|Home | About | Journals | Submit | Contact Us | Français|
The purpose of this study was to examine the demographic and hospitalization characteristics of children hospitalized with lower extremity fractures in the United States in 2006.
Children aged 0 to 20 years with a diagnosis of lower extremity fracture in the 2006 Healthcare Cost and Utilization Project Kids’ Inpatient Database (KID) were included. Lower extremity fractures were defined by International Classification of Diseases, 9th Revision, Clinical Modification codes 820-829 under “Injury and Poisoning (800-999).” Patient demographic and hospitalization-related data were analyzed by chi-square testing and unbalanced analysis of variance.
There were more boys than girls with lower extremity fractures and 53% had private insurance as their primary payer. About one half of the children were between the ages of 13 and 20 years, but all ages were represented from age 0 to 20. White children accounted for 56%. Urban hospitalizations accounted for 93% of cases and 66 percent of admissions were to teaching hospitals. All patients had an average length of stay (LOS) 4.04 days, and infant patients had the longest average LOS of 5.46 days. The average number of diagnoses per patient was 3.07, and the average number of procedures per patient was 2.21. The average charge per discharge was $35,236, and the oldest patients had the largest average charge of $41,907. The average number of comorbidities increased with increasing patient age. There was a 55.6% greater mortality risk in non-teaching hospitals than in teaching hospitals and there was at least ten times the mortality risk in rural hospitals than in urban hospitals.
This study provides an understanding of the demographic and hospitalization characteristics of children with lower extremity fractures in the United States in 2006. This information may be useful in implementing measures to help prevent similar injuries in the future. Further research is required to determine causality of the associations found including increased mortality risk for this population at rural and non-teaching hospitals.
Lower extremity fractures are quite common in the pediatric population due to falls, non-accidental trauma, and motor vehicle collisions.1,2,3 To our knowledge, no national population-based study had been conducted to date that examined the demographic and hospitalization characteristics of pediatric patients who were hospitalized for lower extremity fractures. We report the specific demographic and hospitalization characteristics for children with lower extremity fractures in 2006 in the United States. The results were derived from a database of inpatient hospitalization usage by children across the United States. We hypothesized that inpatient mortality risk in patients with lower extremity fractures was lower in urban and teaching hospitals than rural and non-teaching hospitals.
The Healthcare Cost and Utilization Project (HCUP) Kids’ Inpatient Database4 (KID) is the only dataset on hospital use, outcomes, and charges designed to study children’s use of hospital services in the United States. The 2006 KID contains approximately 3.1 million pediatric discharges from 3,739 community, non-rehabilitation hospitals in 38 states, representing all four geographic census regions (northeast, midwest, west, and south). It includes a sampling of all hospital discharges where the patient was age 20 or less at admission during the year 2006. This can be extrapolated to a national estimation of 7.6 million pediatric hospital discharges. Patient demographic variables included age at time of admission, sex, race, and median household income quartiles based on the ZIP code of the family’s residence. Hospitalization variables included admission month and source, diagnostic and procedure codes, duration of stay, total charges, expected payer, and discharge disposition. Hospitals included in this database were divided into strata using six characteristics: ownership/control, bed size, teaching status, rural/urban location, US region, and hospital type (pediatric vs other). Bed capacity was categorized into small, medium, or large, and varied in specific bed capacity depending on whether the hospital was located in a rural area or was an urban non-teaching or urban teaching hospital.
Patients were included in this study if they were age 0 to 20 and had at least one of the above lower extremity fracture codes from the 15 available diagnosis codes in the database. The study group was then further divided into four age groups6 based on ages that would likely participate in common activities from a developmental standpoint. The four groups were: (1) infants (age <1 yr), (2) toddlers (age 1-2 yrs), (3) children (age 3-12 yrs), and (4) adolescents (age, 13-20 yrs). Data was presented in two categories: patient demographics and hospitalization-related data. Descriptive statistics were reported, and chi-square testing and unbalanced analysis of variance was used for determining the differences between age groups. Univariate logistic regression was used to analyze the impact of teaching hospital status and hospital location on mortality. A p-value less than 0.05 was considered statistically significant for all analyses. Data were analyzed using SAS software.7
There were a total of 11,903 patient admissions associated with lower extremity fractures out of 3.1 million records in the KID. Since there were 391 records without age, the final study group consisted of 11,512 records. Generally, increasing age was associated with increasing incidence of lower extremity fracture (Figure 1). The correlation between age and the admission volume was 0.8054 (p<0.0001).
Table 2 shows 60.41% of lower extremity fractures occurred between the ages of 13 and 20. The vast majority of these injuries (72.77%) occurred in boys and there was a male predominance in each of the four age groups (P<0.0001). This is consistent with previous studies8-12. The most prevalent racial group was white, accounting for 56% of admissions and 53% of all patients had private insurance. Increasing age was associated with a progressive shift of patients from low income households to high income households (p=0.0057). This implies that the household income is greater for older children with lower extremity fractures than their younger counterparts. There were more lower extremity fracture admissions in summer than other seasons.
Non-elective admission accounted for 92% (Table 3) of all admissions (p<0.0001) and also predominated in all for four age groups. Large capacity hospitals accounted for 61% of admissions. (p<0.0001) and the majority of patients in each age group were admitted to large capacity hospitals. The vast majority of hospitals (93%) were located in urban areas (p=0.0001). Most hospitals (66%) had teaching status (p<0.0001). The deaths among the four age groups were not uniformly distributed with the youngest group having the highest death rate 1.82% (P<0.0001).
A separate analysis was performed for length of stay (LOS), number of diagnoses, number of procedures, total charges, and number of comorbidities (Table 4).
The average LOS for all patients was 4.04 days. The infants stayed the longest in hospital.
The average number of diagnoses for all patients was 3.07. The average number of procedures each patient received was 2.21. Strictly, the average number of procedures patients received increased with patient age. The average charge for all patients is $35,235 with the oldest patients having the largest charges. The average number of comorbidities was 0.25. The unbalanced analysis of variance model showed these age-specific means within each parameter category were significantly different (p<0.0001).
We further tabulated the number of patients by age group and number of comorbidities (Table 5). The majority (81%) of admissions were not associated with a comorbidity, only seven were associated with five or more comorbidities, and these seven admissions were in the oldest age group. The adolescent group had a smaller proportion of admissions not associated with a comorbidity while having higher proportions of higher number of comorbidities than any other patient age categories. Chi square testing showed these proportions were significantly different (p<0.0001).
Table 6 shows the anatomic distribution of the fractures. Approximately half (49.9%) of fractures were related to the tibia, fibula, or ankle.
We also made a preliminary analysis on the place-of-injury using the E-code, see Table 7. The results showed that (62%) of the infants had accidents at their home, other residential, or an institutional location, while adolescents tended to be injured on a street or highway. These findings are similar to those previously published.13-15
There was a 55.6% greater mortality risk in non-teaching hospitals than in teaching hospitals and there was at least a ten times greater risk of mortality in rural hospitals than in urban hospitals.
This study analyzed the KID data with children aged 0 to 20 years who were hospitalized with lower extremity fractures in the United States in 2006. This included 11,512 admissions and $405 million in hospital charges. These injuries have a major impact on the health and well-being of children and their families including mental, physical and economic burdens.17
This study has limitations. Costs, reported in the KID as billed charges, may differ from the actual amount paid because of discounts, deductibles, copayments, and coinsurances. The true costs are likely to be underestimated in this study because uncovered charges, professional fees, time lost from school/work by caregivers, and other societal costs are not included in these estimates. Although the database allows for extrapolation to national estimates there is the potential for error in the extrapolation with this particular population since the KID does not include all US states and does not sample the regions of the United States equally. Some degree of regional variations in lower extremity fracture incidence is likely.
Studying racial disparity was particularly difficult given that nine out of 38 states did not report information on race in the KID. This creates a potential source of bias as racial distribution and other demographics vary by region and state. There were 2394 records missing race (20.8%). According to the 2000 US Census data18 regarding children younger than 20 years, the racial breakdown is 61% white, 15% black, 17% Hispanic, and 7% other races, compared with 56% white, 16% black, 21% Hispanic, and 7% other races in this study with race information. Hispanic children are relatively over-represented in this study, however, the lack of complete data regarding race makes drawing conclusions on this finding difficult.
Additionally, the KID database contains discharge-level records, not patient-level records. Therefore, individual patients who are hospitalized multiple times in one year may account for multiple records in KID. There is no uniform patient identifier available that allows a patient-level analysis with the KID.
The results show an increasing incidence of lower extremity fractures as children age, along with a clear gender disparity in which males were considerably more likely to sustain these injuries than females. Our findings show that children with lower extremity fractures were more likely to be admitted to large, urban, teaching hospitals. Additionally, urban and teaching hospitals were associated with a lower risk of inpatient mortality. These findings may guide hospital triage and transfer agreements.
There were more admissions for lower extremity fracture in summer than in other seasons, likely reflecting climate variation. The physical damage and financial burden that result from such an injury have a significant and sometimes lifelong impact on children and their families. Research has demonstrated that most of these injuries are preventable.19,20 Orthopaedists and pediatricians can be instrumental in preventing pediatric injuries by participating in patient education, research, and programs that promote safe play. For example, driving-habit education may have a big effect on reducing highway accidents. More research is needed to identify factors that are associated with different age groups so that age-specific measures may be designed and implemented. A quantitative model analyzing total charges, length of stay and number of procedures would be useful in quality improvement efforts.