Abstract
Objective
Although neurologic disorders are among the most serious acute pediatric illnesses, epidemiologic data are scarce. We sought to determine the scope and outcomes of children with these disorders in the US.
Design
Retrospective cohort study
Setting
All non-federal hospitals in 11 states encompassing 38% of the US pediatric population.
Patients
Children 29 days-19 years old hospitalized in 2005
Interventions
None
Measurements and Main Results
Using ICD-9-CM codes, we identified admissions with neurological diagnoses, analyzed patient and hospitalization characteristics, and generated age- and sex-adjusted national estimates. Of 960,020 admissions in the 11 states, 10.7% (103,140) included a neurological diagnosis, which yields a national estimate of 273,900 admissions of children with neurological diagnoses. The most common were seizures (53.9%) and traumatic brain injury (17.3%). Children with neurological diagnoses had nearly 3 times greater intensive care unit (ICU) use than other hospitalized children (30.6% vs. 10.6%, p<0.001). Neurological diagnoses were associated with nearly half of deaths (46.2%, n=1,790). Among ICU patients, children with neurological diagnoses had more than 3 times the mortality of other patients (4.8% vs.1.5%, p<.001). Children with neurological diagnoses had a significantly longer median hospital LOS than other children (3 days [1, 5] vs. 2 days [2,4], p<.001) and greater median hospital costs ($4,630 [$2,380, $9,730] vs. $2,840 [$1,520, $5,550], p<.001).
Conclusions
Children with neurological diagnoses account for a disproportionate amount of ICU stays and deaths compared to children hospitalized for other reasons.
Keywords: epidemiology, neurological outcome, pediatric, intensive care, pediatric neurocritical care, hospitalization
INTRODUCTION
Neurological diagnoses are leading causes of morbidity and mortality among children and include both traumatic and non-traumatic injuries and conditions. According to a recent Centers for Disease Control and Prevention report, traumatic brain injury (TBI) is responsible for over 60,000 hospitalizations and 6,000 deaths of US children up to age 19 (1). Schenler et al. found similar figures and specified both a case-fatality of 3.8% and over $1 billion in total inpatient charges per year for TBI patients (2). Compared to TBI, stroke has a lower incidence but higher mortality, with over 5,000 stroke hospitalizations among children and young adults (3) and a mortality rate of 14.6% for ischemic stroke and 19.5% for intracerebral hemorrhage (4).
Two single-center studies found that a substantial proportion of in-hospital deaths were associated with at least one neurological disorder. In a study of a tertiary children’s hospital in the United Kingdom, Ramnarayan, et al. found that the primary diagnosis at death among hospitalized children was neurological in 7.0% of ward patients and 9.3% of intensive care unit (ICU) patients. However, incidence and results by specific diagnosis were not investigated (5). In a US tertiary pediatric ICU (PICU), Au et al. found that, regardless of primary diagnosis, neurological diagnoses were the proximate cause of death in the majority of patients (6).
Given the high morbidity and mortality associated with pediatric neurological diagnoses and recent advances in neurological monitoring and treatment (7, 8), we sought to gain a more comprehensive understanding of the scope and characteristics of children that could benefit from such care. We studied a large sample of hospitalizations in the United States to estimate the incidence, patient characteristics, mortality, and costs of caring for children hospitalized with neurological diagnoses.
MATERIALS AND METHODS
Patients
We constructed a patient database using discharge data for calendar year 2005 from 11 states: Texas (9), Virginia (10), and 9 states from the State Inpatient Databases (SID) of the Healthcare Cost and Utilization Project (HCUP), Agency for Health Care Research and Quality (11) (Arizona, Florida, Iowa, Maryland, Massachusetts, New Jersey, New York, North Carolina, and Washington). The databases include information about all admissions to non-federal hospitals (i.e., excludes military, Indian Health Service, and Veterans Administration hospitals). These 11 states comprised 37.6% of the 2005 US pediatric population (12) and were selected on the basis of their geographic diversity, population sizes, and availability of needed data elements. Data included demographics (age, race, sex, insurance status); up to 9 discharge diagnoses as classified by the International Classification of Diseases, 9th revision, clinical modification (ICD-9-CM) (13); hospital mortality; hospital and ICU lengths of stay (LOS); and associated charges. Texas provided subject age by age group only; all other states provided age in years. We also obtained US Census population data to determine age- and sex-adjusted incidences and generate weighted national estimates of the number of cases, deaths, and resource use (12).
Case Selection and Definitions
We studied children ages 29 days to 19 years old at hospital admission to take advantage of age groupings used by the US Census and to remove perinatal neurological diagnoses from analysis, because organization of care for neonates differs substantially from that of older children. We use the term “infants” to refer to children between ages 29 and 364 days old. Our goal was to identify patients who were treated for active, as opposed to chronic, neurological disease during hospitalization. We sought to do this by excluding all cases of neurological disease that were considered unlikely to have been a primary reason for hospitalization. Examples include ataxia, dystonia, spinal muscular atrophy, and muscular dystrophy. A full list of selected diagnoses and ICD-9-CM codes is provided in the Appendix. We classified patients as having traumatic and/or non-traumatic neurological diagnose using ICD-9-CM codes. Non-traumatic diagnoses were categorized as meningitis; encephalitis, myelitis, and encephalomyelitis; brain and spinal cord abscess; CNS encephalopathies; CNS neoplasms; seizures; ischemic and hemorrhagic strokes; hydrocephalus; migraines; syncope; and other non-traumatic neurological diagnoses. Traumatic diagnoses included blunt and penetrating head and spinal cord trauma. We identified severe TBI using the American College of Surgeons Committee on Trauma field triage decision scheme, mapping the anatomic and physiologic indications to ICD-9 Clinical Modification codes (14). Diagnoses included TBI accompanied by diagnoses indicating prolonged loss of consciousness, open fracture of the skull, or traumatic shock. Over half (58.2%) of neurological diagnoses were the primary discharge diagnosis, and 70.5% were the first or second diagnosis. We report the incidence of each diagnosis to account for children admitted with multiple diagnoses, but we include each admission only once when reporting the overall incidence of admissions with neurological disorders.
We identified children with underlying conditions using Pediatric Complex Chronic Conditions definitions developed by Feudtner et al., excluding diagnoses used for acute seizures (15). Neurological disease status and complex chronic conditions status were determined independently. We considered children previously healthy if they lacked a complex chronic condition. We identified children receiving ICU care by the presence of charges for ICU care. We classified hospitals as children’s hospitals or general hospitals based on the National Association of Children’s Hospitals and Related Institutions children’s hospital classification (http://www.childrenshospitals.net).
Statistical Analyses
We used data from the US Census to generate age- and sex-adjusted national estimates of hospital and ICU admission by neurological diagnosis. We estimated costs by multiplying charges by hospital-specific cost-to-charge ratios derived from the Centers for Medicare and Medicaid Services Provider Specific File. Because continuous data were not normally distributed, we made comparisons using the Mann-Whitney U test, and we compared categorical data by chi-square or Fisher’s exact tests as appropriate. When values were missing (e.g., race was missing from 7.0% of admissions and payer from 10.6%), percentages were calculated based only on subjects with non-missing values. We conducted analyses in Stata 11.2 (Stata Corp., College Station, TX).
RESULTS
Of 960,020 hospitalizations in the 11 states, 103,140 (10.7%) had neurological diagnoses. This represents an overall age- and sex-adjusted incidence of 3.4/1,000 children per year (Table 1). The highest incidence was among infants (8.6/1,000, n=13,422), and the lowest was among 10-14 year olds (2.4/1,000, n=18,568). The median age was 5 years older for those with neurological diagnoses than for those without neurological diagnoses (9 [2, 15] vs. 4 [0,14] years, p<.001) (Table 1). Children with neurological diagnoses made up a fourth (25.8%, n=31,530) of ICU admissions. Children with neurological diagnoses had a longer median hospital LOS than other children (3 [1,5] vs. 2 [2,4] days, p<.001) and greater median hospital costs ($4,360 [$2,380, $9,730] vs. $2,840 [$1,520, $5,550], p<.001). From these results, we estimate that 273,900 admissions of children with neurological diagnoses occurred in the US in 2005 (Table 2); 83,780 (30.6%) included ICU care, and 4,750 (1.7%) children died. There were 1.5 million hospitalization days and $3.1 billion of incurred hospital costs.
Table 1.
Patient characteristics by presence of a neurological diagnosis and ICU status
| All Admissions | ICU Admissions | |||||
|---|---|---|---|---|---|---|
| With Neurological Diagnosis N=103,140 |
Without Neurological Diagnosis N=856,880 |
p | With Neurological Diagnosis N=31,530 |
Without Neurological Diagnosis N=90,560 |
p | |
|
| ||||||
| Percent of admissions | 10.7 | 89.3 | 25.8 | 74.2 | ||
|
| ||||||
| Annual incidence (per 1,000 children) |
3.4 | 27.9 | 1.0 | 3.0 | ||
|
| ||||||
| Male, % | 55.7 | 53.3 | <.001 | 58.6 | 55.5 | <.001 |
|
| ||||||
| Age, y | <.001 | <.001 | ||||
| Mean ± SD | 8.9 ± 6.6 | 6.9 ± 7.0 | 9.2 ± 6.7 | 7.1 ± 7.1 | ||
| Median (IQR) | 9 (2, 15) | 4 (0, 14) | 9 (2, 16) | 4 (0, 15) | ||
|
| ||||||
| Age group, % by diagnosis status | <.001 | <.001 | ||||
| 29-364 d | 13.0 | 29.3 | 12.9 | 29.2 | ||
| 1-4 y | 23.7 | 21.9 | 22.8 | 21.1 | ||
| 5-9 y | 17.6 | 12.6 | 16.8 | 11.7 | ||
| 10-14 y | 18.0 | 13.8 | 17.8 | 13.7 | ||
| 15-19 y | 27.7 | 22.4 | 29.8 | 24.3 | ||
|
| ||||||
| Race/ethnicity (%) | <.001 | <.001 | ||||
| White | 49.4 | 47.9 | 49.4 | 46.8 | ||
| Black | 18.4 | 19.7 | 18.1 | 21.3 | ||
| Hispanic | 22.0 | 23.0 | 21.9 | 21.7 | ||
| Other | 10.2 | 9.4 | 10.6 | 10.3 | ||
|
| ||||||
| Insurance coverage (%) | <.001 | <.001 | ||||
| Medicare | 0.2 | 0.3 | 0.4 | 0.4 | ||
| Medicaid | 42.9 | 42.7 | 41.8 | 42.1 | ||
| Private insurance | 48.7 | 49.0 | 49.1 | 48.9 | ||
| Self-Pay/Uninsured | 8.1 | 8.1 | 8.8 | 8.6 | ||
|
| ||||||
| Underlying condition (%) | <.001 | <.001 | ||||
| Neuromuscular | 19.7 | 1.7 | 23.1 | 4.0 | ||
| Cardiovascular | 3.9 | 3.7 | 6.1 | 14.7 | ||
| Respiratory | 1.3 | 1.1 | 1.9 | 3.3 | ||
| Renal | 0.5 | 0.8 | 0.6 | 1.2 | ||
| Gastrointestinal | 0.4 | 1.1 | 0.3 | 1.5 | ||
| Hematologic/Immunologic | 0.7 | 0.9 | 0.7 | 1.2 | ||
| Metabolic | 1.2 | 0.8 | 1.3 | 1.7 | ||
| Other congenital or genetic | 2.7 | 1.7 | 3.0 | 5.2 | ||
| Malignancy | 7.2 | 4.2 | 8.4 | 5.7 | ||
| Any underlying condition | 25.6 | 12.2 | 29.9 | 25.9 | ||
|
| ||||||
| Hospital mortality, % | 1.7 | 0.2 | <.001 | 4.8 | 1.5 | <.001 |
|
| ||||||
| Mechanical ventilation, % | 8.4 | 1.8 | <.001 | 23.9 | 9.9 | <.001 |
|
| ||||||
| Hospital length of stay (d) | <.001 | <.001 | ||||
| Mean ± SD | 5.5 ± 11.9 | 4.3 ± 9.9 | 8.4 ± 14.3 | 7.9 ± 14.6 | ||
| Median (IQR) | 3 (1, 5) | 2 (2, 4) | 4 (2, 9) | 4 (2, 8) | ||
|
| ||||||
| Hospital Cost, US dollars | <.001 | <.001 | ||||
| Mean ± SD | $11,360 ± $26,780 | $6,530 ± $18,680 | 22,050 ± 38,650 | $18,540 ± $37,820 | ||
| Median (IQR) | $4,360 ($2,380, $9,730) |
$2,840 ($1,520, $5,550) |
$10,140 ($4,780, $23,440) |
$7,560 ($3,740, $18,140) |
||
y = year; SD = standard deviation; IQR = interquartile range; d = day
Table 2.
US incidence, number of cases, and deaths associated with neurological diagnoses by age
| Annual incidence (per 1,000 population) |
Number of US cases |
Number of US deaths |
|
|---|---|---|---|
| Total hospitalized | 3.4 | 273,940 | 4,750 |
| Age group n (%) | |||
| 29-364 d | 8.6 | 35,260 | 750 |
| 1-4 y | 4.0 | 64,030 | 870 |
| 5-9 y | 2.5 | 48,110 | 590 |
| 10-14 y | 2.4 | 49,710 | 710 |
| 15-19 y | 3.7 | 76,830 | 1,830 |
|
| |||
| ICU hospitalizations | 1.0 | 83,780 | 4,030 |
| Age group | |||
| 29-364 d | 2.6 | 10,650 | 590 |
| 1-4 y | 1.2 | 18,790 | 760 |
| 5-9 y | 0.7 | 14,050 | 510 |
| 10-14 y | 0.7 | 15,020 | 590 |
| 15-19 y | 1.2 | 25,280 | 1,580 |
ICU = intensive care unit; d = day; y = year
The most common neurological diagnosis was seizure (53.9% of admissions with a neurological diagnosis), and 2.4% of patients with seizure had status epilepticus (Table 3). TBI was the second most common diagnosis. Collectively, traumatic injuries to the brain and/or spinal cord were present in 17.9% of children with neurological diagnoses. The second most common non-traumatic diagnosis was meningitis (8.6% of admissions with neurological diagnoses). Most cases of meningitis (75.6%) were viral. Stroke was relatively uncommon (1.8% of admissions with neurological diagnoses). Age distribution varied by diagnosis. Nearly half of children with seizures (43.7%) and status epilepticus (49.4%) were under 5 years old. In contrast, nearly half of traumatic injuries (45.3%) were among 15-19 year olds. Infants were much more likely than older children with neurological diagnoses to have had meningitis (23.3% vs. 6.3%, p<.001).
Table 3.
Diagnoses, ICU care, and mortality among children with neurological diagnoses (n=103,140)
| Patients with diagnosis (%) |
ICU care (%) | Children’s hospital care (%) |
Hospital mortality (%) |
|
|---|---|---|---|---|
| Meningitis | 8.6% | 14.4% | 44.6 | 0.8% |
| Bacterial | 1.3% | 45.9% | 60.5 | 3.7% |
| Viral | 6.5% | 6.6% | 41.7 | 0.0% |
| Other | 0.9% | 25.2% | 62.5 | 2.2% |
|
| ||||
| Encephalitis, myelitis, or encephalomyelitis | 0.5% | 41.2% | 33.3 | 1.9% |
|
| ||||
| Abscess | 0.3% | 59.6% | 72.5 | 0.6% |
|
| ||||
| Encephalopathy | 5.1% | 48.9% | 87.0 | 9.7% |
|
| ||||
| CNS neoplasm | 5.4% | 35.8% | 90.9 | 2.2% |
|
| ||||
| Seizures without status epilepticus | 51.6% | 25.7% | 55.0 | 0.9% |
|
| ||||
| Status epilepticus | 2.4% | 63.4% | 61.2 | 1.8% |
|
| ||||
| Stroke | 1.8% | 64.1% | 69.3 | 10.6% |
| Ischemic stroke | 0.9% | 59.4% | 70.5 | 9.9% |
| Hemorrhagic stroke | 0.7% | 76.1% | 68.8 | 14.3% |
| Venous thrombosis | 0.2% | 48.1% | 73.8 | 2.2% |
|
| ||||
| Hydrocephalus | 6.5% | 43.3% | 74.9 | 2.2% |
|
| ||||
| Migraine | 4.7% | 9.9% | 43.9 | 0.1% |
|
| ||||
| Syncope | 3.7% | 19.7% | 42.0 | 0.1% |
|
| ||||
| Traumatic injury | 17.9% | 51.5% | 48.5 | 3.5% |
| TBI | 17.3% | 51.2% | 49.4 | 3.9% |
| Severe TBI | 0.7% | 85.1% | 54.5 | 29.1% |
| Spinal cord injury | 0.8% | 64.4% | 47.1 | 4.1% |
|
| ||||
| Other neurological diagnosis | 6.6% | 44.4% | 65.1 | 5.9% |
Percentages do not add up to 100% because some children had more than one neurological diagnosis
ICU = intensive care unit; CNS = central nervous system; TBI = traumatic brain injury
Almost a third (30.6%) of admissions with neurological diagnoses included ICU care, which is nearly 3 times greater than the rate of ICU care in hospitalizations for other conditions (10.6%). Rates of ICU care were highest among patients with severe TBI (85.1%), hemorrhagic stroke (76.1%), traumatic spinal cord injury (64.4%), and status epilepticus (63.4%) (Table 3). The most commonly treated neurological diagnosis in the ICU was seizure (12.6% of ICU admissions). ICU patients with neurological diagnoses had greater than twice the rate of mechanical ventilation than ICU patients with other diagnoses (23.8% vs. 9.9%, p<.001) (Table 1).
Nearly half (46.2%) of hospital deaths and more than half (53.6%) of deaths among children receiving ICU care were associated with neurological diagnoses. Children with neurological diagnoses had a hospital mortality rate of 1.7%, over 8 times that of children with other conditions (0.2%) (Table 1). Among ICU patients, mortality was also considerably higher among children with neurological diagnoses (4.8% vs. 1.5%, p<.001). Mortality was highest among children treated for severe TBI (29.1%), hemorrhagic stroke (14.3%), ischemic stroke (9.9%), and encephalopathy (9.7%) (Table 3). Mortality also varied by presence of an underlying condition. Among all admissions of children with neurological diagnoses, previously healthy children had similar mortality to those with underlying conditions (1.8% vs. 1.6%, p=.09). However, among more severely ill patients with neurological diagnoses, those admitted to an ICU, previously healthy children had a higher mortality than those with underlying conditions (5.6% vs. 3.7%, p<.001). Among children without neurological diagnoses, previously healthy children had lower mortality than those with underlying conditions (0.1% vs. 0.9%, p<.001).
Although the 77 children’s hospitals made up only 6.3% of hospitals in the 11 states, they provided care for 54.4% (n=56,123) of admissions with neurological diagnoses. In contrast, children’s hospitals provided care for only 40.0% of admissions without neurological diagnoses. Children’s hospitals admitted an average of 729 patients per year with neurological diagnoses, whereas general hospitals admitted only 43 per year. Compared to children with neurological diagnoses admitted to general hospitals, those admitted to children’s hospitals had higher rates of ICU admission (35.6% vs. 24.5%, p<.001), mechanical ventilation (9.6% vs. 7.1%, p<.001), underlying conditions (31.7% vs. 18.4%, p<.001), and death (2.0% vs. 1.4%, p<.001) (Table 4). Children with CNS neoplasms and encephalopathy were almost universally cared for at children’s hospitals (90.9% and 87.0% respectively). The majority of children with bacterial or other meningitis, abscess, seizures with or without status epilepticus, stroke, hydrocephalus, and severe TBI were cared for at children’s hospitals. Additionally, children with neurological diagnoses admitted to children’s hospitals had a longer median hospital LOS (3 [2, 6] vs. 2 [1,4] days p<.001) and median hospital costs ($5,290 [$2,790, $12,090] vs. $3,570 [$2,050, $7,310], p<.001).
Table 4.
Characteristics, outcomes, and resource use among children with neurological diagnoses admitted to children’s vs. general hospitals
| Children’s Hospitals 77 hospitals, 56,123 admissions |
General Hospitals 1096 hospitals, 47,020 admissions |
|
|---|---|---|
| Admissions per hospital, mean | 729 | 43 |
| Patient age, y | ||
| Mean ± SD | 8.1 ± 6.3 | 9.7 ± 6.9 |
| Median (IQR) | 7 (2,14) | 10 (2,17) |
|
| ||
| Underlying conditions (%) | 31.7 | 18.4 |
|
| ||
| Mechanical ventilation (%) | 9.6 | 7.1 |
|
| ||
| ICU Stay (%) | 35.6 | 24.5 |
|
| ||
| Hospital mortality (%) | 2.0 | 1.4 |
|
| ||
| LOS, days | ||
| Mean ± SD | 5.9 ± 11.9 | 5.0 ± 12.0 |
| Median (IQR) | 3 (2,6) | 2 (1,4) |
|
| ||
| Hospital cost, US dollars | ||
| Mean ± SD | $13,470 ± $30,770 | $8,960 ± $21,070 |
| Median (IQR) | $5,290 ($2,790, $12,090) | $3,570 ($2,050, $7,310) |
All differences between children’s and general hospitals were statistically significant (p<.001)
y = years; LOS = length of stay; SD = standard deviation; IQR = interquartile range
DISCUSSION
Children hospitalized with neurological diagnoses have a substantially greater use of ICU services, higher risk of death, and higher overall resource use than children hospitalized with other conditions. Even among ICU patients, children with neurological diagnoses account for a disproportionate number of children undergoing mechanical ventilation and deaths. The complexity of care for children neurological diagnoses is also reflected by the higher admission rate to children’s hospitals, longer lengths of stay, and higher hospital costs.
No prior studies have investigated the hospital or ICU incidence of neurological diagnoses collectively in a large population of patients. Our findings related to the incidence of pediatric stroke hospitalization are comparable to results of Lo et al., who studied a large, national dataset from 2003. They found that 3.7/100,000 children were hospitalized with ischemic stroke and 2.4/100,000 were hospitalized with hemorrhagic stroke (3, 16). A study of stroke using California state-wide data from 1991-2000 found mortality rates of 14.6% for ischemic stroke, 19.5% for intracerebral hemorrhage, and 16.6% for subarachnoid hemorrhage (4). While these rates are slightly higher than the rates that we found, they are not surprising given the different time period studied, and may also reflect differences between California and the 11 states that we studied. The TBI incidence rate that we found is similar to that published by the Centers for Disease Control (1), and our mortality results are similar to what Schenler et al. found in year 2000 (2).
While the literature lacks studies regarding the ICU admission rate among children with neurological diagnoses, Au et al. recently reported that 28.2% of ICU admissions at their center were for neurological conditions (17), similar to our finding that 26.1% of ICU patients had a neurological diagnosis. While their finding that brain injury was the proximate cause of death in 65.4% of ICU patients was higher than our finding of 53.7%, their facility is a tertiary, level 1 pediatric trauma center with a dedicated neurocritical care service, and they had access to more detailed clinical data than that available in our administrative dataset.
The lower prevalence of underlying conditions, mortality, and resource use suggest that children admitted to general hospitals, on average, are less severely ill or have less complicated hospital courses than those admitted to children’s hospitals. This is supported by our findings that the majority of children with less common and/or more severe neurological diagnoses were preferentially admitted to children’s hospitals. However, many children with neurological diagnoses still received ICU care and mechanical ventilation at hospitals caring for few such patients annually, some of whom died.
There are several limitations to this study related to the use of administrative data. First, although these data enable the conduct of broad, population-based analyses, they contain limited clinical information. We chose diagnoses with high likelihood of being active during hospital admission, but it is possible that some may have been chronic, stable conditions. For example, while a large majority (70%) of cases of neurological conditions were coded as the first or second discharge diagnosis, suggesting they were acute and active during the hospitalization, the degree to which that holds for diagnoses in other positions is less certain. As with all studies of discharge data, we also do not know the extent to which coding was influenced by payer reimbursement, leading to documentation bias. In addition, the states that we studied may not be fully representative of the entire US, which would affect our national estimates. In effort to minimize this, we adjusted for sex and age differences between the 11-states and the entire country. We also selected states from geographically diverse locations, with populations that collectively amounted to over a third of the pediatric US population. Our estimates slightly underestimate the national burden of neurological disease because children hospitalized in military and Indian Health Service medical centers were not included in our data set. Data were from 2005, and we do not know if the epidemiology has changed in subsequent years.
Conclusions
This study shows the high prevalence, mortality, and cost of treating neurological diagnoses among children. Our findings will provide a baseline for monitoring the hospitalizations, outcomes, and care patterns for children with neurological diagnoses as new therapies and health policies evolve.
Acknowledgments
Ms. Moreau received funding from the National Institutes of Health T-35 Start-Up Program (#5TL1RR024155).
Abbreviations used
- ICD-9
International Classification of Diseases, Ninth Revision, Clinical Modification
- ICU
intensive care unit
- PICU
pediatric intensive care unit
- TBI
traumatic brain injury
- LOS
length of stay
- NCHS
National Center for Health Statistics
APPENDIX.
| Diagnosis Category Non-Traumatic |
Specific Diagnosis | ICD-9-CM Code |
|---|---|---|
| Meningitis | Bacterial | 036.0, 098.82, 320.0/320.9 |
| Viral | 047.0/047.9, 49.0, 049.1, 053.0, 054.72, 072.1 |
|
| Other | 013.00/013.26, 091.81, 094.2, 112.83, 114.20, 115.01, 115.11, 115.91, 321.0/322.9 |
|
| Encephalitis, myelitis or encephalomyelitis |
Bacterial encephalitis or encephalomyelitis |
036.1 |
| Viral encephalitis or encephalomyelitis |
046.2, 046.3, 052.0, 054.3, 055.0, 056.01, 058.21, 058.29, 062.0/064 066.41, 072.2, 323.01, 323.02 |
|
| Other encephalitis or encephalomyelitis |
013.60/013.66, 323.2, 323.41, 323.51, 323.61, 323.61, 323.62, 323.71, 323.81, 323.9 |
|
| Myelitis | 341.20/341.22, 323.42, 323.52, 323.63, 323.72, 323.82 |
|
| Abscess | Brain | 013.30/013.36, 324.0, 324.9 |
| Spinal cord | 013.50/013.56, 324.1 | |
| Encephalopathy | Transmissible Spongiform | 046.11, 046.19, 046.71/046.79 |
| Hypoxic ischemic | 348.1 | |
| Metabolic & toxic | 333.72,348.31, 349.82 | |
| NOS | 348.30 348.39 | |
| Hypertensive | 437.2 | |
| CNS Neoplasm | Benign | 191.0/191.9,192.1/192.9, 194.3,198.3,198.4 |
| Malignant | 225.0, 225.2/225.8 | |
| Unspecified | 237.0/237.2, 237.5, 237.6 | |
| Stroke | Hemorrhagic | 433.01/434.91, 436 |
| Ischemic | 430, 431 | |
| CNS venous thrombosis | 437.6, 325 | |
| Seizure | Seizure | 345.00, 345.10, 345.40, 345.50, 345.60, 345.61, 345.80, 345.90, 780.39 |
| Intractable seizure | 345.01, 345.11, 345.41, 345.51, 345.81, 345.91 |
|
| Status epilepticus | 345.2, 345.3 | |
| Partial status epilepticus | 345.70, 345.71 | |
| Febrile convulsions | 780.31, 780.32 | |
| Hydrocephalus | Hydrocephalus | 331.3, 331.4, 331.5 |
| Migraine | Migraine | 346.00/346.93 |
| Syncope | Syncope | 780.2 |
| Other | Spinal cord TB | 013.40/013.46 |
| Other CNS TB | 013.80/013.96 | |
| Other CNS viral | 046.8, 046.9, 048, 049.8, 049.9, 056.00, 056.09, 066.42 |
|
| Other CNS bacterial | 090.40, 094.89, 094.9 | |
| Congenital syphilitic encephalitis |
090.41 | |
| Congenital syphilitic meningitis |
090.42 | |
| Hypoglycemic coma | 251.0 | |
| Type 1 Diabetic coma | 250.31, 250.33, 249.30, 249.31 | |
| Type 2 diabetic coma | 250.30, 250.32 | |
| Reyes Syndrome | 331.81 | |
| Myoclonus | 333.2 | |
| Other choreas | 333.5 | |
| Genetic torsion dystonia | 333.6 | |
| Other cerebellar ataxia | 334.3 | |
| Other spinal cord | 335.8/336.9 | |
| Demyelinating disorders | 340, 341.1, 341.8, 341.9 | |
| Pseudotumor cerebri | 348.2 | |
| Compression of brain | 348.4 | |
| Cerebral edema | 348.5 | |
| Nervous system complications from surgically implanted device |
349.1 | |
| Disorders of meninges, not elsewhere classified |
349.2 | |
| Unspecified disorders of nervous system |
349.90/349.93 | |
| Hemorrhage | 432.0/432.9 | |
| Ischemia | 435.0/435.9, 437.1 | |
| Cerebral aneurysm | 437.3 | |
| Cerebral arteritis | 437.4 | |
| Moyamoya disorder | 437.5 | |
| Coma NOS | 780.01 | |
| Aphasia | 784.3 | |
| Traumatic Brain | ||
| Injury | ||
| Anatomically severe head injury |
Open fracture of the vault of the skull with cerebral laceration/contusion |
800.60/800.69 |
| Open fracture of the vault of the skull with subarachnoid subdural and extradural hemorrhage |
800.70/800.79 | |
| Open fracture of the vault of the skull with subarachnoid subdural and extradural hemorrhage |
800.80/800.99 | |
| Open fracture of base of the skull with other/unspecified intracranial hemorrhage |
801.60/801.99 | |
| Other open skull fracture | 803.60/803.99 | |
| Open fractures involving skull or face with other bones |
804.60/804.99 | |
| Cortex (cerebral) contusion with open intracranial wound |
851.10/851.19 | |
| Cortex (cerebral) laceration with open intracranial wound |
851.30/851.39 | |
| Cerebellar or brain stem contusion with open intracranial wound |
851.50/851.59 | |
| Cerebellar or brain stem laceration with open intracranial wound |
851.70/851.79 | |
| Other and unspecified cerebral laceration and contusion with open intracranial wound |
851.90/851.99 | |
| Subarachnoid hemorrhage following injury with open intracranial wound |
852.10/852.19 | |
| Subdural hemorrhage following injury with open intracranial wound |
852.30/852.39 | |
| Extradural hemorrhage following injury with open intracranial wound |
852.50/852.59 | |
| Other and unspecified intracranial hemorrhage following injury with open intracranial wound |
853.10/853.19 | |
| Intracranial injury of other and unspecified nature with open intracranial wound |
854.10/854.19 | |
| Physiologically severe head injury |
Concussion with prolonged loss of consciousness |
850.3 850.4 |
| Traumatic Spinal Cord Injury |
Fracture of vertebral column with spinal cord injury |
806.00/806.9 |
| Spinal cord injury without evidence of spinal bone injury |
952.00/952.9 |
Footnotes
The other authors have no financial relationships relevant to this article to disclose. No authors have potential conflicts of interest.
This research was exempt from approval at the University of Pittsburgh.
Contributor’s Statement: The dataset was cleaned and coded by Ms. Moreau, Dr. Hartman, Mr. Linde-Zwirble, and Dr. Watson. Analysis and data interpretation was performed by Ms. Moreau and Dr. Watson. The manuscript was drafted by Ms. Moreau. The manuscript was revised by Drs. Fink, Angus, Bell, Hartman, and Watson. All authors approved of the final version.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Faul MXL, Wald MM, Coronado VG. Traumatic Brain Injury in the United States: Emergency Department Visits, Hospitalizations, and Deaths. Centers for Disease Control and Prevention; Atlanta, GA: 2010. [Google Scholar]
- 2.Schneier AJ, Shields BJ, Hostetler SG, et al. Incidence of pediatric traumatic brain injury and associated hospital resource utilization in the United States. Pediatrics. 2006;118(2):483–492. doi: 10.1542/peds.2005-2588. [DOI] [PubMed] [Google Scholar]
- 3.Lo W, Stephens J, Fernandez S. Pediatric stroke in the United States and the impact of risk factors. Journal of Child Neurology. 2009;24(2):194–203. doi: 10.1177/0883073808322665. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Fullerton HJ, Wu YW, Zhao S, et al. Risk of stroke in children: ethnic and gender disparities. Neurology. 2003;61(2):189–194. doi: 10.1212/01.wnl.0000078894.79866.95. [DOI] [PubMed] [Google Scholar]
- 5.Ramnarayan P, Craig F, Petros A, et al. Characteristics of deaths occurring in hospitalised children: changing trends. J Med Ethics. 2007;33(5):255–260. doi: 10.1136/jme.2005.015768. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Au AK, Carcillo JA, Clark RS, et al. Brain injuries and neurological system failure are the most common proximate causes of death in children admitted to a pediatric intensive care unit. Pediatr Crit Care Med. 2011;12(5):566–571. doi: 10.1097/PCC.0b013e3181fe3420. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Roach ES, Golomb MR, Adams R, et al. Management of stroke in infants and children: a scientific statement from a Special Writing Group of the American Heart Association Stroke Council and the Council on Cardiovascular Disease in the Young. Stroke. 2008;39(9):2644–2691. doi: 10.1161/STROKEAHA.108.189696. [DOI] [PubMed] [Google Scholar]
- 8.Kochanek PM, Tasker RC. Pediatric neurointensive care: 2008 update for the Rogers’ Textbook of Pediatric Intensive Care. Pediatr Crit Care Med. 2009;10(4):517–523. doi: 10.1097/PCC.0b013e3181ab88cd. [DOI] [PubMed] [Google Scholar]
- 9.Texas Hospital Inpatient Discharge Public Use Data File. Texas Department of State Health Services; Austin, Texas: 2005. [Google Scholar]
- 10.Public use file-PUF1 patient level data. Virginia Health Information; Richmond, VA: 2005. [Google Scholar]
- 11.HCUP State Inpatient Databases (SID) Healthcare Cost and Utilization Project (HCUP) Agency for Healthcare Research and Quality; Rockville, MD: [Accessed on April 14, 2013]. 2005. http://www.hcup-us.ahrq.gov/sidoverview.jsp. [Google Scholar]
- 12.United States Census Bureau [cited 2012 April 26];Methodology Summary for the Interim Population Projections for States by Age and Sex: 2004-2030. Available from: http://www.census.gov/population/www/projections/methodology.html.
- 13.World Health Organization . International classification of diseases, 9th revision, clinical modification. 5th ed PMIC; Los Angeles: 1998. [Google Scholar]
- 14.Trauma. ACoSCo. Resources for Optimal Care of the Injured Patient. American College of Surgeons; Chicago, IL: 1999. [PubMed] [Google Scholar]
- 15.Feudtner C, Hays RM, Haynes G, et al. Deaths attributed to pediatric complex chronic conditions: national trends and implications for supportive care services. Pediatrics. 2001;107(6):E99. doi: 10.1542/peds.107.6.e99. [DOI] [PubMed] [Google Scholar]
- 16.Perkins E, Stephens J, Xiang H, et al. The cost of pediatric stroke acute care in the United States. Stroke. 2009;40(8):2820–2827. doi: 10.1161/STROKEAHA.109.548156. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Au AK, Carcillo JA, Clark RS, et al. Brain injuries and neurological system failure are the most common proximate causes of death in children admitted to a pediatric intensive care unit. Pediatr Crit Care Med. 2010 doi: 10.1097/PCC.0b013e3181fe3420. [DOI] [PMC free article] [PubMed] [Google Scholar]