Race, Ethnicity, and Intensive Care Utilization for Common Pediatric Diagnoses: U.S. Pediatric Health Information System 2019 Database Study.

OBJECTIVES: Racial and ethnic disparities in healthcare delivery for acutely ill children are pervasive in the United States; it is unknown whether differential critical care utilization exists. DESIGN: Retrospective study of the Pediatric Health Information System (PHIS) database. SETTING: Multicenter database of academic children's hospitals in the United States. PATIENTS: Children discharged from a PHIS hospital in 2019 with one of the top ten medical conditions where PICU utilization was present in greater than or equal to 5% of hospitalizations. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Race and ethnicity categories included Asian, Black, Hispanic, White, and other. Primary outcomes of interest were differences in rate of PICU admission, and for children requiring PICU care, total hospital length of stay (LOS). One-quarter (n = 44,200) of the 178,134 hospital discharges included a PICU admission. In adjusted models, Black children had greater adjusted odds ratio (aOR [95% CI]) of PICU admission in bronchiolitis (aOR, 1.08 [95% CI, 1.02-1.14]; p = 0.01), respiratory failure (aOR, 1.18 [95% CI, 1.10-1.28]; p < 0.001), seizure (aOR, 1.28 [95% CI, 1.08-1.51]; p = 0.004), and diabetic ketoacidosis (DKA) (aOR, 1.18 [95% CI, 1.05-1.32]; p = 0.006). Together, Hispanic, Asian, and other race children had greater aOR of PICU admission in five of the diagnostic categories, compared with White children. The geometric mean (± sd) hospital LOS ranged from 47.7 hours (± 2.1 hr) in croup to 206.6 hours (± 2.8 hr) in sepsis. After adjusting for demographics and illness severity, non-White children had longer LOS in respiratory failure, pneumonia, DKA, and sepsis. CONCLUSIONS: The need for critical care to treat acute illness in children may be inequitable. Additional studies are needed to understand and eradicate differences in PICU utilization based on race and ethnicity.

Child Opportunity Index and Pediatric Intensive Care Outcomes: A Multicenter Retrospective Study in the United States.

OBJECTIVES: To evaluate for associations between a child's neighborhood, as categorized by Child Opportunity Index (COI 2.0), and 1) PICU mortality, 2) severity of illness at PICU admission, and 3) PICU length of stay (LOS). DESIGN: Retrospective cohort study. SETTING: Fifteen PICUs in the United States. PATIENTS: Children younger than 18 years admitted from 2019 to 2020, excluding those after cardiac procedures. Nationally-normed COI category (very low, low, moderate, high, very high) was determined for each admission by census tract, and clinical features were obtained from the Virtual Pediatric Systems LLC (Los Angeles, CA) data from each site. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Among 33,901 index PICU admissions during the time period, median patient age was 4.9 years and PICU mortality was 2.1%. There was a higher percentage of admissions from the very low COI category (27.3%) than other COI categories (17.2-19.5%, p < 0.0001). Patient admissions from the high and very high COI categories had a lower median Pediatric Index of Mortality 3 risk of mortality (0.70) than those from the very low, low, and moderate COI groups (0.71) ( p < 0.001). PICU mortality was lowest in the very high (1.7%) and high (1.9%) COI groups and highest in the moderate group (2.5%), followed by very low (2.3%) and low (2.2%) ( p = 0.001 across categories). Median PICU LOS was between 1.37 and 1.50 days in all COI categories. Multivariable regression revealed adjusted odds of PICU mortality of 1.30 (95% CI, 0.94-1.79; p = 0.11) for children from a very low versus very high COI neighborhood, with an odds ratio [OR] of 0.996 (95% CI, 0.993-1.00; p = 0.05) for mortality for COI as an ordinal value from 0 to 100. Children without insurance coverage had an OR for mortality of 3.58 (95% CI, 2.46-5.20; p < 0.0001) as compared with those with commercial insurance. CONCLUSIONS: Children admitted to a cohort of U.S. PICUs were often from very low COI neighborhoods. Children from very high COI neighborhoods had the lowest risk of mortality and observed mortality; however, odds of mortality were not statistically different by COI category in a multivariable model. Children without insurance coverage had significantly higher odds of PICU mortality regardless of neighborhood.

Measuring Social Health Following Pediatric Critical Illness: A Scoping Review and Conceptual Framework.

OBJECTIVE: Social health is an important component of recovery following critical illness as modeled in the pediatric Post-Intensive Care Syndrome framework. We conducted a scoping review of studies measuring social outcomes (measurable components of social health) following pediatric critical illness and propose a conceptual framework of the social outcomes measured in these studies. DATA SOURCES: PubMed, EMBASE, PsycINFO, CINAHL, and the Cochrane Registry. STUDY SELECTION: We identified studies evaluating social outcomes in pediatric intensive care unit (PICU) survivors or their families from 1970-2017 as part of a broader scoping review of outcomes after pediatric critical illness. DATA EXTRACTION: We identified articles by dual review and dual-extracted study characteristics, instruments, and instrument validation and administration information. For instruments used in studies evaluating a social outcome, we collected instrument content and described it using qualitative methods adapted to a scoping review. DATA SYNTHESIS: Of 407 articles identified in the scoping review, 223 (55%) evaluated a social outcome. The majority were conducted in North America and the United Kingdom, with wide variation in methodology and population. Among these studies, 38 unique instruments were used to evaluate a social outcome. Specific social outcomes measured included individual (independence, attachment, empathy, social behaviors, social cognition, and social interest), environmental (community perceptions and environment), and network (activities and relationships) characteristics, together with school and family outcomes. While many instruments assessed more than one social outcome, no instrument evaluated all areas of social outcome. CONCLUSIONS: The full range of social outcomes reported following pediatric critical illness were not captured by any single instrument. The lack of a comprehensive instrument focused on social outcomes may contribute to under-appreciation of the importance of social outcomes and their under-representation in PICU outcomes research. A more comprehensive evaluation of social outcomes will improve understanding of overall recovery following pediatric critical illness.

Association of Interventions With Outcomes in Children At-Risk for Pediatric Acute Respiratory Distress Syndrome: A Pediatric Acute Respiratory Distress Syndrome Incidence and Epidemiology Study.

OBJECTIVES: Describe the frequency with which transfusion and medications that modulate lung injury are administered to children meeting at-risk for pediatric acute respiratory distress syndrome (ARF-PARDS) criteria and evaluate for associations of transfusion, fluid balance, nutrition, and medications with unfavorable clinical outcomes. DESIGN: Secondary analysis of the Pediatric Acute Respiratory Distress Syndrome Incidence and Epidemiology study, a prospective point prevalence study. All enrolled ARF-PARDS patients were included unless they developed subsequent pediatric acute respiratory distress syndrome (PARDS) within 24 hours of PICU admission or PICU length of stay was less than 24 hours. Univariate and multivariable analyses were used to identify associations between therapies given during the first 2 calendar days after ARF-PARDS diagnosis and subsequent PARDS diagnosis (primary outcome), 28-day PICU-free days (PFDs), and 28-day ventilator-free days (VFDs). SETTING: Thirty-seven international PICUs. PATIENTS: Two hundred sixty-seven children meeting Pediatric Acute Lung Injury Consensus Conference ARF-PARDS criteria. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: During the first 2 days after meeting ARF-PARDS criteria, 55% of subjects received beta-agonists, 42% received corticosteroids, 28% received diuretics, and 9% were transfused. Subsequent PARDS (15%) was associated with platelet transfusion (n = 11; adjusted odds ratio: 4.75 [95% CI 1.03-21.92]) and diuretics (n = 74; 2.55 [1.19-5.46]) in multivariable analyses that adjusted for comorbidities, PARDS risk factor, initial oxygen saturation by pulse oximetry:Fio2 ratio, and initial type of ventilation. Beta-agonists were associated with lower adjusted odds of subsequent PARDS (0.43 [0.19-0.98]). Platelets and diuretics were also associated with fewer PFDs and fewer VFDs in the multivariable models, and TPN was associated with fewer PFDs. Corticosteroids, net fluid balance, and volume of enteral feeding were not associated with the primary or secondary outcomes. CONCLUSIONS: There is an independent association between platelet transfusion, diuretic administration, and unfavorable outcomes in children at risk for PARDS, although this may be related to treatment bias and unmeasured confounders. Nevertheless, prospective evaluation of the role of these management strategies on outcomes in children with ARF-PARDS is needed.

Hospital Charges Associated With Critical Bronchiolitis From 2009 to 2019.

OBJECTIVES: To evaluate the contribution of PICU care to increasing hospital charges for patients with bronchiolitis over a 10-year study period. DESIGN: In this retrospective multicenter study, changes in annual hospital charges (adjusted for inflation) were analyzed using linear regression for subjects admitted to the PICU with invasive mechanical ventilation (PICU + IMV) and without IMV (PICU - IMV), and for children not requiring PICU care. SETTING: Free-standing children's hospitals contributing to the Pediatric Health Information System (PHIS) database. SUBJECTS: Children less than 2 years with bronchiolitis discharged from a PHIS hospital between July 2009 and June 2019. Subjects were categorized as high risk if they were born prematurely or had a chronic complex condition. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: PICU patients were 26.5% of the 283,006 included subjects but accrued 66% of the total $14.83 billion in charges. Annual charges increased from $1.01 billion in 2009-2010 to $2.07 billion in 2018-2019, and PICU patients accounted for 83% of this increase. PICU + IMV patients were 22% of all PICU patients and accrued 64% of all PICU charges, but PICU - IMV patients without a high-risk condition had the highest relative increase in annual charges, increasing from $76.7 million in 2009-2010 to $377.9 million in 2018-2019 (374% increase, ptrend < 0.001). CONCLUSIONS: In a multicenter cohort study of children hospitalized with bronchiolitis, PICU patients, especially low-risk children without the need for IMV, were the highest driver of increased hospital charges over a 10-year study period.

Derivation and Validation of the Critical Bronchiolitis Score for the PICU.

OBJECTIVES: To derive and internally validate a bronchiolitis-specific illness severity score (the Critical Bronchiolitis Score) that out-performs mortality-based illness severity scores (e.g., Pediatric Risk of Mortality) in measuring expected duration of respiratory support and PICU length of stay for critically ill children with bronchiolitis. DESIGN: Retrospective database study using the Virtual Pediatric Systems (VPS, LLC; Los Angeles, CA) database. SETTING: One-hundred twenty-eight North-American PICUs. PATIENTS: Fourteen-thousand four-hundred seven children less than 2 years old admitted to a contributing PICU with primary diagnosis of bronchiolitis and use of ICU-level respiratory support (defined as high-flow nasal cannula, noninvasive ventilation, invasive mechanical ventilation, or negative pressure ventilation) at 12 hours after PICU admission. INTERVENTIONS: Patient-level variables available at 12 hours from PICU admission, duration of ICU-level respiratory support, and PICU length of stay data were extracted for analysis. After randomly dividing the cohort into derivation and validation groups, patient-level variables that were significantly associated with the study outcomes were selected in a stepwise backward fashion for inclusion in the final score. Score performance in the validation cohort was assessed using root mean squared error and mean absolute error, and performance was compared with that of existing PICU illness severity scores. MEASUREMENTS AND MAIN RESULTS: Twelve commonly available patient-level variables were included in the Critical Bronchiolitis Score. Outcomes calculated with the score were similar to actual outcomes in the validation cohort. The Critical Bronchiolitis Score demonstrated a statistically significantly stronger association with duration of ICU-level respiratory support and PICU length of stay than mortality-based scores as measured by root mean squared error and mean absolute error. CONCLUSIONS: The Critical Bronchiolitis Score performed better than PICU mortality-based scores in measuring expected duration of ICU-level respiratory support and ICU length of stay. This score may have utility to enrich interventional trials and adjust for illness severity in observational studies in this very common PICU condition.

Use of Negative Pressure Ventilation in Pediatric Critical Care: Experience in 56 PICUs in the Virtual Pediatric Systems Database (2009-2019).

OBJECTIVES: Negative pressure ventilation may be more physiologic than positive pressure ventilation, but data describing negative pressure ventilation use in the PICU are limited. We aimed to describe the epidemiology and outcomes of PICU patients receiving negative pressure ventilation. DESIGN: Descriptive analysis of a large, quality-controlled multicenter database. SETTING: Fifty-six PICUs in the Virtual Pediatric Systems database who reported use of negative pressure ventilation. PATIENTS: Children admitted to a participating PICU between 2009 and 2019 who received negative pressure ventilation. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Among 788 subjects, 71% were less than 2 years old, and 45% had underlying health conditions. Two concurrent aspiration events were the only adverse events reported. After excluding one over-represented center, the 3 years with the most negative pressure ventilation usage were 2017-2019 (all > 25 cases/yr and ≥ 13 centers reporting usage). Among those 187 children, the most common primary diagnoses were bronchiolitis and cardiac disease (both 15.5%), 24.1% required endotracheal intubation after negative pressure ventilation, and 9.1% died. CONCLUSIONS: Negative pressure ventilation is being used in many PICUs, most commonly for pulmonary infections or cardiac disease, in children with high rates of subsequent intubation and mortality and with few documented adverse events. Use at individual centers is rare but increasing, suggesting need for prospective collaboration to better evaluate the role of negative pressure ventilation in the PICU.

Prevalence of Reintubation Within 24 Hours of Extubation in Bronchiolitis: Retrospective Cohort Study Using the Virtual Pediatric Systems Database.

OBJECTIVES: High-flow nasal cannula and noninvasive positive pressure ventilation are used to support children following liberation from invasive mechanical ventilation. Evidence comparing extubation failure rates between patients randomized to high-flow nasal cannula and noninvasive positive pressure ventilation is available for adult and neonatal patients; however, similar pediatric trials are lacking. In this study, we employed a quality controlled, multicenter PICU database to test the hypothesis that high-flow nasal cannula is associated with higher prevalence of reintubation within 24 hours among patients with bronchiolitis. DESIGN: Secondary analysis of a prior study utilizing the Virtual Pediatric Systems database. SETTING: One-hundred twenty-four participating PICUs. PATIENTS: Children less than 24 months old with a primary diagnosis of bronchiolitis who were admitted to one of 124 PICUs between January 2009 and September 2015 and received invasive mechanical ventilation. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Among 759 patients, median age was 2.4 months (1.3-5.4 mo), 41.2% were female, 39.7% had greater than or equal to 1 comorbid condition, and 43.7% were Caucasian. Median PICU length of stay was 8.7 days (interquartile range, 5.8-13.7 d) and survival to PICU discharge was 100%. Median duration of intubation was 5.5 days (3.4-9.0 d) prior to initial extubation. High-flow nasal cannula was used following extubation in most (656 [86.5%]) analyzed subjects. The overall prevalence of reintubation within 24 hours was 5.9% (45 children). Extubation to noninvasive positive pressure ventilation was associated with greater prevalence of reintubation than extubation to high-flow nasal cannula (11.7% vs 5.0%; p = 0.016) and, in an a posteriori model that included Pediatric Index of Mortality 2 score and comorbidities, was associated with increased odds of reintubation (odds ratio, 2.43; 1.11-5.34; p = 0.027). CONCLUSIONS: In this secondary analysis of a multicenter database of children with bronchiolitis, extubation to high-flow nasal cannula was associated with a lower prevalence of reintubation within 24 hours compared with noninvasive positive pressure ventilation in both unmatched and propensity-matched analysis. Prospective trials are needed to determine if post-extubation support modality can mitigate the risk of extubation failure.

Outcomes of Children With Critical Bronchiolitis Meeting at Risk for Pediatric Acute Respiratory Distress Syndrome Criteria.

OBJECTIVES: New definitions of pediatric acute respiratory distress syndrome include criteria to identify a subset of children "at risk for pediatric acute respiratory distress syndrome." We hypothesized that, among PICU patients with bronchiolitis not immediately requiring invasive mechanical ventilation, those meeting at risk for pediatric acute respiratory distress syndrome criteria would have worse clinical outcomes, including higher rates of pediatric acute respiratory distress syndrome development. DESIGN: Single-center, retrospective chart review. SETTING: Mixed medical-surgical PICU within a tertiary academic children's hospital. PATIENTS: Children 24 months old or younger admitted to the PICU with a primary diagnosis of bronchiolitis from September 2013 to April 2014. Children intubated before PICU arrival were excluded. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Collected data included demographics, respiratory support, oxygen saturation, and chest radiograph interpretation by staff radiologist. Oxygen flow (calculated as FIO2 × flow rate [L/min]) was calculated when oxygen saturation was 88-97%. The median age of 115 subjects was 5 months (2-11 mo). Median PICU length of stay was 2.8 days (1.5-4.8 d), and median hospital length of stay was 5 days (3-10 d). The criteria for at risk for pediatric acute respiratory distress syndrome was met in 47 of 115 subjects (40.9%). Children who were at risk for pediatric acute respiratory distress syndrome were more likely to develop pediatric acute respiratory distress syndrome (15/47 [31.9%] vs 1/68 [1.5%]; p < 0.001), had longer PICU length of stay (4.6 d [2.8-10.2 d] vs 1.9 d [1.0-3.1 d]; p < 0.001) and hospital length of stay (8 d [5-16 d] vs 4 d [2-6 d]; p < 0.001), and increased need for invasive mechanical ventilation (16/47 [34.0%] vs 2/68 [2.9%]; p < 0.001), compared with those children who did not meet at risk for pediatric acute respiratory distress syndrome criteria. CONCLUSIONS: Our data suggest that the recent definition of at risk for pediatric acute respiratory distress syndrome can successfully identify children with critical bronchiolitis who have relatively unfavorable clinical courses.

Outcomes of Children With Bronchiolitis Treated With High-Flow Nasal Cannula or Noninvasive Positive Pressure Ventilation.

OBJECTIVES: Initial respiratory support with noninvasive positive pressure ventilation or high-flow nasal cannula may prevent the need for invasive mechanical ventilation in PICU patients with bronchiolitis. However, it is not clear whether the initial choice of respiratory support modality influences the need for subsequent invasive mechanical ventilation. The purpose of this study is to compare the rate of subsequent invasive mechanical ventilation after initial support with noninvasive positive pressure ventilation or high-flow nasal cannula in children with bronchiolitis. DESIGN: Analysis of the Virtual Pediatric Systems database. SETTING: Ninety-two participating PICUs. PATIENTS: Children less than 2 years old admitted to a participating PICU between 2009 and 2015 with a diagnosis of bronchiolitis who were prescribed high-flow nasal cannula or noninvasive positive pressure ventilation as the initial respiratory treatment modality. INTERVENTIONS: None. Subsequent receipt of invasive mechanical ventilation was the primary outcome. MEASUREMENTS AND MAIN RESULTS: We identified 6,496 subjects with a median age 3.9 months (1.7-9.5 mo). Most (59.7%) were male, and 23.4% had an identified comorbidity. After initial support with noninvasive positive pressure ventilation or high-flow nasal cannula, 12.3% of patients subsequently received invasive mechanical ventilation. Invasive mechanical ventilation was more common in patients initially supported with noninvasive positive pressure ventilation compared with high-flow nasal cannula (20.1% vs 11.0%: p < 0.001). In a multivariate logistic regression model that adjusted for age, weight, race, viral etiology, presence of a comorbid diagnosis, and Pediatric Index of Mortality score, initial support with noninvasive positive pressure ventilation was associated with a higher odds of subsequent invasive mechanical ventilation compared with high-flow nasal cannula (odds ratio, 1.53; 95% CI, 1.24-1.88). CONCLUSIONS: In this large, multicenter database study of infants with acute bronchiolitis that received initial respiratory support with high-flow nasal cannula or noninvasive positive pressure ventilation, noninvasive positive pressure ventilation use was associated with higher rates of invasive mechanical ventilation, even after adjusting for demographics, comorbid condition, and severity of illness. A large, prospective, multicenter trial is needed to confirm these findings.

Neurologic and Functional Morbidity in Critically Ill Children With Bronchiolitis.

OBJECTIVES: Neurologic and functional morbidity occurs in ~30% of PICU survivors, and young children may be at particular risk. Bronchiolitis is a common indication for PICU admission among children less than 2 years old. Two single-center studies suggest that greater than 10-25% of critical bronchiolitis survivors have neurologic and functional morbidity but those estimates are 20 years old. We aimed to estimate the burden of neurologic and functional morbidity among more recent bronchiolitis patients using two large, multicenter databases. DESIGN: Analysis of the Pediatric Health Information System and the Virtual Pediatric databases. SETTING: Forty-eight U.S. children's hospitals (Pediatric Health Information System) and 40 international (mostly United States) children's hospitals (Virtual Pediatric Systems). PATIENTS: Previously healthy PICU patients less than 2 years old admitted with bronchiolitis between 2009 and 2015 who survived and did not require extracorporeal membrane oxygenation or cardiopulmonary resuscitation. INTERVENTIONS: None. Neurologic and functional morbidity was defined as a Pediatric Overall Performance Category greater than 1 at PICU discharge (Virtual Pediatric Systems subjects), or a subsequent hospital encounter involving developmental delay, feeding tubes, MRI of the brain, neurologist evaluation, or rehabilitation services (Pediatric Health Information System subjects). MEASUREMENTS AND MAIN RESULTS: Among 3,751 Virtual Pediatric Systems subjects and 9,516 Pediatric Health Information System subjects, ~20% of patients received mechanical ventilation. Evidence of neurologic and functional morbidity was present at PICU discharge in 707 Virtual Pediatric Systems subjects (18.6%) and more chronically in 1,104 Pediatric Health Information System subjects (11.6%). In both cohorts, neurologic and functional morbidity was more common in subjects receiving mechanical ventilation (27.5% vs 16.5% in Virtual Pediatric Systems; 14.5% vs 11.1% in Pediatric Health Information System; both p < 0.001). In multivariate models also including demographics, use of mechanical ventilation was the only variable that was associated with increased neurologic and functional morbidity in both cohorts. CONCLUSIONS: In two large, multicenter databases, neurologic and functional morbidity was common among previously healthy children admitted to the PICU with bronchiolitis. Prospective studies are needed to measure neurologic and functional outcomes using more precise metrics. Identification of modifiable risk factors may subsequently lead to improved outcomes from this common PICU condition.

Understanding Disparities in the Pediatric ICU: A Scoping Review.

BACKGROUND AND OBJECTIVES: Health disparities are pervasive in pediatrics. We aimed to describe disparities among patients who are likely to be cared for in the PICU and delineate how sociodemographic data are collected and categorized. METHODS: Using MEDLINE as a data source, we identified studies which included an objective to assess sociodemographic disparities among PICU patients in the United States. We created a review rubric, which included methods of sociodemographic data collection and analysis, outcome and exposure variables assessed, and study findings. Two authors reviewed every study. We used the National Institute on Minority Health and Health Disparities Research Framework to organize outcome and exposure variables. RESULTS: The 136 studies included used variable methods of sociodemographic data collection and analysis. A total of 30 of 124 studies (24%) assessing racial disparities used self- or parent-identified race. More than half of the studies (52%) dichotomized race as white and "nonwhite" or "other" in some analyses. Socioeconomic status (SES) indicators also varied; only insurance status was used in a majority of studies (72%) evaluating SES. Consistent, although not uniform, disadvantages existed for racial minority populations and patients with indicators of lower SES. The authors of only 1 study evaluated an intervention intended to mitigate health disparities. Requiring a stated objective to evaluate disparities aimed to increase the methodologic rigor of included studies but excluded some available literature. CONCLUSIONS: Variable, flawed methodologies diminish our understanding of disparities in the PICU. Meaningfully understanding and addressing health inequity requires refining how we collect, analyze, and interpret relevant data.

Nutritional Support in Children Meeting the At-Risk for Pediatric Acute Respiratory Distress Syndrome Criteria.

Pediatric acute respiratory distress syndrome (PARDS) is a prevalent condition in the PICU with a high morbidity and mortality, but effective preventative strategies are lacking. OBJECTIVES: To examine associations between early enteral nutrition (EN) and PICU outcomes in a cohort of children meeting the 2015 Pediatric Acute Lung Injury Consensus Conference "at-risk" for pediatric acute respiratory distress syndrome (ARF-PARDS) criteria. DESIGN SETTING AND PARTICIPANTS: This was a single-center, electronic health record-based retrospective chart review. We included children less than or equal to 18 years-old admitted to our mixed medical-surgical PICU from January 2017 to December 2018 who met ARF-PARDS criteria within 48 hours of admission. Children were categorized as receiving "early" EN if feeds were initiated within 48 hours of admission. All others were categorized as "delayed" EN. MAIN OUTCOMES AND MEASURES: Extracted data included demographics, illness characteristics including primary diagnosis and Pediatric Risk of Mortality (PRISM) III score, respiratory support and oxygenation indices, nutritional data, and PICU length of stay (LOS). The primary outcome of interest was subsequent diagnosis of PARDS. RESULTS: Of 201 included children, 152 (75.6%) received early EN. The most common admission diagnoses were pneumonia, bronchiolitis, and influenza. Overall, 21.4% (n = 43) of children developed PARDS. Children receiving early EN had a subsequent diagnosis of PARDS less often then children receiving delayed EN (15.1% vs 40.8%; p < 0.001), an association that persisted after adjusting for patient demographics and illness characteristics, including PRISM III and diagnosis (adjusted odds ratio, 0.24; 95% CI, 0.10-0.58; p = 0.002). Early EN was also associated with a shorter PICU LOS in univariate analysis (2.2 d [interquartile range, 1.5-3.4 d] vs 4.2 d [2.7-8.9 d]; p < 0.001). CONCLUSIONS AND RELEVANCE: In this single-center, retrospective cohort study, compared with children with ARF-PARDS who received late EN, those who received early EN demonstrated a reduced odds of subsequent diagnosis of PARDS, and an unadjusted reduction in PICU LOS when compared with delayed EN. Prospective studies should be designed to confirm these findings.

Child Opportunity Index and Hospital Utilization in Children With Traumatic Brain Injury Admitted to the PICU.

The need to understand how Community-based disparities impact morbidity and mortality in pediatric critical illness, such as traumatic brain injury. Test the hypothesis that ZIP code-based disparities in hospital utilization, including length of stay (LOS) and hospital costs, exist in a cohort of children with traumatic brain injury (TBI) admitted to a PICU using the Child Opportunity Index (COI). DESIGN: Multicenter retrospective cohort study. SETTING: Pediatric Health Information System (PHIS) database. PATIENTS: Children 0-18 years old admitted to a PHIS hospital with a diagnosis of TBI from January 2016 to December 2020 requiring PICU care. To identify the most severely injured children, a study-specific definition of "Complicated TBI" was created based on radiology, pharmacy, and procedure codes. INTERVENTIONS: None. Main Outcomes and Measures: Using nationally normed ZIP code-level COI data, patients were categorized into COI quintiles. A low COI ZIP code has low childhood opportunity based on weighted indicators within educational, health and environmental, and social and economic domains. Population-averaged generalized estimating equation (GEE) models, adjusted for patient and clinical characteristics examined the association between COI and study outcomes, including hospital LOS and accrued hospital costs. The median age of this cohort of 8,055 children was 58 months (interquartile range [IQR], 8-145 mo). There were differences in patient demographics and rates of Complicated TBI between COI levels. The median hospital LOS was 3.0 days (IQR, 2.0-6.0 d) and in population-averaged GEE models, children living in very low COI ZIP codes were expected to have a hospital LOS 10.2% (95% CI, 4.1-16.8%; p = 0.0142) longer than children living in very high COI ZIP codes. For the 11% of children with a Complicated TBI, the relationship between COI and LOS was lost in multivariable models. COI level was not predictive of accrued hospital costs in this study. CONCLUSIONS: Children with TBI requiring PICU care living in low-opportunity ZIP codes have higher injury severity and longer hospital LOS compared with children living in higher-opportunity ZIP codes. Additional studies are needed to understand why these differences exist.

Screening for Social Determinants of Health in the Pediatric Intensive Care Unit: Recommendations for Clinicians.

Social determinants of health (SDoH) play a significant role in the health and well-being of children in the United States. Disparities in the risk and outcomes of critical illness have been extensively documented but are yet to be fully explored through the lens of SDoH. In this review, we provide justification for routine SDoH screening as a critical first step toward understanding the causes of, and effectively addressing health disparities affecting critically ill children. Second, we summarize important aspects of SDoH screening that need to be considered before implementing this practice in the pediatric critical care setting.

Machine learning models to predict and benchmark PICU length of stay with application to children with critical bronchiolitis.

OBJECTIVE: To create models for prediction and benchmarking of pediatric intensive care unit (PICU) length of stay (LOS) for patients with critical bronchiolitis. HYPOTHESIS: We hypothesize that machine learning models applied to an administrative database will be able to accurately predict and benchmark the PICU LOS for critical bronchiolitis. DESIGN: Retrospective cohort study. PATIENTS: All patients less than 24-month-old admitted to the PICU with a diagnosis of bronchiolitis in the Pediatric Health Information Systems (PHIS) Database from 2016 to 2019. METHODOLOGY: Two random forest models were developed to predict the PICU LOS. Model 1 was developed for benchmarking using all data available in the PHIS database for the hospitalization. Model 2 was developed for prediction using only data available on hospital admission. Models were evaluated using R2 values, mean standard error (MSE), and the observed to expected ratio (O/E), which is the total observed LOS divided by the total predicted LOS from the model. RESULTS: The models were trained on 13,838 patients admitted from 2016 to 2018 and validated on 5254 patients admitted in 2019. While Model 1 had superior R2 (0.51 vs. 0.10) and (MSE) (0.21 vs. 0.37) values compared to Model 2, the O/E ratios were similar (1.18 vs. 1.20). Institutional median O/E (LOS) ratio was 1.01 (IQR 0.90-1.09) with wide variability present between institutions. CONCLUSIONS: Machine learning models developed using an administrative database were able to predict and benchmark the length of PICU stay for patients with critical bronchiolitis.

Human Trafficking ICD-10 Code Utilization in Pediatric Tertiary Care Centers Within the United States.

BACKGROUND: Human trafficking is a global public health issue that affects pediatric patients widely. The International Labor Organization estimates children comprise approximately 25% of the identified trafficked persons globally, with domestic estimates including over 2000 children a year. Trafficked children experience a broad range of health consequences leading to interface with healthcare systems during their exploitation. In June 2018, International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) released diagnostic codes for human trafficking. OBJECTIVE: To use a large, multicenter database of US pediatric hospitalizations to describe the utilization of the ICD-10-CM codes related to child trafficking, as well as the demographic and clinical characteristics of these children. METHODS: This study was descriptive in nature. Encounters using data from the Pediatric Health Information System database (PHIS) with ICD-10-CM codes indicating trafficking from June 1, 2018 to March 1st, 2020 were included in the study cohort, with data collection continuing for 30 days after first hospital encounter, until March 31st, 2020. Patients 19 years old and younger were included. Condition-specific prevalence as well as demographic and clinical characteristics for patient encounters were analyzed. Study subjects were followed for 30 days after first hospital encounter to describe healthcare utilization patterns. RESULTS: During the study period, 0.005% (n = 293) of patient encounters in the PHIS database were identified as trafficked children. The children of our cohort were mostly female (90%), non-Hispanic Black (38%), and had public insurance (59%). Nearly two-thirds of patients (n = 190) had a documented mental health disorder at the initial encounter, with 32.1% classified as the principal diagnosis. Our cohort had a 30-day hospital inpatient, overnight observation, or emergency department readmission rate of 16% (n = 48). DISCUSSION: Our study demonstrates a low utilization of human trafficking ICD-10-CM codes in academic children's health centers, with code usage predominantly assigned to Non-Hispanic Black teenage girls. As comparison, in 2019 the National Human Trafficking Hotline identified 2,582 trafficked US children in a single year. These results suggest widespread under-recognition of child trafficking in health care settings, including the intensive care unit, in addition to racial and socioeconomic disparities amongst trafficked children.