baraff and colleagues1,2 published widely known evidence-based guidelines for the treatment of febrile infants. These guidelines recommended sepsis evaluation and hospitalization for infants <28 days of age who have a fever without a source. These guidelines also recommended outpatient-based treatment for 28- to 90-day-old infants who met “low-risk” criteria. Studies demonstrated that considerable variation in the evaluation and treatment of febrile infants continues to exist despite the presence of these guidelines.3,4 In a survey of pediatricians, family practitioners, and emergency medicine physicians, the majority indicated that they would hospitalize 3- and 7-week-old infants presenting with a fever without a source.5 For older infants, however, there was more variation in practice within and across the 3 provider types studied. A survey by Witter et al3 found similar results, as did a provider survey performed before the publication of the Baraff guidelines.1
In a study of adherence to the Baraff guidelines among pediatricians, family physicians, and emergency medicine providers, the authors found that 59% of pediatricians, 29% of family physicians, and 49% of emergency medicine providers described themselves as aware or somewhat aware of the guidelines.6 Despite these reportedly high levels of awareness, only 36.7% of pediatricians reported that they followed the guidelines 100% of the time. Across all specialties, approximately one half of respondents cited the guideline's perceived invasiveness as a reason for failing to comply with the recommendations. Among pediatricians, nearly 40% thought that the guidelines would not affect their clinical practice. Although there was considerable disagreement with the guidelines in some areas, there was a high degree of agreement with the guidelines regarding the treatment of very young infants.
These studies highlight some potential underlying causes of the failure of providers to follow published guidelines for the evaluation and treatment of febrile infants and the observed variation in the evaluation and treatment of this condition. Although lack of awareness of the guidelines does contribute to noncompliance, disagreement with the guideline's aggressiveness and necessity seems to play a larger role in noncompliance among pediatric providers. Zerr et al6 also found that 15.4% and 38.8% of pediatricians felt comfortable with their ability to diagnose serious bacterial illness among infants <4 weeks and 4 to 12 weeks of age, respectively. These findings suggest that some pediatric providers rely on their own clinical judgment more than on the published guidelines. Therefore, variability in the treatment of febrile infants may be intentional, reflecting differences in the clinical presentation of patients.
The observed variability in the diagnosis and treatment of febrile infants either may represent deviations from optimal care, as embodied in clinical guidelines, or may reflect appropriate modifications of care in response to the severity of the patient's clinical presentation. By better understanding the source of variability, we should be able to improve the evaluation and treatment of febrile infants. To accomplish this goal, we analyzed the care of febrile infants ≤3 months of age in a prospective cohort study of pediatric office-based practices. This study examined the relationship between observed variability in the diagnosis and treatment of febrile infants and differences in clinical presentation, demographic features, practice/practitioner variables, and geographic region. We hypothesized that differences in clinical assessments account for much of the explainable variability in the diagnosis and treatment of febrile infants and that providers' failure to comply with existing guidelines for this condition results, in part, from variability in patients' clinical presentations. The relationship between the clinical outcomes of the study subjects and compliance with clinical guidelines was presented in a previously published report.7
METHODS
Data
This study used data from the Pediatric Research in Office Settings (PROS), the practice-based research network of the American Academy of Pediatrics (AAP). Currently PROS includes >1900 physicians, nurse practitioners, physician assistants, and other health professionals who have an interest in conducting primary care research in their offices. For this study, >800 providers expressed interest in participating and 577 provided data on infants meeting the eligibility requirements. Although PROS members are not a random sample of AAP members, the participating providers in this sample were similar to AAP general pediatrician members who completed a 1995 Periodic Survey of Fellows,8 with respect to all demographic characteristics measured except for age, race, gender, and practice arrangement. Fewer study clinicians (7.3% vs 12%; P < .001) practiced in urban inner-city areas. Study providers tended to be slightly older than typical AAP pediatricians, were less likely to be female, were more likely to be white, and were less likely to be Asian American.
With a prospective, observational, cohort study design, we monitored the febrile episodes of patients who presented to study providers between February 28, 1995, and April 25, 1998. Providers enrolled eligible febrile infants sequentially in the study during this time period. A patient was eligible for the study if he or she was ≤3 months of age, had a temperature of ≥38°C either at home or in the provider's office, and had no other major comorbidities, such as significant congenital anomalies, extreme prematurity, or other conditions associated with organ-system failure. PROS participants recorded their observations of the infant and managed the patient's care in their customary manner. The clinician recorded a few critical observations during the patient encounter. Other office staff members recorded substantial demographic and laboratory data. A telephone call was required as the minimal follow-up contact, to ensure that the clinician knew the outcome of the febrile episode. To assess data quality, staff members at the AAP offices and at the University of California, San Francisco, reviewed data for clinical logic and additional missing or erroneous data, contacted PROS practice coordinators regarding problematic information, and edited the data appropriately. An outside group entered the data by using double entry and data logic checks.
At the end of the study, participants had collected data on 3131 infants, of whom 3066 met eligibility requirements. Of these, 2712 (88%) had complete data for all dependent and independent variables in our analyses. These infants were treated by 484 providers, located in 194 practices. The characteristics of the practices and providers were described in a previously published report.7
Dependent Variable
To characterize diagnostic and therapeutic management, we chose data on 5 dichotomous variables, ie, urinalysis or urine culture, blood culture, lumbar puncture, initial administration of antibiotics, and hospitalization. Of these 5 evaluation and treatment options, blood culture was used most frequently (74% of cases), followed by urinalysis and/or urine culture (57%), initial use of antibiotics (52%), hospitalization (34%), and lumbar puncture (33%). We then used principal-components analyses to combine the information on these 5 variables to generate an overall measure of intervention intensity. One can think of the first principal component as an intervention intensity score equal to the sum of the normalized scores (ie, adjusted for frequency) of the 5 evaluation and treatment variables, weighted with the standardized factor loadings. This allowed us to place the mean averages for these variables on the same scale and to give them weights according to the degree to which they explained the overall variability. The score summarizes the data for each of these 5 variables to create a single measure that maximizes the degree of the variance that can be explained by the evaluation and treatment variables. The evaluation and treatment intensity scores ranged from a low of −1.37 for cases in which providers used none of the 5 evaluation and treatment variables to a high of 1.45 for cases in which providers used all 5 variables. The first principal component, which served as our measure of evaluation and treatment intensity, explained 56% of the overall variance in the 5 variables.
Independent Variables
Our independent variables fell into 4 classes, ie, clinical presentation, demographic features, provider and practice characteristics, and regional variables. Clinical variables included provider rating of the patient's initial appearance (coded as mildly, moderately, or very ill), patient age (in days), and patient temperature, represented by the highest of the values reported by the patient's caregiver or recorded during the office visit (with addition of 0.5°C to axillary temperatures). Temperature was included in the model as both a linear term and a squared term, to account for the attenuation of its effects at higher levels. The model also included a variable that indicated the presence of any maternal risk factor for possible sepsis, as reported by the physicians, such as positive cultures for group B streptococci, chorioamnionitis, fever, or intrapartum antibiotic treatment. Demographic factors included gender, race, ethnicity, and insurance type. Provider and practice variables included provider age, gender, training (ie, physician, nurse practitioner, or physician assistant), and practice type (ie, solo or 2-person practice, group pediatric practice, or multispecialty practice). Regional variables included urban location and region of the country.
Analyses
We developed a statistical model to quantify the independent effects of clinical presentation, demographic features, provider and practice characteristics, and regional variables on practice variability in the diagnosis and treatment of febrile infants. In addition to these 4 covariate classes, we included practice fixed effects (ie, a dummy variable for each practice site) to control for the correlation across multiple cases from the same practice site. These site fixed effects also allowed us to determine whether treatment decisions were similar among providers within the same practice. Ordinary least-squares regression analysis was used to estimate the fixed-effects model.
RESULTS
Descriptive Statistics
Providers described nearly three fourths of the febrile infants in the study as mildly ill (Table 1). One fourth of infants were described as moderately ill, and <2% were described as very ill. Infant ages ranged from 0 to 93 days, with a mean of 48.9 days. Temperatures ranged from a low of 38°C, which was the minimal temperature required for inclusion in the study, to a high of 41°C. Approximately one tenth of the infants were at increased risk because of a maternal intrapartum risk factor.
Descriptive Statistics
White infants represented the majority of infants in the sample. The sample had disproportionately fewer black infants than the overall population proportion in the United States (8.3% vs 12.7%). The under-representation of black subjects in this sample may stem from the small number of urban, inner-city practices that participated in the study. Hispanic infants were slightly over-represented in the study, compared with their overall population proportion in the United States (14.5% vs 11.2%). Infants were fairly evenly distributed across health maintenance organization (HMO), private/commercial, and Medicaid insurance. A small number of infants' families had other forms of insurance or were uninsured.
Most of the providers in the sample were physicians and were male. The average age among providers was 45 years. The majority of providers worked in group pediatric practices. Very few practices were located in urban, inner-city areas. Of the geographic regions, practices in the north central region contributed the fewest patients to the study.
As mentioned previously, the Baraff guidelines call for routine hospitalization of all febrile infants <28 days of age. In our sample, providers hospitalized 61.3% of such infants. Providers described 86.7% of nonhospitalized febrile neonates as mildly ill but described only 56.1% of hospitalized neonates as mildly ill (P < .001). In our sample, providers hospitalized 26.6% of febrile infants 28 to 93 days of age. Providers described 48.2% of those hospitalized infants as mildly ill. In contrast, providers described 83.9% of nonhospitalized infants 28 to 93 days of age as mildly ill (P < .001).
Multiple Regression Analyses
Overall, our model explained 46% of the observed variance in evaluation and treatment intensity. Measures of clinical presentation alone explained 29% of the observed variance in evaluation and treatment intensity (Table 2). As a group, practice site fixed effects contributed significantly to the model and explained ∼15% of the overall variance in evaluation and treatment intensity. Practice location measures and provider and practice characteristics each explained ∼1% of the variance.
Partial R2 Values
As stated previously, the clinical presentation measures in the model contributed the most to the overall explanatory power of the model. Of the clinical presentation variables, the patient's appearance had the largest effect on the evaluation and treatment intensity score (Table 3). Compared with children whom providers characterized as appearing minimally ill, those who appeared very ill had expected evaluation and treatment intensity scores 0.92 points higher. Children who appeared moderately ill had significantly higher treatment scores; however, the magnitude of the effect was smaller than that among very ill children (0.69-point increase, compared with 0.92-point increase).
Fixed-Effects Results for Estimations of Evaluation and Treatment Intensity
Older infants had lower expected treatment scores than their younger peers. For example, a 4-week-old infant had an expected treatment score 0.29 points lower than a 1-week-old infant. Similarly, infants with higher temperatures had higher evaluation and treatment intensity scores. An infant with a temperature of 39°C had a ∼0.69-point higher expected evaluation and treatment intensity score, compared with an infant with a temperature of 38°C. The results of the analysis suggested that the relationship between temperature and evaluation and treatment intensity was curvilinear. Therefore, evaluation and treatment intensities may differ considerably between infants with temperatures of 38°C and 38.5°C but may be relatively similar for infants with temperatures of 39.5°C and 40°C. Finally, infants whose birth history included any maternal risk factor had an intensity score 0.12 points higher than those whose mothers had no risk factors.
Few demographic features were related significantly to evaluation and treatment intensity. Neither black nor Asian American infants had significantly higher evaluation and treatment intensity scores, compared with their white counterparts. Infants of “other races” had significantly lower evaluation and treatment intensity scores, compared with white infants. Hispanic infants had expected evaluation and treatment intensity scores 0.13 points higher than non-Hispanic infants. Of the insurance variables, only Medicaid was related significantly to higher evaluation and treatment intensity scores. Medicaid-covered children had evaluation and treatment intensity scores ∼0.15 points higher than those with private insurance. The treatment scores for infants covered by HMOs did not differ significantly from scores for infants covered by private insurance.
Among the provider and practice characteristics analyzed, only children treated in solo or 2-provider practices had significantly different evaluation and treatment intensity scores. Children treated in these settings had expected scores 0.91 points higher than those treated in other practice settings (eg, HMOs, medical schools, community health centers, and hospitals). Physician age and gender had no effect on evaluation and treatment intensity. Moreover, scores for infants treated by nurse practitioners and physician assistants did not differ significantly from scores for infants treated by physicians.
Children treated in inner-city practices had significantly higher expected treatment scores than those treated in other nonurban practices, even after correction for type of insurance. Geographic region had no significant association with evaluation and treatment intensity.
DISCUSSION
Our findings demonstrate that differences in the clinical presentation of the patient account for a significant proportion of the variability in the evaluation and treatment of febrile infants. The relationship between clinical assessments, evaluations, and clinical outcomes in this study has been described previously.7 Although these results may not surprise practicing clinicians, they carry important implications for efforts to assess and to improve quality. In the past decade, various quality initiatives have attempted to improve patient care through the development and dissemination of evidence-based practice guidelines. The primary goal of these guidelines has been to close the gap between evidence-based care and actual clinical practice. From these efforts, the Agency for Healthcare Research and Quality has cataloged thousands of guidelines and made them accessible to clinicians over the Internet.9 Despite these prodigious efforts to produce evidence-based recommendations, considerable variability in clinical practice remains and compliance with guidelines remains low.
As in other studies, we found substantial variability in the implementation of existing guideline recommendations in clinical practice. In a previous report from this study, we showed that these deviations from recommended practice resulted in similar outcomes, in terms of identification and hospitalization of patients with serious bacterial infections, compared with those expected if the guidelines had been followed.10 Our findings clearly demonstrate that differences in clinical presentation account for much of the explainable variation in the treatment of febrile infants. These findings are consistent with the work by McCarthy et al10,11 and others12 that documented the validity of a clinical observation score in combination with laboratory findings in predicting serious bacterial illness among infants >3 months of age. Although the pediatricians in this study did not use a scoring system, it was clear that clinical appearance and clinical evaluation results played important roles in explaining the variations in clinical practice. Of some concern was the finding that 13% of infants <30 days of age who were assessed as being moderately or severely ill were not hospitalized, despite research showing that this is a particularly high-risk group for serious bacterial illness.7
In contrast to clinical characteristics, nonclinical characteristics of the patient, provider and practice characteristics, and regional factors explained little of the variation in evaluation and treatment intensity. Among these other covariates, we noted some potentially important associations. Patients examined in urban, inner-city practices and those covered by Medicaid had higher evaluation and treatment intensity scores than did those from nonurban areas and those with other forms of insurance coverage. These findings may reflect the appropriate application of a more aggressive treatment approach with low-income populations. Such an approach has been advocated for patients whose families may experience barriers to obtaining appropriate follow-up care.13,14
Practice site fixed effects explained ∼15% of the explainable variance in our model. The significance of these fixed effects suggests that providers within the same practice tend to evaluate and to treat febrile infants in similar ways. Our study cannot provide insight into the reasons for this tendency, but we speculate that certain underlying phenomena may contribute to the observed tendency for practice partners to practice similarly. Providers may tend to join partners with similar attitudes and levels of risk aversion. Therefore, the observed similarity across providers in the same practices may merely reflect providers' attraction to similar providers when joining a practice. Similarities in practice patterns may reflect sociodemographic considerations more local to the practice or clinic than those accounted for in the basic racial, ethnic, and geographic variables and may also reflect informal or formal agreements among members of a practice to follow practice-specific guidelines.
Whatever the underlying reason for the similarities in practice patterns among partners, the finding has both clinical and research relevance. From a clinical perspective, efforts to improve quality of care can capitalize on these similarities and use opinion leaders and academic detailing in changing clinician behavior. From a research perspective, these results demonstrate the importance of incorporating the “clustering” effects of providers and sites when using multiple patients from the same providers and sites as the unit of analysis in practice-based research. Failure to do so can cause researchers to draw inappropriate inferences from their data.
Our study did not document variability in practice according to region of the country. These results run contrary to prior research that documented significant regional variability in hospitalization rates and rates of surgical procedures.15 Those studies, however, focused on procedures that frequently required referral from a primary care physician to a specialist. Regional differences in enthusiasm for and belief in a given procedure and/or availability of specialists may explain the observed variability in these rates.
Our study is unique for several reasons. First, our national data source involved a large diverse sample of pediatric providers. Second, our sample focused on office-based providers rather than those based in academic medical centers, enhancing the generalizability of our findings. Finally, we examined the actual practice decisions of providers through the use of detailed data collection for patients. We think that this method provides a more valid assessment of provider behavior than those that rely on provider responses to case scenarios. With these data, we think we can provide a thorough assessment of which factors underlie variability in the treatment of febrile infants.
The greatest limitations of our study are the lack of a uniform definition of the initial clinical appearance variable and the lack of a standardized method for assessing this variable. Providers were not given any definition of the categories of mildly, moderately, or very ill. Therefore, more cautious providers with greater propensities to treat aggressively may characterize patients differently than less risk-averse providers. For this reason, our measure of initial appearance may be less than ideal. The low frequency of patients classified as very ill is reassuring and suggests that providers consider this presentation to be relatively rare. An additional limitation may be the inclusion of infants with a reported temperature of >38°C. It is possible that these infants did not have a true fever, which might have biased the sample toward less sick infants. The use of 0.5°C as a conversion factor for changing axillary temperatures to rectal temperatures might also have contributed to potential inaccuracies in the assessment of temperature.
It is important to note that our model could not assess the impact of psychosocial and environmental variables on evaluation and treatment intensity. For example, we cannot estimate the influence of parental concern or pressure on provider behavior.14 The associations of increased evaluation and treatment intensity with Medicaid insurance type and urbanicity hint at the possible importance of psychosocial and environmental variables or may suggest that these infants have a greater burden of serious illness. The impact of these variables on clinician decision-making and the costs and outcomes of care is an important area for additional research. It is also likely that there are additional clinician and infant variables that contribute to management decisions that were not assessed in this study.
Finally, the PROS network consists of a sample of volunteer providers. Few urban, inner-city practices participated in our study. The use of this sample limits the generalizability of our study; however, we think that our findings are highly relevant despite this limitation.
CONCLUSIONS
The results of this study suggest that differences in clinical presentation and severity of illness underlie much of the explainable practice variability in the evaluation and treatment of febrile infants. Practice location and provider and practice characteristics account for only a small percentage of this variability. Our findings imply that observed variability in clinical practice in the evaluation and treatment of this common and potentially serious condition may reflect providers' tendencies to tailor their treatment of febrile infants to the clinical presentation of the patient. These findings may offer to consumers of health care the reassurance that variability in treatment patterns reflects, in large part, the clinical presentation of the patient, rather than nonclinical characteristics of the patient, characteristics of the provider and his or her practice, or regional differences.
Acknowledgments
This study was supported by a grant from the Agency for Healthcare Research and Quality (grant R01 HS06485), with additional support from the Health Resources and Services Administration Maternal and Child Health Bureau (grant MCJ-177022).
We acknowledge Richard A. Olshen and Alfred Lin of the Department of Biostatistics at Stanford University for assistance with this research and Nanci Echeverri for editorial assistance. We especially appreciate the help of the PROS practitioners, and their pediatric practices, who made this study possible.
The pediatric practices or individual practitioners who enrolled subjects in this study are as follows (listed according to AAP chapter): Alabama: Drs Heilpern and Reynolds (Birmingham), Growing Up Pediatrics (Birmingham), University of Alabama (Birmingham); Alaska: Anchorage Neighborhood Health Center (Anchorage), Anchorage Pediatric Group (Anchorage), La Touche Pediatrics (Anchorage), Eielson Clinic (Eielson); Arizona: Mesa Pediatrics Professional Associates (Mesa), Orange Grove Pediatrics (Tucson), Tanque Verde Pediatrics (Tucson), 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, and Brennan (Salinas), Marin Community Clinic (Greenbrae); California-2: Clinic Sierra Vista (Lamont), Rose City Pediatrics Medical Group (Pasadena), Touraj Shafai, MD (Riverside), Facey Medical Group (Northridge); California-4: Edinger Medical Group (Fountain Valley), Southern Orange County Pediatric Associates (Lake Forest); Colorado: Rocky Mountain Youth Medical Providers (Thornton); Connecticut: St Francis Pediatric Primary Care Center (Hartford), Hemant Panchal, MD (Enfield), Uwe Koepke, MD (Danbury); Delaware: Pediatric Associates (Newark); District of Columbia: George Washington University Health Plan (Washington); Florida: Sawgrass Pediatrics (Coral Springs), Jonathan Rubin, MD (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 and Scarano Pediatrics (Bradenton), Emilio Del Valle, MD (Fort Myers), Arnold Palmer Women and 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), University of Hawaii (Honolulu); Illinois: Southwest Pediatrics SC (Orland Park), Southern Illinois University Physicians and Surgeons-Auburn (Auburn), LaGrange Pediatrics (Western Springs), Sidney Smith, MD (Carbondale), Signature Medical Associates (Elgin); Indiana: Georgetown Pediatrics (Indianapolis), Pediatric Advocates (Peru), Southern Indiana Pediatrics (Bedford), Bloomington Pediatric Associates (Bloomington), Lynn Ryan, MD (Lawrenceburg), Marshall County Pediatrics (Plymouth), Jeffersonville Pediatrics (Jeffersonville), Children's Health Care (Batesville), Northpoint Pediatrics-Fischers (Fischers), Southern Indiana Pediatrics (Bloomington); Iowa: David Kelly, MD (Marshalltown), West Des Moines Family Physicians (West Des Moines); Kansas: Bethel Pediatrics (Newton), Ashley Clinic (Chanute); Kentucky: Pediatric and Adolescent Medicine (Lexington); Maine: John Salvato, MD (Waterville), Intermed Pediatrics (Portland); Maryland: Clinical Associates (Towson), Drs Andorsky, Finkelstein, and Cardin (Owings Mills), Children's Medical Group (Cumberland), Steven Caplan, MD (Baltimore), Shore Pediatrics (Easton), O'Donovan and Ahluwalia, MD (Baltimore), Drs Wiczer, Korengold, and Mayol (Bethesda), D'Albora and Osha, MD (Rockville), Shady Side Medical Associates (Shady Side), The Children's Doctors (Westminster), Drs Coleman, Sachs, and Thillairajah (Rockville), Potomac Physicians (Severna Park); Massachusetts: Framingham Pediatric (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), Winthrop Community Health Center (Winthrop); Michigan: Botsford Pediatrics (Farmington), H. M. Hildebrandt, MD (Ypsilanti), Essexville Medical Clinic (Bay City), Downriver Pediatric Associates (Lincoln Park), Child Health Associates (Ann Arbor), Pediatric and Family Care of Rochester Hills (Rochester Hills), Drs Lee and Kim Associates (Warren), Orchard Pediatrics (West Bloomfield); Minnesota: South Lake Clinic (Minnetonka); Missouri: Pediatric Associates of SW Missouri (Joplin), Children's Clinic (Springfield), 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 and Adolescent Medicine (Kingston), Lahey-Hitchcock Clinic-Keene (Keene); New Jersey: Kids Care Pediatrics (Egg Harbor Township), Salem Road Pediatrics (Burlington), Coventry Family Practice (Phillipsburg); New Mexico: Albuquerque Pediatric Associates (Albuquerque), University of New Mexico Hospital (Albuquerque); Nevada: Physician's Center West (Fallon), 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 (Kingston), North Country Children's Clinic (Watertown), Springville Pediatrics (Springville); New York-2: Women and Children's Health Center (Long Island City), Gary Mirkin, MD (Great Neck), Southampton Pediatric Associates (Southampton), Sonia Vinas, MD (Brooklyn); New York-3: Saint Vincent's Pediatric Associates (New York); North Carolina: Novant Health, Eastover Pediatrics (Charlotte), Triangle Pediatric Center (Cary), Peace Haven Family Health Center (Winston-Salem); North Dakota: Medical Arts Clinic (Minot), Altru Clinic (Grand Forks), Dakota Clinic-Jamestown (Jamestown); Ohio: Bryan Medical Group (Bryan), South Dayton Pediatrics (Dayton), Oxford Pediatrics and Adolescents (Oxford), John DiTraglia (Portsmouth), Family Health Center (Idaho), Oberlin Clinic (Oberlin), Children's Hospital Physicians Associates (Twinsburg), North Central Ohio Family Care Center (Galion), Drs Harris and Rhodes (Lancaster); Oklahoma: Pediatric and Adolescent Care (Tulsa), Medical Care Associates of Tulsa (Tulsa), Oklahoma University Pediatric Clinic (Tulsa); Oregon: Eugene Clinic (Eugene), North Bend Medical Center (Coos Bay); Pennsylvania: Pennridge Pediatric Associates (Sellersville), Praful Bhatt, MD (Lock Haven), Reading Pediatrics (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 (Quakertown), Pediatric Group Services (Philadelphia), VNA KidsCare (Bethlehem), Laurel Health Center (Blossburg); Puerto Rico: Edna Zayas, MD (San Juan), Pediatric Ambulatory Clinic (San Juan); Rhode Island: Marvin Wasser, MD (Cranston); South Carolina: Anderson Pediatric Group (Anderson), Grand Strand Pediatrics and Adolescent Medicine (Surfside Beach), Barnwell Pediatrics (Barnwell); Tennessee: Johnson City Pediatrics (Johnson City); Texas: The Pediatric Clinic (Greenville), Winnsboro Pediatrics (Winnsboro), Pediatrics (Sherman), Sarah Helfand, MD (Dallas), Cleveland Pediatric and Adolescent Clinic (Cleveland), University of Texas Health Center at Tyler Pediatrics (Tyler), Pediatric Clinic (Mineral Wells), White Rock Pediatrics (Dallas), University of North Texas Health Science Center at Fort Worth Pediatric Clinic (Fort Worth), Family Medical Center (Big Spring); Utah: Utah Valley Pediatrics (American Fork), Mountain View Pediatrics (Sandy), Granger Medical Center (Salt Lake City), Willow Creek Pediatrics (Salt Lake City), John Weipert, MD (American Fork), University of Utah Health Sciences Center (Salt Lake City); Vermont: University Pediatrics, University HealthSystem Campus (Burlington), Pediatric Medicine (South Burlington), Timber Lane Pediatrics (South Burlington), Hagan and Rinehart Pediatrics (South Burlington), Brattleboro Pediatrics (Brattleboro), University Pediatrics-Williston Office (Williston), Rebecca Collman, MD (Colchester), Mousetrap Pediatrics (Milton), Green Mountain Pediatrics (Bennington), St Johnsbury Pediatrics (St Johnsbury); Virginia: Pediatric Association of Richmond (Richmond), Alexandria Lakeridge Pediatrics (Alexandria), Drs Casey, Goldman, Lischwe, Garrett, and Kim (Arlington), Fishing Bay Family Practice (Deltaville), Stafford Pediatrics (Stafford), Pediatric Clinic (Arlington); Washington: Valley Children's Clinic (Renton), Rockwood Clinic (Spokane), Yakima Valley Farm Workers Clinic (Toppenish), Paulouse Pediatrics (Pullman), Columbia Health Center (Seattle); West Military: Wilford Hall Medical Center (Lackland Air Force Base); West Virginia: Grant Memorial Pediatrics (Petersburg), Tess Alejo, MD (Martinsburg); Wisconsin: Beloit Clinic SC (Beloit), Lutheran Hospital (La Crosse), Waukesha Pediatric Associates (Waukesha), Children's Hospital of Wisconsin Pediatric Clinic (Milwaukee), Aurora Health Center-Waukesha (Waukesha); Wyoming: Jackson Pediatrics (Jackson), Cheyenne Children's Clinic (Cheyenne).
Footnotes
- Accepted August 3, 2005.
- Address correspondence to David A. Bergman, MD, Division of General Pediatrics, 750 Welch Rd, Suite 325, Palo Alto, CA 94304. E-mail: david.bergman{at}stanford.edu
The authors have indicated they have no financial relationships relevant to this article to disclose.
REFERENCES
- Copyright © 2006 by the American Academy of Pediatrics
In this issue
Jump to section
Related Articles
Cited By...
- Accuracy of Clinician Suspicion of Lyme Disease in the Emergency Department
- The Yale Observation Scale Score and the Risk of Serious Bacterial Infections in Febrile Infants
- Febrile Infants: Aligning Science, Guidelines, and Cost Reduction With Quality of Individualized Care
- Practice Variations in the Treatment of Febrile Infants Among Pediatric Emergency Physicians
- Office-Based Treatment and Outcomes for Febrile Infants With Clinically Diagnosed Bronchiolitis
- Effect of prolonged and exclusive breast feeding on risk of allergy and asthma: cluster randomised trial
- Study looks at factors that influence treatment of febrile infants