RESUMEN
BACKGROUND AND OBJECTIVES: Children with new tracheostomy and invasive mechanical ventilation (IMV) require transitional care involving caregiver education and nursing support. To better understand hospital resource use during this transition, our study aimed to: (1) define and characterize low-resource days (LRDs) for this population and (2) identify factors associated with LRD occurrence. METHODS: This retrospective cohort analysis included children ≤21 years with new tracheostomy and IMV dependence admitted to an ICU from 2017 to 2022 using the Pediatric Health Information System database. A LRD was defined as a post tracheostomy day that accrued nonroom charges <10% of each patient's accrued nonroom charges on postoperative day 1. Factors associated with LRDs were analyzed using negative binomial regression. RESULTS: Among 4048 children, median post tracheostomy stay was 69 days (interquartile range 34-127.5). LRDs were common: 38.6% and 16.4% experienced ≥1 and ≥7 LRDs, respectively. Younger age at tracheostomy (0-7 days rate ratio [RR] 2.42 [1.67-3.51]; 8-28 days RR 1.8 (1.2-2.69) versus 29-365 days; Asian race (RR 1.5 [1.04-2.16]); early tracheostomy (0-7 days RR 1.56 [1.2-2.04]), and longer post tracheostomy hospitalizations (31-60 days RR 1.85 [1.44-2.36]; 61-90 days RR 2.14 [1.58-2.91]; >90 days RR 2.21 [1.71-2.86]) were associated with more LRDs. CONCLUSIONS: Approximately 1 in 6 children experienced ≥7 LRDs. Younger age, early tracheostomy, Asian race, and longer hospital stays were associated with increased risk of LRDs. Understanding the postacute phase, including bed utilization, serves as an archetype to explore care models for children with IMV dependence.
RESUMEN
RATIONALE: More targeted management of severe acute pediatric asthma could improve clinical outcomes. OBJECTIVES: To identify distinct clinical phenotypes of severe acute pediatric asthma using variables obtained in the first 12 h of hospitalization. METHODS: We conducted a retrospective cohort study in a quaternary care children's hospital from 2014 to 2022. Encounters for children ages 2-18 years admitted to the hospital for asthma were included. We used consensus k means clustering with patient demographics, vital signs, diagnostics, and laboratory data obtained in the first 12 h of hospitalization. MEASUREMENTS AND MAIN RESULTS: The study population included 683 encounters divided into derivation (80%) and validation (20%) sets, and two distinct clusters were identified. Compared to Cluster 1 in the derivation set, Cluster 2 encounters (177 [32%]) were older (11 years [8; 14] vs. 5 years [3; 8]; p < .01) and more commonly males (63% vs. 53%; p = .03) of Black race (51% vs. 40%; p = .03) with non-Hispanic ethnicity (96% vs. 84%; p < .01). Cluster 2 encounters had smaller improvements in vital signs at 12-h including percent change in heart rate (-1.7 [-11.7; 12.7] vs. -7.8 [-18.5; 1.7]; p < .01), and respiratory rate (0.0 [-20.0; 22.2] vs. -11.4 [-27.3; 9.0]; p < .01). Encounters in Cluster 2 had lower percentages of neutrophils (70.0 [55.0; 83.0] vs. 85.0 [77.0; 90.0]; p < .01) and higher percentages of lymphocytes (17.0 [8.0; 32.0] vs. 9.0 [5.3; 14.0]; p < .01). Cluster 2 encounters had higher rates of invasive mechanical ventilation (23% vs. 5%; p < .01), longer hospital length of stay (4.5 [2.6; 8.8] vs. 2.9 [2.0; 4.3]; p < .01), and a higher mortality rate (7.3% vs. 0.0%; p < .01). The predicted cluster assignments in the validation set shared the same ratio (~2:1), and many of the same characteristics. CONCLUSIONS: We identified two clinical phenotypes of severe acute pediatric asthma which exhibited distinct clinical features and outcomes.
RESUMEN
OBJECTIVE: Pediatric poison exposures are a common reason for pediatric intensive care unit (PICU) -admission. The purpose of this study was to examine the exposure trends and patient outcomes in 2018-2019 compared with 2020-2021 amidst the COVID-19 pandemic. METHODS: This was a retrospective cohort study of patients 18 years of age or younger with a suspected toxicologic exposure from January 2018 to March 2021. The primary endpoint was rate of PICU admissions between the 2 cohorts. Secondary endpoints included medical outcome stratified by severity, PICU length of stay, and need for mechanical ventilation. RESULTS: Our study included a total of 340 patients with median age 14.5 (IQR, 11.9-16.1) years. There was no significant difference in age, sex, or race between the 2 cohorts. The percentage of patients admitted to the PICU for poison exposures was significantly higher in the COVID-19 cohort compared with the pre-COVID-19 cohort (8.4% vs 3.7%, p < 0.01). Severity of medical outcomes differed between the groups; the COVID-19 cohort had more extreme clinical presentations of no effect or death (p < 0.01). No significant difference was found among the remaining secondary outcomes. Classes of substances ingested were comparable with baseline poison center data. CONCLUSIONS: Poisoning-related PICU admissions occurred at more than twice the pre-pandemic rate. This may emphasize the effect of the COVID-19 pandemic on pediatric access and exposure to poisons.
RESUMEN
Objective: Long COVID, marked by persistent, recurring, or new symptoms post-COVID-19 infection, impacts children's well-being yet lacks a unified clinical definition. This study evaluates the performance of an empirically derived Long COVID case identification algorithm, or computable phenotype, with manual chart review in a pediatric sample. This approach aims to facilitate large-scale research efforts to understand this condition better. Methods: The algorithm, composed of diagnostic codes empirically associated with Long COVID, was applied to a cohort of pediatric patients with SARS-CoV-2 infection in the RECOVER PCORnet EHR database. The algorithm classified 31,781 patients with conclusive, probable, or possible Long COVID and 307,686 patients without evidence of Long COVID. A chart review was performed on a subset of patients (n=651) to determine the overlap between the two methods. Instances of discordance were reviewed to understand the reasons for differences. Results: The sample comprised 651 pediatric patients (339 females, M age = 10.10 years) across 16 hospital systems. Results showed moderate overlap between phenotype and chart review Long COVID identification (accuracy = 0.62, PPV = 0.49, NPV = 0.75); however, there were also numerous cases of disagreement. No notable differences were found when the analyses were stratified by age at infection or era of infection. Further examination of the discordant cases revealed that the most common cause of disagreement was the clinician reviewers' tendency to attribute Long COVID-like symptoms to prior medical conditions. The performance of the phenotype improved when prior medical conditions were considered (accuracy = 0.71, PPV = 0.65, NPV = 0.74). Conclusions: Although there was moderate overlap between the two methods, the discrepancies between the two sources are likely attributed to the lack of consensus on a Long COVID clinical definition. It is essential to consider the strengths and limitations of each method when developing Long COVID classification algorithms.
RESUMEN
OBJECTIVES: Identification of children with sepsis-associated multiple organ dysfunction syndrome (MODS) at risk for poor outcomes remains a challenge. We sought to the determine reproducibility of the data-driven "persistent hypoxemia, encephalopathy, and shock" (PHES) phenotype and determine its association with inflammatory and endothelial biomarkers, as well as biomarker-based pediatric risk strata. DESIGN: We retrained and validated a random forest classifier using organ dysfunction subscores in the 2012-2018 electronic health record (EHR) dataset used to derive the PHES phenotype. We used this classifier to assign phenotype membership in a test set consisting of prospectively (2003-2023) enrolled pediatric septic shock patients. We compared profiles of the PERSEVERE family of biomarkers among those with and without the PHES phenotype and determined the association with established biomarker-based mortality and MODS risk strata. SETTING: Twenty-five PICUs across the United States. PATIENTS: EHR data from 15,246 critically ill patients with sepsis-associated MODS split into derivation and validation sets and 1,270 pediatric septic shock patients in the test set of whom 615 had complete biomarker data. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: The area under the receiver operator characteristic curve of the modified classifier to predict PHES phenotype membership was 0.91 (95% CI, 0.90-0.92) in the EHR validation set. In the test set, PHES phenotype membership was associated with both increased adjusted odds of complicated course (adjusted odds ratio [aOR] 4.1; 95% CI, 3.2-5.4) and 28-day mortality (aOR of 4.8; 95% CI, 3.11-7.25) after controlling for age, severity of illness, and immunocompromised status. Patients belonging to the PHES phenotype were characterized by greater degree of systemic inflammation and endothelial activation, and were more likely to be stratified as high risk based on PERSEVERE biomarkers predictive of death and persistent MODS. CONCLUSIONS: The PHES trajectory-based phenotype is reproducible, independently associated with poor clinical outcomes, and overlapped with higher risk strata based on prospectively validated biomarker approaches.
Asunto(s)
Biomarcadores , Hipoxia , Fenotipo , Choque Séptico , Humanos , Biomarcadores/sangre , Femenino , Masculino , Niño , Preescolar , Lactante , Choque Séptico/sangre , Choque Séptico/mortalidad , Choque Séptico/diagnóstico , Hipoxia/diagnóstico , Hipoxia/sangre , Unidades de Cuidado Intensivo Pediátrico , Insuficiencia Multiorgánica/diagnóstico , Insuficiencia Multiorgánica/mortalidad , Insuficiencia Multiorgánica/sangre , Adolescente , Sepsis/diagnóstico , Sepsis/complicaciones , Sepsis/sangre , Sepsis/mortalidad , Reproducibilidad de los Resultados , Medición de Riesgo/métodos , Estudios Prospectivos , Encefalopatía Asociada a la Sepsis/sangre , Encefalopatía Asociada a la Sepsis/diagnóstico , Curva ROC , Puntuaciones en la Disfunción de ÓrganosRESUMEN
In September 2023, CDC's Advisory Committee on Immunization Practices recommended updated 2023-2024 (monovalent XBB.1.5) COVID-19 vaccination for all persons aged ≥6 months to prevent COVID-19, including severe disease. As with past COVID-19 vaccines, additional doses may be considered for persons with immunocompromising conditions, who are at higher risk for severe COVID-19 and might have decreased response to vaccination. In this analysis, vaccine effectiveness (VE) of an updated COVID-19 vaccine dose against COVID-19-associated hospitalization was evaluated during September 2023-February 2024 using data from the VISION VE network. Among adults aged ≥18 years with immunocompromising conditions, VE against COVID-19-associated hospitalization was 38% in the 7-59 days after receipt of an updated vaccine dose and 34% in the 60-119 days after receipt of an updated dose. Few persons (18%) in this high-risk study population had received updated COVID-19 vaccine. All persons aged ≥6 months should receive updated 2023-2024 COVID-19 vaccination; persons with immunocompromising conditions may get additional updated COVID-19 vaccine doses ≥2 months after the last recommended COVID-19 vaccine.
Asunto(s)
COVID-19 , Vacunas contra la Influenza , Gripe Humana , Adulto , Estados Unidos/epidemiología , Humanos , Adolescente , Gripe Humana/epidemiología , Vacunas contra la COVID-19 , COVID-19/epidemiología , COVID-19/prevención & control , Vacunación , HospitalizaciónRESUMEN
OBJECTIVES: To characterize the complex interplay between multiple clinical conditions in a time-to-event analysis framework using data from multiple hospitals, we developed two novel one-shot distributed algorithms for competing risk models (ODACoR). By applying our algorithms to the EHR data from eight national children's hospitals, we quantified the impacts of a wide range of risk factors on the risk of post-acute sequelae of SARS-COV-2 (PASC) among children and adolescents. MATERIALS AND METHODS: Our ODACoR algorithms are effectively executed due to their devised simplicity and communication efficiency. We evaluated our algorithms via extensive simulation studies as applications to quantification of the impacts of risk factors for PASC among children and adolescents using data from eight children's hospitals including the Children's Hospital of Philadelphia, Cincinnati Children's Hospital Medical Center, Children's Hospital of Colorado covering over 6.5 million pediatric patients. The accuracy of the estimation was assessed by comparing the results from our ODACoR algorithms with the estimators derived from the meta-analysis and the pooled data. RESULTS: The meta-analysis estimator showed a high relative bias (â¼40%) when the clinical condition is relatively rare (â¼0.5%), whereas ODACoR algorithms exhibited a substantially lower relative bias (â¼0.2%). The estimated effects from our ODACoR algorithms were identical on par with the estimates from the pooled data, suggesting the high reliability of our federated learning algorithms. In contrast, the meta-analysis estimate failed to identify risk factors such as age, gender, chronic conditions history, and obesity, compared to the pooled data. DISCUSSION: Our proposed ODACoR algorithms are communication-efficient, highly accurate, and suitable to characterize the complex interplay between multiple clinical conditions. CONCLUSION: Our study demonstrates that our ODACoR algorithms are communication-efficient and can be widely applicable for analyzing multiple clinical conditions in a time-to-event analysis framework.
Asunto(s)
Algoritmos , Hospitales , Adolescente , Niño , Humanos , Reproducibilidad de los Resultados , Simulación por Computador , Factores de RiesgoRESUMEN
BACKGROUND: Noninvasive respiratory support (NRS) for pediatric critical asthma includes CPAP; bi-level positive airway pressure (BPAP); and heated, humidified, high-flow nasal cannula (HFNC). We used the Virtual Pediatric System database to estimate NRS by prescribing rates for pediatric critical asthma and characterize patient clinical features and in-patient outcomes by the initial NRS device applied. METHODS: We performed a retrospective cohort study from 125 participating pediatric ICUs among children 2-17 years of age hospitalized for critical asthma and prescribed NRS from 2017 through 2021. The primary outcomes were NRS modality prescribing rates and trends. Secondary outcomes were descriptive and included demographics, comorbidities, severity of illness indices, and NRS failure rates (defined as escalation from the initial NRS modality to invasive ventilation, HFNC to BPAP or CPAP, or CPAP to BPAP). RESULTS: Of the 10,083 encounters studied, the initial NRS modalities prescribed varied widely by hospital center (HFNC: 69.7 ± 29.6%; BPAP: 27.2 ± 7.1%; CPAP: 3.1 ± 5.9%). The mean rates of HFNC use increased from 59.7% in 2017 to 71.9% in 2021 (+2.5%/y). In contrast, BPAP (-1.6%/y) and CPAP (-0.8%/y) utilization declined throughout the study period. Older children who were obese and with a higher Pediatric Risk of Mortality III-Probability of Mortality score were more frequently prescribed BPAP and CPAP compared with HFNC. Those children on HFNC experienced higher noninvasive respiratory support failure rates versus BPAP (7.3% vs 2.4%; P < .001) but a lower subsequent invasive ventilation rate versus BPAP (0.8% vs 2.4%; P < .001). CONCLUSIONS: In this multi-center cohort study, we observed that children with critical asthma are increasingly exposed to HFNC compared with BPAP and CPAP. Rates of HFNC failure were greater than those of BPAP failure, but a majority were transitioned to BPAP without subsequent invasive ventilation. The next steps include prospective trials, including practical end points such as patient comfort and optimal delivery of nebulized treatments to distinguish device superiority and suitable NRS utilization.
RESUMEN
Rationale: Over 20,000 children are hospitalized in the United States for asthma every year. Although initial treatment guidelines are well established, there is a lack of high-quality evidence regarding the optimal respiratory support devices for these patients.Objectives: The objective of this study was to evaluate institutional and temporal variability in the use of respiratory support modalities for pediatric critical asthma.Methods: We conducted a retrospective cohort study using data from the Virtual Pediatrics Systems database. Our study population included children older than 2 years old admitted to a VPS contributing pediatric intensive care unit from January 2012 to December 2021 with a primary diagnosis of asthma or status asthmaticus. We evaluated the percentage of encounters using a high-flow nasal cannula (HFNC), continuous positive airway pressure (CPAP), noninvasive bilevel positive pressure ventilation (NIV), and invasive mechanical ventilation (IMV) for all institutions, then divided institutions into quintiles based on the volume of patients. We created logistic regression models to determine the influence of institutional volume and year of admission on respiratory support modality use. We also conducted time-series analyses using Kendall's tau.Results: Our population included 77,115 patient encounters from 163 separate institutions. Institutional use of respiratory modalities had significant variation in HFNC (28.3%, interquartile range [IQR], 11.0-49.0%; P < 0.01), CPAP (1.4%; IQR, 0.3-4.3%; P < 0.01), NIV (8.6%; IQR, 3.5-16.1%; P < 0.01), and IMV (5.1%; IQR, 3.1-8.2%; P < 0.01). Increased institutional patient volume was associated with significantly increased use of NIV (odds ratio [OR], 1.33; 1.29-1.36; P < 0.01) and CPAP (OR, 1.20; 1.15-1.25; P < 0.01), and significantly decreased use of HFNC (OR, 0.80; 0.79-0.81; P < 0.01) and IMV (OR, 0.82; 0.79-0.86; P < 0.01). Time was also associated with a significant increase in the use of HFNC (11.0-52.3%; P < 0.01), CPAP (1.6-5.4%; P < 0.01), and NIV (3.7-21.2%; P < 0.01), whereas there was no significant change in IMV use (6.1-4.0%; P = 0.11).Conclusions: Higher-volume centers are using noninvasive positive pressure ventilation more frequently for pediatric critical asthma and lower frequencies of HFNC and IMV. Treatment with HFNC, CPAP, and NIV for this population is increasing in the last decade.
Asunto(s)
Asma , Ventilación no Invasiva , Insuficiencia Respiratoria , Humanos , Niño , Preescolar , Estudios Retrospectivos , Asma/terapia , Respiración Artificial , Hospitalización , Terapia por Inhalación de Oxígeno , Insuficiencia Respiratoria/terapiaRESUMEN
OBJECTIVES: To determine changes in pediatric oncology hospitalizations requiring intensive care over the period 2012-2021. DESIGN: Retrospective study of hospital admission. SETTING: Registry data from 36 children's hospitals in the U.S. Pediatric Health Information Systems database. PATIENTS: Children 18 years or younger admitted to any of 36 hospitals with an oncology diagnosis. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: There were a total of 55,827 unique patients accounted for 281,221 pediatric oncology hospitalizations over the 10-year period, and 16.6% of hospitalizations included admission to the PICU. Hospitalizations and PICU admissions steadily increased over this decade. Between 2012 and 2016, 15.1% of oncology hospitalizations were admitted to the PICU compared with 18.0% from 2017 to 2021 (difference 2.9% [95% CI, 2.6-3.2%] p ≤ 0.0001). Support with invasive mechanical ventilation also increased over time with 3.7% during 2012-2016 compared with 4.1% from 2017 to 2021 (difference 0.4% [95% CI, 0.2-0.5%] p ≤ 0.0001). Similar results were seen with cardiorespiratory life support using extracorporeal membrane oxygenation (difference 0.05% [95% CI, 0.02-0.07%] p = 0.0002), multiple vasoactive agent use (difference 0.3% [95% CI, 0.2-0.4%] p < 0.0001), central line placement (difference 5.3% [95% CI, 5.1-5.6%], p < 0.001), and arterial line placement (difference 0.4% [95% CI, 0.3-0.4%], p < 0.001). Year-on-year case fatality rate was unchanged over time (1.3%), but admission to the PICU during the second 5 years, compared with the first 5 years, was associated with lower odds of mortality (difference 0.7% [95% CI, 0.3-1.1%]) (odds ratio 0.82 [95% CI, 0.75-0.90%] p < 0.001). CONCLUSIONS: The percentage of pediatric oncology hospitalizations resulting in PICU admission has increased over the past 10 years. Despite the increasing use of PICU admission and markers of acuity, and on comparing 2017-2021 with 2012-2016, there are lower odds of mortality.
Asunto(s)
Sistemas de Información en Salud , Neoplasias , Niño , Humanos , Lactante , Estudios Retrospectivos , Neoplasias/terapia , Cuidados Críticos , Unidades de Cuidado Intensivo Pediátrico , Hospitales PediátricosRESUMEN
OBJECTIVES: To characterize the epidemiology of suicide and self-harm among adolescents admitted to PICUs during the first 2 years of the COVID-19 pandemic in the United States. DESIGN: Descriptive analysis of a large, multicenter, quality-controlled database (Virtual Pediatric Systems [VPS]), and of a national public health dataset (U.S. Centers for Disease Control and Prevention web-based Wide-ranging ONline Data for Epidemiology Research [CDC WONDER]). SETTING: The 69 PICUs participating in the VPS database that contributed data for the entire the study period, January 1, 2016, to December 31, 2021. PATIENTS: Adolescents older than 12 years to younger than 18 years old admitted to a participating PICU during the study period with a diagnosis involving self-harm or a suicide attempt (VPS sample), or adolescent suicide deaths over the same period (CDC WONDER sample). INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: We identified 10,239 suicide deaths and 7,692 PICU admissions for self-harm, including 5,414 admissions in the pre-pandemic period (Q1-2016 to Q1-2020) and 2,278 in the pandemic period (Q2-2020 to Q4-2021). Compared with the pre-pandemic period, there was no increase in the median (interquartile range) number of suicide deaths per quarter (429 [399-453] vs. 416 [390-482]) or PICU admissions for self-harm per quarter (315 [289-353] vs. 310 [286-387]) during the pandemic period, respectively. There was an increase in the ratio of self-harm PICU admissions to all-cause PICU admissions per quarter during the pandemic (1.98 [1.43-2.12]) compared with the pre-pandemic period per quarter (1.59 [1.46-1.74]). We also observed a significant decrease in all-cause PICU admissions per quarter early in the pandemic compared with the pre-pandemic period (16,026 [13,721-16,297] vs. 19,607 [18,371-20,581]). CONCLUSIONS: The number of suicide deaths and PICU admissions per quarter for self-harm remained relatively constant during the pandemic, while the number of all-cause PICU admissions per quarter decreased compared with the pre-pandemic period. The resultant higher ratio of self-harm admissions to all-cause PICU admissions may have contributed to the perception that more adolescents required critical care for mental health-related conditions early in the pandemic.
Asunto(s)
COVID-19 , Conducta Autodestructiva , Suicidio , Adolescente , Niño , Humanos , COVID-19/epidemiología , Unidades de Cuidado Intensivo Pediátrico , Estudios Multicéntricos como Asunto , Pandemias , Conducta Autodestructiva/epidemiología , Estados Unidos/epidemiología , Bases de Datos Factuales , Suicidio/estadística & datos numéricosRESUMEN
OBJECTIVES: Children with status asthmaticus refractory to first-line therapies of systemic corticosteroids and inhaled beta-agonists often receive additional treatments. Because there are no national guidelines on the use of asthma therapies in the PICU, we sought to evaluate institutional variability in the use of adjunctive asthma treatments and associations with length of stay (LOS) and PICU use. DESIGN: Multicenter retrospective cohort study. SETTING: Administrative data from the Pediatric Health Information Systems (PHIS) database. PATIENTS: All inpatients 2-18 years old were admitted to a PHIS hospital between 2013 and 2021 with a diagnostic code for asthma. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: This study included 213,506 inpatient encounters for asthma, of which 29,026 patient encounters included care in a PICU from 39 institutions. Among these PICU encounters, large variability was seen across institutions in both the number of adjunctive asthma therapies used per encounter (min: 0.6, median: 1.7, max: 2.5, p < 0.01) and types of adjunctive asthma therapies (aminophylline, ipratropium, magnesium, epinephrine, and terbutaline) used. The center-level median hospital LOS ranged from 1 (interquartile range [IQR]: 1, 3) to 4 (3, 6) days. Among all the 213,506 inpatient encounters for asthma, the range of asthma admissions that resulted in PICU admission varied between centers from 5.2% to 47.3%. The average number of adjunctive therapies used per institution was not significantly associated with hospital LOS ( p = 0.81) nor the percentage of encounters with PICU admission ( p = 0.47). CONCLUSIONS: Use of adjunctive therapies for status asthmaticus varies widely among large children's hospitals and was not associated with hospital LOS or the percentage of encounters with PICU admission. Wide variance presents an opportunity for standardizing care with evidence-based guidelines to optimize outcomes and decrease adverse treatment effects and hospital costs.
Asunto(s)
Asma , Estado Asmático , Niño , Humanos , Preescolar , Adolescente , Estudios Retrospectivos , Estado Asmático/terapia , Estado Asmático/diagnóstico , Asma/tratamiento farmacológico , Aminofilina , Terbutalina , Tiempo de Internación , Unidades de Cuidado Intensivo PediátricoRESUMEN
OBJECTIVE: Perform a scoping review of supervised machine learning in pediatric critical care to identify published applications, methodologies, and implementation frequency to inform best practices for the development, validation, and reporting of predictive models in pediatric critical care. DESIGN: Scoping review and expert opinion. SETTING: We queried CINAHL Plus with Full Text (EBSCO), Cochrane Library (Wiley), Embase (Elsevier), Ovid Medline, and PubMed for articles published between 2000 and 2022 related to machine learning concepts and pediatric critical illness. Articles were excluded if the majority of patients were adults or neonates, if unsupervised machine learning was the primary methodology, or if information related to the development, validation, and/or implementation of the model was not reported. Article selection and data extraction were performed using dual review in the Covidence tool, with discrepancies resolved by consensus. SUBJECTS: Articles reporting on the development, validation, or implementation of supervised machine learning models in the field of pediatric critical care medicine. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Of 5075 identified studies, 141 articles were included. Studies were primarily (57%) performed at a single site. The majority took place in the United States (70%). Most were retrospective observational cohort studies. More than three-quarters of the articles were published between 2018 and 2022. The most common algorithms included logistic regression and random forest. Predicted events were most commonly death, transfer to ICU, and sepsis. Only 14% of articles reported external validation, and only a single model was implemented at publication. Reporting of validation methods, performance assessments, and implementation varied widely. Follow-up with authors suggests that implementation remains uncommon after model publication. CONCLUSIONS: Publication of supervised machine learning models to address clinical challenges in pediatric critical care medicine has increased dramatically in the last 5 years. While these approaches have the potential to benefit children with critical illness, the literature demonstrates incomplete reporting, absence of external validation, and infrequent clinical implementation.
Asunto(s)
Enfermedad Crítica , Sepsis , Adulto , Recién Nacido , Humanos , Niño , Ciencia de los Datos , Estudios Retrospectivos , Cuidados Críticos , Sepsis/diagnóstico , Sepsis/terapia , Aprendizaje Automático SupervisadoRESUMEN
BACKGROUND: Asthma is the most common chronic disease of childhood, and has several social, environmental, and demographic factors potentially influential to its disease burden. This study sought to determine the influence of these factors on hospital admissions and readmissions for pediatric asthma. METHODS: This was a retrospective cohort study using data from the Indiana Network for Patient Care, a state-wide health information exchange in the United States. Study participants were children 2-18 years old admitted to the hospital with a respiratory diagnostic code between 2010 and 2021. Clinical variables were obtained from electronic health record data, and social and environmental determinants of health data were obtained from the Indiana Social Assets and Vulnerabilities Indicators using geocoding systems. Negative binomial models were used to examine community level social and environmental risk factors modifying the relationship between patient characteristics and the risk of asthma-related hospitalizations and 30-day readmissions. RESULTS: The study sample included 25,063 patients with an average follow-up of 9 (SD = 5) years. Of these, there were 17,816 asthma-related admissions. There were a total of 1,037 asthma-related 30-day readmissions, with an incidence rate of readmissions relative to total visits of 0.028 per person-year. A high social vulnerability index (SVI) was associated with an increased rate of hospital admissions (Proportion attributable ratio: 1.09, 95%CI (1.03,1.15), p < 0.05). No environmental determinants of health were significantly associated with hospitalization rate. CONCLUSION: High SVI was significantly associated with increased risk of total hospital admissions for pediatric asthma.
RESUMEN
Objective: This systematic review aims to identify social risk factors that influence pediatric asthma exacerbations. Methods: Cohort studies published between 2010 and 2020 were systematically searched on the OVID Medline, Embase, and PsycInfo databases. Using our established phased inclusion and exclusion criteria, studies that did not address a pediatric population, social risk factors, and asthma exacerbations were excluded. Out of a total of 707 initially retrieved articles, 3 prospective cohort and 6 retrospective cohort studies were included. Results: Upon analysis of our retrieved studies, two overarching domains of social determinants, as defined by Healthy People 2030, were identified as major risk factors for pediatric asthma exacerbations: Social/Community Context and Neighborhood/Built Environment. Social/Community factors including African American race and inadequate caregiver perceptions were associated with increased risk for asthma exacerbations. Patients in high-risk neighborhoods, defined by lower levels of education, housing, and employment, had higher rates of emergency department readmissions and extended duration of stay. Additionally, a synergistic interaction between the two domains was found such that patients with public or no health insurance and residence in high-risk neighborhoods were associated with excess hospital utilization attributable to pediatric asthma exacerbations. Conclusion: Social risk factors play a significant role in influencing the frequency and severity of pediatric asthma exacerbations.
RESUMEN
Background: Identifying phenotypes in sepsis patients may enable precision medicine approaches. However, the generalisability of these phenotypes to specific patient populations is unclear. Given that paediatric cancer patients with sepsis have different host response and pathogen profiles and higher mortality rates when compared to non-cancer patients, we determined whether unique, reproducible, and clinically-relevant sepsis phenotypes exist in this specific patient population. Methods: We studied patients with underlying malignancies admitted with sepsis to one of 25 paediatric intensive care units (PICUs) participating in two large, multi-centre, observational cohorts from the European SCOTER study (n = 383 patients; study period between January 1, 2018 and January 1, 2020) and the U.S. Novel Data-Driven Sepsis Phenotypes in Children study (n = 1898 patients; study period between January 1, 2012 and January 1, 2018). We independently used latent class analysis (LCA) in both cohorts to identify phenotypes using demographic, clinical, and laboratory data from the first 24 h of PICU admission. We then tested the association of the phenotypes with clinical outcomes in both cohorts. Findings: LCA identified two distinct phenotypes that were comparable across both cohorts. Phenotype 1 was characterised by lower serum bicarbonate and albumin, markedly increased lactate and hepatic, renal, and coagulation abnormalities when compared to phenotype 2. Patients with phenotype 1 had a higher 90-day mortality (European cohort 29.2% versus 13.4%, U.S. cohort 27.3% versus 11.4%, p < 0.001) and received more vasopressor and renal replacement therapy than patients with phenotype 2. After adjusting for severity of organ dysfunction, haematological cancer, prior stem cell transplantation and age, phenotype 1 was associated with an adjusted OR of death at 90-day of 1.9 (1.04-3.34) in the European cohort and 1.6 (1.2-2.2) in the U.S. cohort. Interpretation: We identified two clinically-relevant sepsis phenotypes in paediatric cancer patients that are reproducible across two international, multicentre cohorts with prognostic implications. These results may guide further research regarding therapeutic approaches for these specific phenotypes. Funding: Part of this study is funded by the Eunice Kennedy Shriver National Institute of Child Health and Human Development.
RESUMEN
As clinical understanding of pediatric Post-Acute Sequelae of SARS CoV-2 (PASC) develops, and hence the clinical definition evolves, it is desirable to have a method to reliably identify patients who are likely to have post-acute sequelae of SARS CoV-2 (PASC) in health systems data. In this study, we developed and validated a machine learning algorithm to classify which patients have PASC (distinguishing between Multisystem Inflammatory Syndrome in Children (MIS-C) and non-MIS-C variants) from a cohort of patients with positive SARS- CoV-2 test results in pediatric health systems within the PEDSnet EHR network. Patient features included in the model were selected from conditions, procedures, performance of diagnostic testing, and medications using a tree-based scan statistic approach. We used an XGboost model, with hyperparameters selected through cross-validated grid search, and model performance was assessed using 5-fold cross-validation. Model predictions and feature importance were evaluated using Shapley Additive exPlanation (SHAP) values. The model provides a tool for identifying patients with PASC and an approach to characterizing PASC using diagnosis, medication, laboratory, and procedure features in health systems data. Using appropriate threshold settings, the model can be used to identify PASC patients in health systems data at higher precision for inclusion in studies or at higher recall in screening for clinical trials, especially in settings where PASC diagnosis codes are used less frequently or less reliably. Analysis of how specific features contribute to the classification process may assist in gaining a better understanding of features that are associated with PASC diagnoses.
Asunto(s)
COVID-19 , Síndrome Post Agudo de COVID-19 , Niño , Humanos , COVID-19/diagnóstico , SARS-CoV-2 , Progresión de la Enfermedad , Aprendizaje Automático , FenotipoRESUMEN
INTRODUCTION: This study aimed to determine if a respiratory therapist (RT)-driven high flow nasal cannula (HFNC) protocol could decrease duration of HFNC use, pediatric intensive care unit (PICU) and hospital length of stay (LOS), and duration of continuous albuterol use in pediatric patients with critical asthma. METHODS: This was a quality improvement project performed at a quaternary academic PICU. Patients admitted to the PICU between 2 and 18 years of age with a diagnosis of asthma requiring continuous albuterol and HFNC were included. Implementation of an RT-driven HFNC protocol [Plan-Do-Study-Act (PDSA) 1] occurred in October 2017. Additional interventions included weaning continuous albuterol and HFNC simultaneously (PDSA 2; March 2019), adjusting HFNC wean rate (PDSA 3; July 2020), and a HFNC holiday (PDSA 4; October 2021). HFNC duration was the primary outcome. Secondary outcomes included LOS data and continuous albuterol duration. Noninvasive ventilation (NIV), invasive mechanical ventilation (IMV), and 7-day PICU and hospital readmission rates were used as balancing measures. RESULTS: A total of 410 patients were included. Patient demographics and adjunct therapy use did not differ among the groups. After PDSA 2, mean HFNC duration decreased (26.8-18.1 h). Mean PICU LOS decreased (41-31.8 h). Mean hospital LOS also decreased (86.5-68 h). These outcomes remained stable during PDSA 3 and 4. Continuous albuterol duration and NIV use were unchanged, while IMV use decreased. CONCLUSIONS: An RT-driven HFNC protocol led to decreased length of HFNC and PICU and hospital LOS for pediatric patients with critical asthma without an increase in adverse events.
RESUMEN
Objective: Identification of children with sepsis-associated multiple organ dysfunction syndrome (MODS) at risk for poor outcomes remains a challenge. Data-driven phenotyping approaches that leverage electronic health record (EHR) data hold promise given the widespread availability of EHRs. We sought to externally validate the data-driven 'persistent hypoxemia, encephalopathy, and shock' (PHES) phenotype and determine its association with inflammatory and endothelial biomarkers, as well as biomarker-based pediatric risk-strata. Design: We trained and validated a random forest classifier using organ dysfunction subscores in the EHR dataset used to derive the PHES phenotype. We used the classifier to assign phenotype membership in a test set consisting of prospectively enrolled pediatric septic shock patients. We compared biomarker profiles of those with and without the PHES phenotype and determined the association with established biomarker-based mortality and MODS risk-strata. Setting: 25 pediatric intensive care units (PICU) across the U.S. Patients: EHR data from 15,246 critically ill patients sepsis-associated MODS and 1,270 pediatric septic shock patients in the test cohort of whom 615 had biomarker data. Interventions: None. Measurements and Main Results: The area under the receiver operator characteristic curve (AUROC) of the new classifier to predict PHES phenotype membership was 0.91(95%CI, 0.90-0.92) in the EHR validation set. In the test set, patients with the PHES phenotype were independently associated with both increased odds of complicated course (adjusted odds ratio [aOR] of 4.1, 95%CI: 3.2-5.4) and 28-day mortality (aOR of 4.8, 95%CI: 3.11-7.25) after controlling for age, severity of illness, and immuno-compromised status. Patients belonging to the PHES phenotype were characterized by greater degree of systemic inflammation and endothelial activation, and overlapped with high risk-strata based on PERSEVERE biomarkers predictive of death and persistent MODS. Conclusions: The PHES trajectory-based phenotype is reproducible, independently associated with poor clinical outcomes, and overlap with higher risk-strata based on validated biomarker approaches.
RESUMEN
BACKGROUND: High flow nasal cannula (HFNC) is a respiratory device increasingly used to treat asthma. Recent mechanistic studies have shown that nebulized medications may have reduced delivery with HFNC, which may impair asthma treatment. This study evaluated the association between HFNC use for pediatric asthma and hospital length of stay (LOS). METHODS: This was a retrospective matched cohort study. Cases included patients aged 2-18 years hospitalized between January 2010 and December 2021 with asthma and received HFNC treatment. Controls were selected using logistic regression propensity score matching based on demographics, vital signs, medications, imaging, and social and environmental determinants of health. The primary outcome was hospital LOS. RESULTS: A total of 23,659 encounters met eligibility criteria, and of these 1766 cases included HFNC treatment with a suitable matched control. Cases were well-matched in demographics, social and environmental determinants of health, and clinical characteristics including use of adjunctive asthma therapies. The median hospital LOS for study cases was significantly higher at 87 h (interquartile range [IQR]: 61-145) compared to 66 h (IQR: 43-105) in the matched controls (p < 0.01). There was no significant difference in the rate of intubation and mechanical ventilation (8.9% vs. 7.6%, p = .18); however, the use of NIV was significantly higher in the cases than the control group (21.3% vs. 6.7%, p < .01). CONCLUSION: In this study of children hospitalized for asthma, HFNC use was associated with increased hospital LOS compared to matched controls. Further research using more granular data and additional relevant variables is needed to validate these findings.