OBJECTIVE. The goal was to measure US emergency department performance in the pediatric care of asthma, bronchiolitis, and croup, by using systematically developed quality indicators.
METHODS. Data on visits to emergency departments by children 1 to 19 years of age with moderate/severe asthma, 3 months to 2 years of age with bronchiolitis, and 3 months to 3 years of age with croup from the 2005 National Hospital Ambulatory Medical Care Survey, with a nationally representative sample of US patients, were analyzed. We used national rates of use of corticosteroids, antibiotics, and radiographs as our main outcome measures.
RESULTS. Physicians prescribed corticosteroids in 69% of the estimated 405 000 annual visits for moderate/severe asthma and in 31% of the estimated 317 000 annual croup visits. Children with bronchiolitis received antibiotics in 53% of the estimated 228 000 annual visits. Physicians obtained radiographs in 72% of bronchiolitis visits and 32% of croup visits.
CONCLUSIONS. Physicians treating children with asthma, bronchiolitis, and croup in US emergency departments are underusing known effective treatments and overusing ineffective or unproven therapies and diagnostic tests.
Respiratory illnesses are among the top 10 diagnoses for emergency department (ED) visits for children <17 years of age in the United States.1 Three of the most common pediatric respiratory illnesses are asthma, bronchiolitis, and croup. Children had >754 000 ED visits related to asthma in 2005, a rate of 103 visits per 10 000 children.2 Almost all children are infected with respiratory syncytial virus in the first 2 years of life, and up to 40% of them develop bronchiolitis,3 which accounts for ∼3% of ED visits in this age group.4 There are an estimated 129 000 ED visits related to croup each year.5
An extensive body of research and a number of systematic reviews provide strong evidence for the optimal treatment of asthma, bronchiolitis, and croup.6–9 The National Heart, Lung, and Blood Institute guidelines for the diagnosis and management of asthma have been available since 1991 and were updated in 1997 and 2007.10–12 Guidelines focused exclusively on pediatric asthma and extrapolated from the National Heart, Lung, and Blood Institute expert panel report have been available since 1999.13,14 Evidence-based clinical practice guidelines for bronchiolitis were published by Cincinnati Children's Hospital Medical Center in 1996, with revision in 2005 and review in 2006,15 and by the American Academy of Pediatrics in 2006.16 A systematic review recommending the use of glucocorticoids for children with croup was first published in 1997 and was updated in 2004.17
Previous studies measuring the Institute of Medicine quality domain of effectiveness of US ED performance in following evidence-based practices for children with asthma were based on physician self-report surveys,18–20 were based on local, state, or regional data,21–25 or were conducted before the availability of systematically determined pediatric quality indicators.4 Recently, however, Guttmann et al26 systematically developed clinical performance indicators for the ED care of children, by using a structured panel process27 and an extensive scientific literature review and incorporating recommendations from the US Emergency Medical Services for Children Managed Care Task Force.28 Their study defined several quality indicators for the ED care of asthma, bronchiolitis, and croup.26
The feasibility of using rigorously defined performance indicators such as those defined by Guttman et al26 depends on the availability of quality data. The National Hospital Ambulatory Medical Care Survey (NHAMCS) is a reliable source of data that captures detailed information about ED visits in a nationally representative sample of US hospitals. NHAMCS has been used for research regarding the ED care of children with bronchiolitis4 and adolescents with sexually transmitted diseases29 and for documentation of parenteral analgesic therapy underuse and inappropriate antibiotic-prescribing patterns for children and adolescents in EDs.30,31 The objective of this study was to assess US ED performance for the pediatric care of asthma, bronchiolitis, and croup, by using systematically developed quality indicators.
A retrospective cross-sectional study of ED encounters compiled from the NHAMCS database for 2005 was performed. The NHAMCS is a national, probability-sample survey of nonfederal, general, short-stay hospitals conducted by the Centers for Disease Control and Prevention, National Center for Health Statistics, Division of Health Care Statistics. We limited the study to 2005 data because that is the only year for which data on drugs given in the ED and at ED discharge are available.
The target of the NHAMCS is visits made to US EDs and outpatient departments of nonfederal, general (medical or surgical) or children's general, short-stay hospitals (hospitals with an average stay of <30 days). These criteria exclude stand-alone urgent care facilities. Multistage probability sampling is used to collect information on visits to EDs, as described elsewhere.32,33 The survey instrument is a patient record form collecting standard data; hospital staff members are trained to complete the form by specially qualified interviewers.33 A sample of 458 hospitals was selected for the 2005 NHAMCS, 386 of which had eligible EDs. Of these eligible EDs, 352 participated in the study, for an unweighted ED response rate of 91.2%. These EDs provided 33605 patient record forms. Visit weights accounting for selection probability, adjustment for nonresponse, and other adjustments allow extrapolation to national estimates for all aspects of the surveys.34,35 NHAMCS data represent a proportion of visits, not a proportion of patients, because these surveys do not track individual patients.
Demographic data such as age, gender, race, and insurance status are included. Hospital information includes geographic region and whether the hospital is in a rural area. Clinical characteristics include up to 3 diagnoses (coded by using International Classification of Diseases, Ninth Revision, Clinical Modification [ICD-9-CM] codes), up to 3 reasons for the visit, up to 8 medications, and the corresponding National Drug Code class numbers for all medications.36
Data quality for the 2005 NHAMCS data sets was ensured with a number of quality checks. A field representative performed an on-site record review and investigated missing or inconsistent data. Code ranges and inconsistencies were screened during the process of data entry. All records underwent a 10% independent verification procedure. The average keying error rates were 0.3% for nonmedical items and ranged between 0.3% and 4.2% for items requiring medical coding.34 The protocol for this study was reviewed and approved by the Children's Mercy Hospitals and Clinics pediatric institutional review board.
Three study populations were defined, to parallel the groups studied by Guttman et al.26 The first population was children 1 to 19 years of age with a diagnosis of asthma, the second was children 3 months to 2 years of age with a diagnosis of bronchiolitis, and the third was children 3 months to 3 years of age with a diagnosis of croup. For all of these conditions, the diagnosis needed to appear as a primary or secondary diagnosis. Table 1 lists the ICD-9-CM codes for these diagnoses.
To exclude patient visits related to mild exacerbations of asthma, we added 2 more conditions to the asthma visits. First, patients needed to receive nebulized albuterol treatment during their ED visit. The NHAMCS does not collect data on the route of administration; therefore, we used “nebulizer therapy” to select albuterol administration. Second, cases must have been triaged as immediate, emergency, or urgent (that is, needing to be seen within 60 minutes after arrival).
We also examined 2 subpopulations. For asthma visits, we examined a subpopulation of the visits that resulted in discharge (rather than admission to the hospital). For bronchiolitis visits, we examined a subpopulation of visits for which otitis media (ICD-9-CM code of 381 or 382) was not included as a primary or secondary diagnosis.
All medications were identified by using the National Drug Code Directory.36 We used the prefix 1032 for corticosteroids, the prefix 1940 for antiasthmatic agents/bronchodilators, and the prefixes 0346, 0347, 0348, and 0353 for antibiotics. Radiographs were identified by the variable x-ray.
We selected performance measures that are available in the NHAMCS database, as shown in Table 2. Guttman et al26 used International Classification of Diseases, 10th Revision (ICD-10) classifications. We obtained their list of ICD-10 classifications, and one of the researchers (Dr Sharma) manually translated the ICD-10 classifications used by those authors to ICD-9-CM classifications. The translation was then reviewed by a second author (Dr Knapp).
Race/ethnicity variables in the NHAMCS were combined into single variables with the following values: white not of Hispanic origin, Hispanic, black not of Hispanic origin, Asian not of Hispanic origin, and other/unknown. The pay type variable was combined into 4 categories, that is, commercial insurance (private insurance), government (Medicare or Medicaid), self-pay, and other/unknown (worker's compensation, blank, no charge, other, or unknown).
SEs were calculated by using a weighted approach recommended by the National Center for Health Statistics.37 All estimates had relative SE values (ie, the SE divided by the estimate, expressed as a percentage of the estimate) of <30% and were based on ≥30 cases in the sample data, as recommended by the National Center for Health Statistics.37 Because of this restriction, we were unable to compare the performance of general (medical or surgical) hospitals and children's hospitals or to perform other subgroup analyses. Ninety-five percent confidence intervals were computed for all proportions by using the Svy commands in Stata; all analyses were performed with Stata for Windows 9 (Stata, College Station, TX).
Cases of asthma, bronchiolitis, and croup meeting the inclusion criteria accounted for 950 000 (3%) of the 31 million pediatric ED visits for children <19 years of age in 2005. Table 3 provides demographic data. For all 3 conditions, the data were also analyzed by using only a primary diagnosis of the condition as the diagnosis of interest. This did not change any of the results.
Table 4 shows the results of ED performance in caring for children with asthma, bronchiolitis, and croup. Sixty-nine percent of asthma visits involved the use of steroids; 40% of patients received steroid treatment in the ED and a prescription for steroid use at home. Antibiotics were administered at 29% of the visits, either in the ED or as a prescription at discharge. Radiographs were obtained 48% of the time.
In 53% of bronchiolitis visits, patients were treated with antibiotics in the ED or were sent home with an antibiotic prescription. In addition, 72% of all bronchiolitis visits documented radiographs. Thirty-one percent of croup visits recorded the use of steroids. In addition, 32% of patients had radiographs obtained during their ED visits.
This is the first study to use rigorously determined quality indicators to examine US ED performance in caring for children with asthma, bronchiolitis, and croup. We chose to study these conditions because these are common pediatric conditions for which substantial evidence defining effective care has been published. We found significant gaps between recommended therapy and the care actually delivered in US EDs. It seems that known effective treatments are being underused and ineffective or unproven therapies are being overused in the ED management of these common respiratory illnesses. Deficits in the quality of US ED care of children with asthma, bronchiolitis, and croup seem to be of similar magnitude as those found in studies of the ambulatory care of children38 and adults.39
A 1995 survey of US EDs regarding the care of children with asthma found that only 45% reported using orally administered steroids if there was a poor response to initial treatment.18 A prospective observational study of Australian ED care of children with moderate exacerbations of asthma documented steroid use during the ED visits at >80%.40 The upper confidence limit for our sample indicates that US ED usage of steroids to treat acute exacerbations of asthma might be similar to that in Australia. If the lower confidence limit of 57% is more accurate, however, then this suggests that the quality gap is much wider. Inadequate therapy for asthma increases the risk for additional ED visits, missed school days, missed work days for parents, and avoidable hospitalizations.11
On the basis of our results, antibiotics are overused in the treatment of asthma and bronchiolitis in US EDs. In the Australian study, <10% of the children with asthma received a discharge prescription for antibiotics.40 In US EDs, the rate was 29%. Mansbach et al4 found a 37% rate of ED use of antibiotics for children <2 years of age with bronchiolitis. This was lower than our overall rate of 53% but was consistent with our rate when a concomitant diagnosis of otitis media was excluded. Overuse of antibiotics leads to unnecessary health care costs and increases the potential for the development of strains of resistant organisms.
A single dose of dexamethasone administered orally or parenterally to children with croup improves symptoms and reduces return visits and hospitalizations.17 We found that only 1 in 3 croup visits had a record of oral or parenteral administration of a steroid. Rates in Canada vary from 22% to 95%, depending on hospital location and volume.23
There was evidence of overuse of radiographs for all 3 conditions. Overuse of chest radiographs for pediatric asthma patients in US EDs was observed previously, usually in association with the presence of fever.18,41 The US rate was almost double the Canadian and Australian rates of 20% and 25%, respectively.26,40 At 72%, there was significant overuse of radiographs for bronchiolitis. The Canadian rate is 42.6%.26 Almost one third of the croup visits had radiographs. The rate of chest or lateral neck radiograph use for croup in Canadian children is 9%.26 Although the Canadian rates for radiographs represent per-patient rates and not per-visit rates, the striking disparity between the US and Canadian rates is unlikely to be accounted for by this factor alone. Overuse of radiographs for children increases the length of the ED stay, exposes the child to unnecessary radiation, and increases the costs of health care. It is possible to reduce the rate of radiograph overuse. ED provider education resulted in a safe 17% reduction in radiograph use in asthma,42 and experience with the implementation of evidence-based guidelines for bronchiolitis reduced radiograph use by 20%.43
The numbers of visits attributable to asthma, bronchiolitis, and croup were lower than expected. This might be explained by our inclusion criteria for all 3 conditions, which were designed to match the age selection criteria used by Guttmann et al.26
This was a secondary data analysis, and there are limitations associated with its use. Because this study examined a proportion of visits, not a proportion of patients, it cannot be determined whether the results were potentially inflated by data for individual patients with severe disease who had multiple visits.
We acknowledge that we cannot confirm the accuracy of the diagnosis of any of the conditions. We used ICD-9-CM codes to obtain our data; however, providers were not instructed to follow any defined diagnostic criteria before making a diagnosis.
NHAMCS data do not distinguish between mild and moderate/severe exacerbations of asthma. To address this, we excluded children with mild exacerbations by requiring the addition of nebulized albuterol treatment and placement in the most urgent triage categories. We recognize that this approach has not been formally validated. Clinical data needed to validate this approach are not available in the NHAMCS database. Furthermore, there might be a subgroup of patients with a history of asthma who might develop bronchiolitis and for whom steroid treatment is warranted. Without patient-specific data, the significance of this population is lost.
It is possible that some of the patients with asthma, bronchiolitis, or croup had comorbidities for which antibiotics or radiographs were indicated or contraindicated. The group of conditions that would need to be examined for a formal analysis of comorbidities was too large and complex to apply to this study. Although it would be possible to analyze data for patients with asthma, bronchiolitis, or croup and no comorbidities, the sample size would be reduced so greatly as to make the analysis meaningless. As multiple years of NHAMCS data become available, we may be able to undertake such an analysis.
The definition of radiograph in the NHAMCS data set is a binary (yes/no) variable and does not distinguish between chest radiographs and other radiographs. A qualitative review of data on the reasons for visits did not identify injuries or other comorbidities to account for the use of radiographs other than chest radiographs for asthma or bronchiolitis or lateral neck radiographs for croup.
Data before 2005 lacked key information about medication use in the ED; therefore, we were limited to a single year of data. This limitation reduced our overall sample size, so that we could not estimate with adequate precision the factors (eg, insurance status, geographic location, and demographic features of the patients being treated) associated with the provision of quality care. As more years of data become available, we should accumulate sufficient sample sizes to examine these factors.
Although the data provided much of the key information for determination of the use of medications, NHAMCS lacks indicators of disease severity and information about previous visits and previous use or failure of prescribed medications. Therefore, it is possible that some children presenting to the ED had already received steroids or had a prescription for steroids at home.
Despite these considerations, the NHAMCS is well suited for the study of practice variations from a public health or policy perspective. The availability of data for large numbers of patients, with detailed clinical information, and sophisticated sampling techniques provide nationally representative data that are unavailable elsewhere.
The failure of translation of evidence to practice in US EDs for such common pediatric respiratory illnesses as asthma, bronchiolitis, and croup is disappointing but not surprising. These results are similar to those of single-site studies that examined the quality of care for a range of conditions being treated in EDs and studies that documented a lack of implementation of evidence-based therapies in EDs.29,44–48
In Emergency Care for Children: Growing Pains, the Institute of Medicine noted that, “Without accountability, participants in the emergency care system need not accept responsibility for failures and can avoid making changes to improve the delivery of care.”1 The experience with the implementation of β-receptor blocker therapy for patients after acute myocardial infarction demonstrates that accountability can produce reliable changes that improve quality. When the US health care system was held accountable for performance to obtain National Committee for Quality Assurance accreditation, the rate of β-receptor blocker usage went from approximately two thirds to virtually 100%.49 There are data that suggest that the resulting relative reduction in mortality rates might be as great as 40%.50 This success has been attributed to a stepwise process including research, support of opinion leaders, development of guidelines and quality measures, use of measures by quality-assurance and accrediting organizations, and implementation of quality initiatives by professional societies. The results of our study quantifying current performance in ED care for common respiratory illnesses in children advance the stepwise process of improving the quality of pediatric ED care.
We conclude that there is underuse of proven therapies and overuse of diagnostic studies and treatments in the care of children with asthma, bronchiolitis, and croup in US EDs. We think that our data can be used by EDs for baseline comparisons of performance. We support the development of benchmarks that define quality care, research to quantify performance in comparison with such benchmarks, and studies on the dissemination and implementation of evidence-based guidelines.
We acknowledge the invaluable assistance of Pam West in the preparation of the manuscript.
- Accepted March 12, 2008.
- Address correspondence to Jane F. Knapp, MD, Department of Pediatrics, Children's Mercy Hospitals and Clinics, 2401 Gillham Rd, Kansas City, MO 64108. E-mail:
The authors have indicated they have no financial relationships relevant to this article to disclose.
What's Known on This Subject
Evidence exists for the optimal treatment of asthma, bronchiolitis, and croup in children. It is not known how well this evidence has been translated into practice in US EDs.
What This Study Adds
This is the first study to apply systematically developed performance indicators for the US ED care of children with common respiratory illnesses.
- ↵Institute of Medicine of the National Academies, Committee on the Future of Emergency Care in the United States Health System. Emergency Care for Children: Growing Pains. Washington, DC: National Academies Press; 2007
- ↵Centers for Disease Control and Prevention. Asthma prevalence, health care use and mortality: United States, 2003–2005. Available at: www.cdc.gov/nchs/products/pubs/pubd/hestats/ashtma03-05/asthma03-05.htm. Accessed October 16, 2008
- ↵Centers for Disease Control and Prevention. Respiratory syncytial virus. Available at: www.cdc.gov/ncidod/dvrd/revb/respiratory/rsvfeat.htm. Accessed February 8, 2008
- ↵Marx A, Török TJ, Holman RC, Clarke MJ, Anderson LJ. Pediatric hospitalizations for croup (laryngotracheobronchitis): biennial increases associated with human parainfluenza virus 1 epidemics. J Infect Dis.1997;176 (6):1423– 1427
- ↵Rowe BH, Spooner C, Ducharme FM, Bretzlaff JA, Bota GW. Early emergency department treatment of acute asthma with systemic corticosteroids. Cochrane Database of Systematic Reviews 2001, Issue 1. Art. No.: CD002178. DOI: 10.1002/14651858.CD002178
- Kairys SW, Olmstead EM, O'Connor GT. Steroid treatment of laryngotracheitis: a meta-analysis of the evidence from randomized trials. Pediatrics.1989;83 (5):683– 693
- ↵Ausejo M, Saenz A, Pham B, et al. The effectiveness of glucocorticoids in treating croup: meta-analysis. BMJ.1999;319 (7210):595– 600
- ↵National Asthma Education Program. Guidelines for the Diagnosis and Management of Asthma. Bethesda, MD: National Heart, Lung, and Blood Institute; 1991. Publication DHHS 91–3042
- ↵National Asthma Education and Prevention Program. Expert Panel Report 2: Guidelines for the Diagnosis and Management of Asthma. Bethesda, MD: National Heart, Lung, and Blood Institute; 1997. Publication DHHS 97–4051
- ↵National Asthma Education and Prevention Program. Expert Panel Report 3: Guidelines for the Diagnosis and Management of Asthma. Bethesda, MD: National Heart, Lung, and Blood Institute; 2007. Publication DHHS 08–4051
- ↵Pediatric Asthma Committee. Pediatric Asthma: Promoting Best Practice Guide for Management of Asthma in Children. Milwaukee, WI: American Academy of Allergy, Asthma, and Immunology; 1999
- ↵Cincinnati Children's Hospital Medical Center, Bronchiolitis Guideline Team. Evidence-Based Clinical Practice Guideline for Medical Management of Bronchiolitis in Infants Less Than 1 Year of Age Presenting With a First Time Episode. Cincinnati, OH: Cincinnati Children's Hospital Medical Center; 2006
- ↵American Academy of Pediatrics, Subcommittee on Diagnosis and Management of Bronchiolitis. Diagnosis and management of bronchiolitis. Pediatrics.2006;118 (4):1774– 1793
- ↵Russell K, Wiebe N, Saenz A, et al. Glucocorticoids for croup [Cochrane review]. Cochrane Database of Systematic Reviews 2004, Issue 1. Art. No.: CD001955. DOI: 10.1002/14651858.CD001955.pub2
- Hampers LC, Faries SG. Practice variation in the emergency management of croup. Pediatrics.2002;109 (3):505– 508
- ↵Guttmann A, Razzaq A, Lindsay P, Zagorsky B, Anderson GM. Development of measures of the quality of emergency department care for children using a structured panel process. Pediatrics.2006;118 (1):114– 123
- ↵National Center for Health Statistics. NHAMCS estimation procedures. Available at: www.cdc.gov/nchs/about/major/ahcd/nhamcs_est_proc.htm. Accessed July 9, 2007
- ↵National Center for Health Statistics. NHAMC scope and sample design. Available at: www.cdc.gov/nchs/about/major/ahcd/sampham.htm. Accessed July 9, 2007
- ↵National Center for Health Statistics. Public Use Microdata File Documentation, National Hospital Ambulatory Medical Care Survey, 2003. Hyattsville, MD: National Technical Information Service; 2005
- ↵National Center for Health Statistics. Public Use Microdata File Documentation, National Ambulatory Medical Care Survey, 2003. Hyattsville, MD: National Technical Information Service; 2005
- ↵US Food and Drug Administration. National Drug Code Directory, 1995 Edition. Washington, DC: Public Health Service; 1995
- ↵Perlstein PH, Kotagal UR, Bolling C, et al. Evaluation of an evidence-based guideline for bronchiolitis. Pediatrics.1999;104 (6):1334– 1341
- ↵Groce JB III. Translating evidence-based guidelines into performance measures for venous thromboembolism and acute coronary syndrome. Am J Health Syst Pharm.2007;64 (11 suppl 7):S25– S29
- Copyright © 2008 by the American Academy of Pediatrics