OBJECTIVE. The goal was to examine the influence of sociodemographic characteristics and health care system factors on the utilization of hospital resources by US children ≤17 years of age with a diagnosis of traumatic brain injury.
METHODS. A retrospective analysis of data from the Healthcare Cost and Utilization Project Kids' Inpatient Database, from January 1, 2000, to December 31, 2000, was performed. National estimates of traumatic brain injury-associated hospitalization rates and resource use were calculated with Kids' Inpatient Database sample weighting methods.
RESULTS. Of 2516833 encounters between January 1, 2000, and December 31, 2000, 25783 cases involved patients ≤17 years of age with a recorded diagnosis of traumatic brain injury. On the basis of these data, there were an estimated 50658 traumatic brain injury-associated hospitalizations among children ≤17 years of age in the United States in 2000. The traumatic brain injury-associated hospitalization rate was 70 cases per 100000 children ≤17 years of age per year; 15- to 17-year-old patients had the highest hospitalization rate (125 cases per 100000 children per year). Pediatric inpatients accrued more than $1 billion in total charges for traumatic brain injury-associated hospitalizations in this study. In the multivariate regression models, older age, Medicaid insurance status, and admission to any type of children's hospital were associated with a longer length of stay for pediatric traumatic brain injury-associated hospitalizations. Older age, longer length of stay, and in-hospital death predicted higher total charges for traumatic brain injury-associated hospitalizations.
CONCLUSION. Pediatric traumatic brain injury is a substantial contributor to the health resource burden in the United States, accounting for more than $1 billion in total hospital charges annually.
- traumatic brain injury
- resource utilization
- length of stay
- total charges
- Kids' Inpatient Database
Traumatic brain injury (TBI) has been referred to as the “silent epidemic” by the Centers for Disease Control and Prevention (CDC) and others, because of its vast incidence and pressing need for additional research.1,2 Annually, an estimated 1.4 million individuals in the United States suffer TBI, resulting in 235000 hospitalizations, 50000 deaths, and $56.3 billion in direct and indirect costs.3 By comparison, an estimated 175000 individuals in the United States are diagnosed as having breast cancer and 44000 as having HIV/AIDS each year.4 TBI is the most common traumatic event involving the central nervous system, surpassing traumatic spinal cord injury by a factor of 10.4 The CDC, Congress, and health professionals have issued numerous calls to reduce the substantial morbidity, mortality, and resource burdens associated with TBI in the United States.2,5,6
TBI is particularly devastating among children. TBIs resulting from unintentional trauma constitute the primary cause of death among US youths <24 years of age and are often the most serious consequences of nonfatal injuries.7–9 Neuropsychological sequelae associated with pediatric TBI can influence key developmental processes, such as learning, emotional awareness, and social functioning.10,11 Furthermore, injury costs for children are often greater than those for adults, because they include expenditures for acute treatment, long-term rehabilitation, and loss of productivity for the parent or guardian.12 Despite the immense costs, both financial and emotional, of TBIs to families and society, there is a paucity of research on the resource burden associated with TBIs, particularly for the estimated 64000 pediatric hospitalizations that occur each year in the United States.3
Although several studies examined trends in TBI hospitalizations and explored the psychosocial sequelae of brain-injured patients, few researchers examined the resource burden associated with TBI events and fewer still focused on pediatric TBI.10,11,13–21 Investigations reporting TBI-associated resource utilization for pediatric inpatients often were limited by small nonrepresentative samples from single institutions or single states or by age categories that were too broad to allow meaningful comparisons according to developmental stage.12,22–26 To our knowledge, no studies have used a large inpatient sample of pediatric patients with TBIs to examine health resource utilization on a national level.
The purpose of this study was to assess the influence of patient and health care system factors on length of stay (LOS) and total hospital charges for TBI-associated hospitalizations for children ≤17 years of age in the United States. LOS and total hospital charges were selected because they are standard variables used by both the private sector and government agencies to monitor quality of care and resource utilization.27–29 This study also provides national estimates of the incidence of pediatric TBI-associated hospitalizations in the United States and the resource utilization associated with those hospitalizations.
Data from the Kids' Inpatient Database (KID), from January 1, 2000, to December 31, 2000, were used in this study.30 The KID is a component of the Healthcare Cost and Utilization Project maintained by the Agency for Healthcare Research and Quality.30 The KID is one of the only hospital administrative data sets designed specifically to assess the use of hospital services by newborns, children, and adolescents.31 Since its implementation in 1997, the KID has undergone only minor changes in the recorded variables, types of participating facilities, and number of hospitals sampled. In 2000, the KID collected hospital discharge information on pediatric treatments and resource utilization from 2784 hospitals in 27 states. The 2000 KID included nonfederal hospitals, short-term hospitals, academic medical centers, and specialty hospitals such as obstetrics/gynecology, ear/nose/throat, orthopedic, and pediatric hospitals.31 The 2000 KID excluded federal hospitals, short-term rehabilitation hospitals, long-term hospitals, psychiatric hospitals, and alcoholism/chemical dependency treatment centers.31 Hospitals were assigned to 6 strata for random selection, on the basis of ownership/control, bed size, teaching status, rural/urban location, US region, and hospital type (pediatric versus other).31 All data were at the discharge level; therefore, individuals who were hospitalized multiple times had multiple records in the KID. The 2000 KID data were selected because 2000 was the most recent year with complete data at the time of the study.
A series of International Classification of Diseases, 9th Revision, Clinical Modification, diagnosis codes were used to identify TBI-associated hospitalizations among children ≤17 years of age, consistent with CDC recommendations.15,32 Researchers reviewed all 15 diagnosis fields in each hospitalization record for the following TBI diagnosis codes: fracture of the vault or base of the skull (codes 800.0–801.9); other unqualified and multiple fractures of the skull (codes 803.0–804.9); intracranial injury, including concussion, contusion, laceration, and hemorrhage (codes 850.0–854.1); injury to the optic chiasm, optic pathways, or visual cortex (codes 950.1–950.3); head injury, unspecified (code 959.01); and shaken infant syndrome (code 995.55).
The KID is one of the only national, all-payer databases of hospital discharges for children.33,34 Total charges reported to the Agency for Healthcare Research and Quality by each state are edited by professional KID coders, to round values to the nearest dollar, to set 0 charges to “missing,” and to set excessively high (more than $1 million) or excessively low (less than $25) charges to “inconsistent.”34 Total charges do not include professional fees and noncovered charges.34
In the KID, hospital bed-size categories (small, medium, and large) are based on the number of short-term patient beds, the geographic region of the hospital, and the teaching status of the hospital. Hospital type follows the children's hospital categorization method used by the National Association of Children's Hospitals and Related Institutions. General children's hospitals are freestanding, general, acute-care children's hospitals. Children's units in general hospitals are nonfreestanding, general, acute-care hospitals, typically within larger hospitals. Specialty children's hospitals include orthopedic, rehabilitative, psychiatric, and other specialty-focused inpatient facilities. Nonchildren's hospitals admit patients of all ages and do not have a specific pediatric focus. The charges considered in our analyses were limited to acute care. If a patient is discharged from acute care to a rehabilitation unit in the same hospital, then generally KID coders consider this a transfer to a specialty hospital. Specialty hospital charges were not included in this study, because our investigation focused on acute care and because very few TBI cases were associated with those hospitals.
Data analyses were conducted with SAS (version 9.1; SAS Institute, Cary, NC) and Stata SE (version 8.0; Stata Corp, College Station, TX) statistical software, to account for the weighting structures of the KID. SAS software was used for calculations of frequencies (unweighted sample and national estimates), 95% confidence intervals (95% CIs) for national estimates, percentages, means, medians, and sums and analyses of unadjusted means. Stata SE software was used for regression analyses and calculations of associated 95% CIs. Statisticians at the KID provided discharge-level and hospital-level statistical weights to account for the complex survey design and sampling procedures. The KID was designed specifically to provide national estimates of hospitalizations with discharge-level statistical weights, as performed in this study.31
The actual sample size is a statistically unweighted number and is noted when presented. All other frequencies, percentages, rates, means, medians, and sums are national estimates calculated with the statistical weights. National estimates of pediatric TBI-associated hospitalizations for children ≤17 years of age were calculated according to age, gender, race, primary expected payer, mortality status, hospital bed size, and hospital type, with 95% CIs. Race was excluded from additional subanalyses because of the relatively large percentage of missing data (15.8%). Rates of TBI-associated hospitalizations per 100000, according to age group and gender, were generated with publicly available US Census 2000 denominator data.35
The distributions of the outcome variables (LOS and total charges) were examined and found to be skewed rightward, with long, heavy, right tails. We chose ordinary least-squares (OLS) linear regression as the primary tool of analysis in this study despite these non-normally distributed, heteroscedastic, outcome data. Although linear regression assumes dependent variable normality and homoscedascity (similar outcome variable variance across all values of a predictor variable), Diehr et al29 demonstrated that OLS linear regression may be used with untransformed utilization data in very large data sets to provide unbiased estimates of regression parameters. Initially, unadjusted mean LOS values and total charges according to patient and hospital characteristics were compared by using Student's t tests and analyses of variance. A P value of <.05 was considered significant. LOS values and total charges were then regressed separately with respect to each of the independent variables of age, gender, race, expected primary payer, mortality status, hospital bed size, and hospital type, to identify significant covariates. Variables that were significant in these latter analyses were included in an OLS multivariate regression model. Separate models were used for LOS and charges. LOS was entered into the multivariate regression model as a continuous covariate during analysis of total hospital charges.
This study was approved by the institutional review board of the Columbus Children's Research Institute.
Demographic Features and General Characteristics
The KID collected data on 2516833 hospitalizations between January 1, 2000, and December 31, 2000. Of those hospitalizations, 25783 cases involved patients ≤17 years of age with a recorded diagnosis of TBI. On the basis of these data, there were an estimated 50658 (95% CI: 46466–54849) TBI-associated hospitalizations among children ≤17 years of age in the United States in 2000 (Table 1). The rate of TBI-associated hospitalizations for all children examined in this study was 70 hospitalizations per 100000 children ≤17 years of age. The rate of TBI-associated hospitalizations for white children was 54 hospitalizations per 100000 children ≤17 years of age, whereas the rate for black children was 51 hospitalizations per 100000 children, although these rates were underestimates because of missing data regarding children's race. Male children had twice the rate of female children (Fig 1). The number of TBI-associated hospitalizations peaked in late adolescence (Table 1). Among those with a recorded race (84.2%), white children accounted for the majority of patients hospitalized for treatment of a TBI (63.3%). Private insurance (63.0%) was the most common expected primary payer, followed by Medicaid (25.7%). Most patients were admitted to large, nonchildren's hospitals (55.3%). Overall, 3.8% of pediatric patients with TBIs died in the hospital, including 5.1% of 15- to 17-year-old patients.
Unadjusted Analyses of LOS and Total Charges
Overall, the mean LOS for pediatric TBI-associated hospitalizations was 4.5 days (median: 2 days). However, the range was 1 to 4 days when the 25% of patients with the shortest LOS and the 25% of patients with the longest LOS were excluded (Table 2). In total, there were 228988 inpatient days for treatment of TBIs in 2000. Only 10% of the encounters studied were >10 days in length, but the top 1% of encounters were >40 days. TBI-associated hospitalizations incurred an average of $20325 (median: $8056) in total charges for inpatient care, with a range of $3676 to $18563 when the 25% of patients with the least expensive total charges and the 25% of patients with the most expensive total charges were excluded. In total, pediatric inpatients with a diagnosis of TBI accrued $1005371887 in hospital charges in 2000. The most costly 1% of hospitalizations incurred more than $200000 per case in total charges.
All patients had a median LOS of 2 days, except for those who died in the hospital (Fig 2). Compared with all other age groups, adolescents 15 to 17 years of age had the longest average LOS (5.3 days; P < .001) and highest average total charges ($26578; median: $11000; P < .001) per hospitalization. LOS and total charges did not differ according to gender (P = .10 and P = .44 for LOS and total charges, respectively). Compared with privately insured children, children covered by Medicaid stayed longer (P < .001) and were charged more per hospitalization (P < .001). Hospitalizations resulting in death had short LOS (mean: 3.0 days; median: 1 day) and high total charges (mean: $36075; median: $23100).
Pediatric patients treated at a children's general hospital or a children's unit in a general hospital incurred higher total charges (P < .001) and stayed longer (P < .001) than did patients admitted to a nonchildren's hospital (Fig 3). Children treated in large hospitals accrued greater average total charges (P < .001) and had longer average LOS (P < .001) than did those treated in smaller hospitals.
Multivariate Analyses of LOS and Total Charges
With adjustment for age, expected primary payer, hospital type, and mortality status (Table 3), older age, a primary payer of Medicaid, and admission to any type of children's hospital were associated significantly with longer LOS for pediatric patients with TBIs. Age was a significant predictor of longer LOS only for 15- to 17-year-old patients (adjusted difference: 1.7 days; 95% CI: 1.2–2.3 days; P < .001, with <1-year-old patients as the reference group). Multivariate analysis showed that admission to a children's general hospital or children's unit in a general hospital predicted >1-day longer LOS, compared with admission to a nonchildren's hospital. Patient death was found to be predictive of shorter LOS for TBI-associated hospitalizations (adjusted difference: −1.8 days; 95% CI: −2.4 to −1.2 days; P < .001), compared with those who were discharged alive.
Controlling for potentially confounding variables (hospital type, mortality status, and LOS), mean total charges increased as age increased, up to an adjusted difference of $7876 more per stay for patients 15 to 17 years of age (95% CI: $6493–$9259; P < .001), compared with patients <1 year of age (Table 3). Hospital type was not associated significantly with total charges in multivariate analysis. Each day a patient spent in the hospital with a TBI was associated with an adjusted increase of $3915 (95% CI: $3495–$4334; P < .001) in mean total charges. Patients who died in the hospital incurred $21828 (95% CI: $18418–$25238; P < .001) more in mean total charges per stay, compared with those who were discharged alive.
Overall, there were an estimated 50658 pediatric TBI-associated hospitalizations among children ≤17 years of age in the United States in 2000, with more than $1 billion in inpatient charges. The actual sample of 25783 encounters is, to our knowledge, the largest pediatric TBI population examined to date, particularly among studies providing national estimates. On the basis of data reported by the Healthcare Cost and Utilization Project using the KID database, the total charges calculated for pediatric hospitalizations attributable to TBI would make it the fifth most expensive hospital diagnosis for children in 2000, behind pneumonia at $1.3 billion and ahead of acute bronchitis at $880 million.36 This calculated pediatric TBI total is consistent with published estimates of inpatient charges for TBI among all age groups ($6.5 billion, adjusted for inflation).3,15,37,38
Hospitalization rates according to age and gender were similar to rates calculated by the CDC in its latest report on TBI in the United States and by authors of other large epidemiologic studies.3,15 In this study, the overall rate for children admitted to US hospitals with a diagnosis of TBI was 70 cases per 100000 children ≤17 years of age. This rate was slightly lower than the rate of 77 cases per 100000 children ≤20 years of age published by the CDC; the difference was most likely attributable to the inclusion by the CDC of young adults 18 and 19 years of age, a group with a known high rate of TBI.3 The rate of TBI-associated hospitalizations for male children was twice that of female children, consistent with CDC publications and previous research.3,12,16,18,39
Also consistent with previous research, 15- to 17-year-old patients had the highest incidence rates of hospitalization and in-hospital deaths, compared with other age groups.3,40 Although our data did not record the external cause of injury, the CDC reported that adolescents of driving age were involved in substantially more motor vehicle crashes than were younger children, presumably because they were both passengers and newly licensed drivers. Motor vehicle crashes are the second leading cause of emergency department visits attributable to TBI in the United States and the leading cause of TBI-associated hospitalizations and deaths.3
The average LOS for all TBI-associated hospitalizations was 4.5 days, consistent with other studies that examined large populations of children hospitalized with TBIs.16 Fifteen- to 17-year-old patients had the longest hospital stays for treatment of TBIs, compared with all other age groups. This finding is supported by previous research that suggested that adolescents of driving age sustain the most severe, resource-intensive, brain injuries, because of the disproportionately high rate of occurrence of motor vehicle crashes in this age group.3,40 There was no statistically significant difference in LOS between male and female patients after adjustment for covariates. Although Morrison et al16 reported a trend toward girls having worse outcomes than boys after TBIs, their only statistically significant finding was a longer ICU LOS for girls.
After adjustment for potentially confounding variables, pediatric patients covered by Medicaid stayed longer in the hospital than did children covered by other insurance types. This is an important finding, because more than one fourth of the TBI-associated hospitalizations examined in this study were billed to Medicaid. Hospital administrators and public health officials may benefit by examining whether longer hospital stays among pediatric patients covered by Medicaid lead to better health outcomes and/or patient and family education or constitute a symptom of the systemic inefficiency associated with treating underserved populations.5,41
Children treated for TBIs in freestanding children's hospitals and children's units in general hospitals stayed longer than did other patients, after adjustment for potentially confounding variables. Case mixture, a measure of the variation in patients' medical conditions within a single hospital, might have played a role in this finding. Previous research indicated that children's hospitals treat sicker patients than do nonchildren's hospitals, which potentially leads to longer LOS.27,42 Our analysis detected a difference in LOS of ∼1 day between children admitted to children's hospitals versus nonchildren's hospitals, but we could not adjust for case mixture because of lack of data. Studies that adjusted for case mixture found no statistically significant difference in patient LOS between children admitted to children's versus nonchildren's hospitals.27
The overall mean and median total charges per hospitalization were slightly higher than the 1996 estimates for similar age groups reported by Schootman et al,15 but these differences are nominal in light of inflation and variations in sampling techniques. Older adolescents sustained the costliest brain injuries, even after adjustment for LOS. The greater frequency of motor vehicle crashes in this age group likely resulted in higher hospital charges because of the often-severe nature of these events. Other studies investigating TBI-associated hospital charges related to external causes of injury reported higher charges among those involved in motor vehicle crashes.6,12,40
Patients who died in the hospital because of TBI had shorter stays and higher total charges, compared with those who were discharged to home or to another facility. This is likely related to the intensive care measures taken routinely by hospital personnel in cases of life-threatening injuries to children. It may be assumed that expensive medical interventions were administered for cases of in-hospital death examined in this study, which led to the higher charges observed for this group of patients. Other TBI studies reported that life-threatening injuries were associated with extremely high hospital charges.12,15,20
Although there were gross differences in total mean (and median) charges according to payer type, these differences disappeared with control for confounding variables. Previous research indicated that payer status was not predictive of costs when charges for a large sample of TBI hospitalizations across all age groups were examined, although researchers investigating other pediatric diagnoses found increases in charges for certain primary payers.15,27,41,42
In this study, there was a trend toward higher mean total charges billed to patients admitted to children's hospitals for treatment of TBIs, but no statistically significant association between hospital type and charges was demonstrated. Merenstein et al27 found that patients admitted to children's hospitals paid more than those admitted to nonchildren's hospitals for common pediatric conditions, although those researchers did not examine trauma specifically in their analyses.
Several methodologic limitations exist in this study. First and most salient, the KID does not record the severity of the TBI. Steiner et al43 cited lack of clinical detail, which might have provided better indicators of severity, as a major drawback of using the KID. Severity unequivocally plays a role in determining the LOS and total charges associated with a hospital visit.12,15,20,44,45 Unfortunately, there is no consensus in the literature regarding how to adjust for severity with databases such as the KID.42,45,46 Second, the external cause of TBIs could not be examined because E-codes were missing for a substantial proportion of records in the KID. Previous TBI studies reported significant correlations between cause of injury and hospital charges.12,16,20 Third, total charges in this study represent billed charges and do not reflect actual payments, professional fees or payments, or noncovered charges. Billed charges do not represent the actual costs of hospital services, such as procedures and consultations, because these are subject to complex accounting practices; this makes them troublesome to quantify in dollar amounts and limits their comparability across providers.28,29 Some facilities offer cost-to-charge ratios for analysis of billed charges. This approach assumes that costs are known for all providers, which is not the case for a large diverse data set such as the KID.29 The Healthcare Cost and Utilization Project has not yet developed a formula for estimating costs from charges in the KID, as it has with certain other data sets under its auspices.27,41 Furthermore, Max et al38 estimated that hospitalization costs represent only 5% of the total direct and indirect economic burden of TBI; therefore, even a comprehensive calculation of hospital costs associated with TBIs would not be sufficient to approximate the total economic burden they engender. Another limitation of the study is that all data are at the discharge level; therefore, individuals who were hospitalized multiple times have multiple records in the KID. Finally, there is no way to measure whether changes in LOS or total charges were associated with changes in the health outcomes of injured patients. Such outcome assessments, although important, are beyond the scope of this study.
Within the ever-changing and increasingly costly milieu of health care, it is essential to identify risk factors and to develop targeted interventions to reduce the morbidity, mortality, and resource burdens associated with pediatric TBI. The inpatient charges alone calculated in this study totaled more than $1 billion, making TBI one of the most costly hospital diagnoses for children. Primary prevention of TBI is the preferred means of reducing its impact on hospitals, families, and society. Intervention strategies to prevent pediatric TBI and evaluations of these strategies are critical, because TBI is especially devastating among children.
A.J.S. received a student stipend from the Samuel J. Roessler Memorial Scholarship Fund of the Ohio State University College of Medicine and Public Health; S.G.H. was supported by an Alpha Omega Alpha Student Research Fellowship.
We gratefully acknowledge statistical advice from Soledad A. Fernandez, PhD (Ohio State University Center for Biostatistics).
- Accepted March 2, 2006.
- Address correspondence to Gary A. Smith, MD, DrPH, Center for Injury Research and Policy, Columbus Children's Hospital, 700 Children's Dr, Columbus, OH 43205. E-mail:
The authors have indicated they have no financial relationships relevant to this article to disclose.
- ↵Langlois JA, Rutland-Brown W, Thomas KE. Traumatic Brain Injury in the United States: Emergency Department Visits, Hospitalizations, and Deaths. Atlanta, GA: Centers for Disease Control and Prevention, National Center for Injury Prevention and Control; 2004
- ↵Brain Injury Association. TBI Incidence. MacLean, VA: Brain Injury Association of America; 2002. Available at: www.biausa.org/BIAUSA.ORG/word.files.to.pdf/good.pdfs/factsheets/TBIincidence.pdf. Accessed January 5, 2005
- ↵Centers for Disease Control and Prevention. Web-Based Injury Statistics Query and Reporting System, 2000. Available at: www.cdc.gov/ncipc/wisqars/. Accessed July 1, 2005
- ↵King WJ, MacKay M, Sirnick A. Shaken baby syndrome in Canada: clinical characteristics and outcomes of hospital cases. CMAJ.2003;168 :155– 159
- ↵Merenstein D, Egleston B, Diener-West M. Lengths of stay and costs associated with children's hospitals. Pediatrics.2005;115 :839– 844
- ↵Brooten D. Methodological issues linking costs and outcomes. Med Care.1997;35 (suppl):NS87–NS95
- ↵Agency for Healthcare Research and Quality. Healthcare Cost and Utilization Project Kids' Inpatient Database (KID), 2000. Available at: www.ahcpr.gov/data/hcup. Accessed May 19, 2005
- ↵Healthcare Cost and Utilization Project. Overview of the HCUP Kids' Inpatient Database (KID), 2000. Rockville, MD: Agency for Healthcare Research and Quality; 2003
- ↵Marr A, Coronado V, eds. Central Nervous System Injury Surveillance Data Submission Standards, 2002. Atlanta, GA: Centers for Disease Control and Prevention, National Center for Injury Prevention and Control; 2004
- ↵Killingsworth JB, Tilford JM, Parker JG, Graham JJ, Dick RM, Aitken ME. National hospitalization impact of pediatric all-terrain vehicle injuries. Pediatrics.2005;115 (3). Available at: www.pediatrics.org/cgi/content/full/115/3/e316
- ↵Healthcare Cost and Utilization Project. Description of KID Data Elements Inpatient Core File. Rockville, MD: Agency for Healthcare Research and Quality; 2003
- ↵US Census Bureau. Census 2000, Summary File 1. Available at: http://factfinder.census.gov/Servlet/DTTable?_bm=y&-geo_id=01000US&-ds_name=DEC_2000_SF1_U&-_lang=en&mt_name=DEC_2000_SF1_U_P012&-format=&-CONTEXT=df. Accessed July 13, 2005
- ↵Agency for Healthcare Research and Quality. Care of Children and Adolescents in US Hospitals. Rockville, MD: Agency for Healthcare Research and Quality; 2003. HCUP Fact Book 4, AHRQ publication 04-0004. Available at: www.ahrq.gov/data/hcup/factbk4/factbk4.htm. Accessed May 21, 2005
- ↵National Aeronautics and Space Administration. Consumer Price Index (CPI) inflation calculator. Available at: www.jsc.nasa.gov/bu2/inflateCPI.html. Accessed January 14, 2005
- ↵Miller TR, Romano EO, Spicer RS. The cost of childhood unintentional injuries and the value of prevention. Future Child.2000;10 :137– 163. Available at: www.futureofchildren.org/information2826/information_show.htm?doc_id=69762. Accessed July 23, 2005
- ↵Meurer JR, Kuhn EM, George V, Yauck JS, Layde PM. Charges for childhood asthma by hospital characteristics. Pediatrics.1998;102 (6). Available at: www.pediatrics.org/cgi/content/full/102/6/e70
- ↵Srivastava R, Homer CJ. Length of stay for common pediatric conditions: teaching versus nonteaching hospitals. Pediatrics.2003;112 :278– 281
- Copyright © 2006 by the American Academy of Pediatrics