The Comparative Effectiveness of Perioperative Antibiotic Regimens to Prevent Surgical Site Infections in Pediatric Liver Transplant Recipients.

INTRODUCTION: Surgical site infections (SSIs) are a common complication in liver transplant(LT) recipients. Lack of pediatric prophylaxis guidelines results in variation in preventative antibiotic regimens. METHODS: We performed a retrospective observational study of LT recipients under 18 years using a merged dataset that included data from PHIS and UNOS between 2006 and 2017. The exposure was defined as the antibiotic(s) received within 24 hours of LT; with 6 categories, ranging from narrow (category 1: cefazolin), to broad). The primary outcome was presence or absence of SSI in the index admission. Mixed-effects logistic regression compared the effectiveness of each category relative to category 1 in preventing SSI. RESULTS: Of the 2586 LT, 284 (11%) met SSI criteria. SSI rate was higher (16.2%) in the younger sub-cohort compared to older (8.6%), necessitating a stratified analysis. Antibiotics from category 5 were most commonly used. In the younger sub-cohort, the adjusted risk was increased in all categories compared to the reference, most notably in category 3 (OR 2.58; 0.69-9.59) and category 6 (OR 2.76; 0.66-11.56). In the older sub-cohort, estimated ORs were also increased for each category, most notably in category 4 (2.49; 0.99-6.27). None of the ORs suggested benefit from broader-spectrum prophylaxis. Our E value assessment suggests it's unlikely there is unmeasured confounding by indication to the degree necessary to revert ORs to protective. CONCLUSION: There was wide variation in antibiotic prophylaxis. Adjusted analyses did not reveal a protective benefit of broader-spectrum prophylaxis in either sub-cohort, suggesting that narrower regimens may be adequate.

Association of the social disorganization index with time to first septic shock event in children with acute myeloid leukemia.

BACKGROUND: Pediatric acute myeloid leukemia (AML) chemotherapy increases the risk of life-threatening complications, including septic shock (SS). An area-based measure of social determinants of health, the social disorganization index (SDI), was hypothesized to be associated with SS and SS-associated death (SS-death). METHODS: Children treated for de novo AML on two Children's Oncology Group trials at institutions contributing to the Pediatric Health Information System (PHIS) database were included. The SDI was calculated via residential zip code data from the US Census Bureau. SS was identified via PHIS resource utilization codes. SS-death was defined as death within 2 weeks of an antecedent SS event. Patients were followed from 7 days after the start of chemotherapy until the first of end of front-line therapy, death, relapse, or removal from study. Multivariable-adjusted Cox regressions estimated hazard ratios (HRs) comparing time to first SS by SDI group. RESULTS: The assembled cohort included 700 patients, with 207 (29.6%) sustaining at least one SS event. There were 233 (33%) in the SDI-5 group (highest disorganization). Adjusted time to incident SS did not statistically significantly differ by SDI (reference, SDI-1; SDI-2: HR, 0.84 [95% confidence interval (CI), 0.51-1.41]; SDI-3: HR, 0.70 [95% CI, 0.42-1.16]; SDI-4: HR, 0.97 [95% CI, 0.61-1.53]; SDI-5: HR, 0.72 [95% CI, 0.45-1.14]). Nine patients (4.4%) with SS experienced SS-death; seven of these patients (78%) were in SDI-4 or SDI-5. CONCLUSIONS: In a large, nationally representative cohort of trial-enrolled pediatric patients with AML, there was no significant association between the SDI and time to SS.

Applying machine learning to identify pediatric patients with newly diagnosed acute lymphoblastic leukemia using administrative data.

Case identification in administrative databases is challenging as diagnosis codes alone are not adequate for case ascertainment. We utilized machine learning (ML) to efficiently identify pediatric patients with newly diagnosed acute lymphoblastic leukemia. We tested nine ML models and validated the best model internally and externally. The optimal model had 97% positive predictive value (PPV) and 99% sensitivity in internal validation; 94% PPV and 82% sensitivity in external validation. Our ML model identified a large cohort of 21,044 patients, demonstrating an efficient approach for cohort assembly and enhancing the usability of administrative data.

The effectiveness and safety of antibody induction immunosuppression in a large cohort of United States pediatric liver transplant recipients.

Data on the potential benefits and risks of induction therapy in pediatric liver transplantation (LT) are limited. This was a retrospective cohort study of 2748 pediatric LT recipients at 26 children's hospitals between January 1, 2006 to May 31, 2017 using data from the pediatric health information system linked to the United Network for Organ Sharing database. The induction regimen was obtained from the pediatric health information system day-by-day pharmacy resource utilization. Cox proportional hazards evaluated the association of induction regimen (none/corticosteroid-only, nondepleting, and depleting) on patient and graft survival. Additional outcomes, including opportunistic infections and posttransplant lymphoproliferative disorder, were studied using multivariable logistic regression. Overall, 64.9% received none/corticosteroid-only induction, whereas 28.1% received nondepleting, 8.3% received depleting, and 2.5% other antibody regimens. Differences in patient characteristics were small, but center practices were heterogeneous. Compared with none/corticosteroid-only induction, nondepleting induction was associated with reduced acute rejection (odd ratio [OR], 0.53; P <.001) but with the increased posttransplant lymphoproliferative disorder (OR, 1.75; P =.021). Depleting induction was associated with improved graft survival (hazard ratio [HR], 0.64; P =.028) but with increased noncytomegalovirus opportunistic infections (OR, 1.46; P =.046). Depleting induction is underused yet may offer long-term benefits in this large multicenter cohort. Greater consensus guidance in this aspect of pediatric LT care is warranted.

Leveraging machine learning to identify acute myeloid leukemia patients and their chemotherapy regimens in an administrative database.

BACKGROUND: Administrative datasets are useful for identifying rare disease cohorts such as pediatric acute myeloid leukemia (AML). Previously, cohorts were assembled using labor-intensive, manual reviews of patients' longitudinal chemotherapy data. METHODS: We utilized a two-step machine learning (ML) method to (i) identify pediatric patients with newly diagnosed AML, and (ii) among the identified AML patients, their chemotherapy courses, in an administrative/billing database. Using 2558 patients previously manually reviewed, multiple ML algorithms were derived from 75% of the study sample, and the selected model was tested in the remaining hold-out sample. The selected model was also applied to assemble a new pediatric AML cohort and further assessed in an external validation, using a standalone cohort established by manual chart abstraction. RESULTS: For patient identification, the selected Support Vector Machine model yielded a sensitivity of 0.97 and a positive predictive value (PPV) of 0.97 in the hold-out test sample. For course-specific chemotherapy regimen and start date identification, the selected Random Forest model yielded overall PPV greater than or equal to 0.88 and sensitivity greater than or equal to 0.86 across all courses in the test sample. When applied to new cohort assembly, ML identified 3016 AML patients with 10,588 treatment courses. In the external validation subset, PPV was greater than or equal to 0.75 and sensitivity was greater than or equal to 0.82 for patient identification, and PPV was greater than or equal to 0.93 and sensitivity was greater than or equal to 0.94 for regimen identifications. CONCLUSION: A carefully designed ML model can accurately identify pediatric AML patients and their chemotherapy courses from administrative databases. This approach may be generalizable to other diseases and databases.

Variation in hospital costs and resource utilisation after congenital heart surgery.

BACKGROUND: Children undergoing cardiac surgery have overall improving survival, though they consume substantial resources. Nationwide inpatient cost estimates and costs at longitudinal follow-up are lacking. METHODS: Retrospective cohort study of children <19 years of age admitted to Pediatric Health Information System administrative database with an International Classification of Diseases diagnosis code undergoing cardiac surgery. Patients were grouped into neonates (≤30 days of age), infants (31-365 days of age), and children (>1 year) at index procedure. Primary and secondary outcomes included hospital stay and hospital costs at index surgical admission and 1- and 5-year follow-up. RESULTS: Of the 99,670 cohort patients, neonates comprised 27% and had the highest total hospital costs, though daily hospital costs were lower. Mortality declined (5.6% in 2004 versus 2.5% in 2015, p < 0.0001) while inpatient costs rose (5% increase/year, p < 0.0001). Neonates had greater index diagnosis complexity, greater inpatient costs, required the greatest ICU resources, pharmacotherapy, and respiratory therapy. We found no relationship between hospital surgical volume, mortality, and hospital costs. Neonates had higher cumulative hospital costs at 1- and 5-year follow-up compared to infants and children. CONCLUSIONS: Inpatient hospital costs rose during the study period, driven primarily by longer stay. Neonates had greater complexity index diagnosis, required greater hospital resources, and have higher hospital costs at 1 and 5 years compared to older children. Surgical volume and in-hospital mortality were not associated with costs. Further analyses comprising merged clinical and administrative data are necessary to identify longer stay and cost drivers after paediatric cardiac surgery.

Late health outcomes after dexrazoxane treatment: A report from the Children's Oncology Group.

BACKGROUND: The objective of this study was to examine long-term outcomes among children newly diagnosed with cancer who were treated in dexrazoxane-containing clinical trials. METHODS: P9404 (acute lymphoblastic leukemia/lymphoma [ALL]), P9425 and P9426 (Hodgkin lymphoma), P9754 (osteosarcoma), and Dana-Farber Cancer Institute 95-01 (ALL) enrolled 1308 patients between 1996 and 2001: 1066 were randomized (1:1) to doxorubicin with or without dexrazoxane, and 242 (from P9754) were nonrandomly assigned to receive dexrazoxane. Trial data were linked with the National Death Index, the Organ Procurement and Transplantation Network, the Pediatric Health Information System (PHIS), and Medicaid. Osteosarcoma survivors from the Childhood Cancer Survivor Study (CCSS; n = 495; no dexrazoxane) served as comparators in subanalyses. Follow-up events were assessed with cumulative incidence, Cox regression, and Fine-Gray methods. RESULTS: In randomized trials (cumulative prescribed doxorubicin dose, 100-360 mg/m2 ; median follow-up, 18.6 years), dexrazoxane was not associated with relapse (hazard ratio [HR], 0.84; 95% confidence interval [CI], 0.63-1.13), second cancers (HR, 1.19; 95% CI, 0.62-2.30), all-cause mortality (HR, 1.07; 95% CI, 0.78-1.47), or cardiovascular mortality (HR, 1.45; 95% CI, 0.41-5.16). Among P9754 patients (all exposed to dexrazoxane; cumulative doxorubicin, 450-600 mg/m2 ; median follow-up, 16.6-18.4 years), no cardiovascular deaths or heart transplantation occurred. The 20-year heart transplantation rate among CCSS osteosarcoma survivors (mean doxorubicin, 377 ± 145 mg/m2 ) was 1.6% (vs 0% in P9754; P = .13). Among randomized patients, serious cardiovascular outcomes (cardiomyopathy, ischemic heart disease, and stroke) ascertained by PHIS/Medicaid occurred less commonly with dexrazoxane (5.6%) than without it (17.6%; P = .02), although cardiomyopathy rates alone did not differ (4.4% vs 8.1%; P = .35). CONCLUSIONS: Dexrazoxane did not appear to adversely affect long-term mortality, event-free survival, or second cancer risk.

Center Variability in Acute Rejection and Biliary Complications After Pediatric Liver Transplantation.

Transplant center performance and practice variation for pediatric post-liver transplantation (LT) outcomes other than survival are understudied. This was a retrospective cohort study of pediatric LT recipients who received transplants between January 1, 2006, and May 31, 2017, using United Network for Organ Sharing (UNOS) data that were merged with the Pediatric Health Information System database. Center effects for the acute rejection rate at 1 year after LT (AR1) using UNOS coding and the biliary complication rate at 1 year after LT (BC1) using inpatient billing claims data were estimated by center-specific rescaled odds ratios that accounted for potential differences in recipient and donor characteristics. There were 2216 pediatric LT recipients at 24 freestanding children's hospitals in the United States during the study period. The median unadjusted center rate of AR1 was 36.92% (interquartile range [IQR], 22.36%-44.52%), whereas that of BC1 was 32.29% (IQR, 26.14%-40.44%). Accounting for recipient case mix and donor factors, 5/24 centers performed better than expected with regard to AR1, whereas 3/24 centers performed worse than expected. There was less heterogeneity across the center effects for BC1 than for AR1. There was no relationship observed between the center effects for AR1 or BC1 and center volume. Beyond recipient and allograft factors, differences in transplant center management are an important driver of center AR1 performance, and less so of BC1 performance. Further research is needed to identify the sources of variability so as to implement the most effective solutions to broadly enhance outcomes for pediatric LT recipients.

Center Variation in Indication and Short-Term Outcomes after Pediatric Heart Transplantation: Analysis of a Merged United Network for Organ Sharing - Pediatric Health Information System Cohort.

The relationship between center-specific variation in indication for pediatric heart transplantation and short-term outcomes after heart transplantation is not well described. We used merged patient- and hospital-level data from the United Network for Organ Sharing and the Pediatric Health Information Systems to analyze outcomes according to transplant indication for a cohort of children (≤ 21 years old) who underwent heart transplantation between 2004 and 2015. Outcomes included 30-day mortality, transplant hospital admission mortality, and hospital length of stay, with multivariable adjustment performed according to patient and center characteristics. The merged cohort reflected 2169 heart transplants at 20 U.S. centers. The median number of transplants annually at each center was 11.6, but ranged from 3.5 to 22.6 transplants/year. Congenital heart disease was the indication in the plurality of cases (49.2%), with cardiomyopathy (46%) and myocarditis (4.8%) accounting for the remainder. There was significant center-to-center variability in congenital heart disease as the principal indication, ranging from 15% to 66% (P < 0.0001). After adjustment, neither center volume nor proportion of indications for transplantation were associated with 30-day or transplant hospital admission mortality. In this large, merged pediatric cohort, variation was observed at center level in annual transplant volume and prevalence of indications for heart transplantation. Despite this variability, center volume and proportion of indications represented at a given center did not appear to impact short-term outcomes.

Presentation acuity, induction mortality, and resource utilization in infants with acute leukemia.

BACKGROUND: Treatment of infants with acute leukemia remains challenging, especially for acute lymphocytic leukemia (ALL). Infants have shown markedly higher rates of induction mortality compared with noninfants. There are limited data on presentation acuity and supportive care utilization in this age group. METHODS: In retrospective analyses of patients treated for new onset ALL or acute myeloid leukemia (AML) at pediatric hospitals contributing to the Pediatric Health Information System, we compared presentation acuity, induction mortality, and resource utilization in infants relative to noninfants less than 10 years at diagnosis. RESULTS: Analyses included 10 359 children with ALL (405 infants, 9954 noninfants) and 871 AML (189 infants, 682 noninfants). Infants were more likely to present with multisystem organ failure compared to noninfants for both ALL (12% and 1%, PR = 10.8, 95% CI: 7.4, 15.7) and AML (6% vs. 3%; PR = 2.0, 95% CI: 1.0, 3.7). Infants with ALL had higher induction mortality compared to noninfants, even after accounting for differences in anthracycline exposure and presentation acuity (2.7% vs. 0.5%, HR = 2.1, 95% CI: 1.0, 4.8). Conversely, infants and noninfants with AML had similar rates of induction mortality (3.2% vs. 2.1%, HR = 1.2, 95% CI: 0.3, 3.9), which were comparable to rates among infants with ALL. Infants with ALL and AML had greater requirements for blood products, diuretics, supplemental oxygen, and ventilation during induction relative to noninfants. CONCLUSIONS: Infants with leukemia present with higher acuity compared with noninfants. Induction mortality and supportive care requirements for infants with ALL were similar to all children with AML, and significantly higher than those for noninfants with ALL.

Identifying relapses and stem cell transplants in pediatric acute lymphoblastic leukemia using administrative data: Capturing national outcomes irrespective of trial enrollment.

INTRODUCTION: Our objectives were to design and validate methods to identify relapse and hematopoietic stem cell transplantation (HSCT) in children with acute lymphoblastic leukemia (ALL) using administrative data representing hospitalizations at US pediatric institutions. METHODS: We developed daily billing and ICD-9 code definitions to identify relapses and HSCTs within a cohort of children with newly diagnosed ALL between January 1, 2004, and December 31, 2013, previously assembled from the Pediatric Health Information System (PHIS) database. Chart review for children with ALL at the Children's Hospital of Philadelphia (CHOP) and Texas Children's Hospital (TCH) was performed to establish relapse and HSCT gold standards for sensitivity and positive predictive value (PPV) calculations. We estimated incidences of relapse and HSCT in the PHIS ALL cohort. RESULTS: We identified 362 CHOP and 314 TCH ALL patients in PHIS and established true positives by chart review. Sensitivity and PPV for identifying both relapse and HSCT in PHIS were > 90% at both hospitals. Five-year relapse incidence in the 10 150-patient PHIS cohort was 10.3% (95% CI 9.8%-10.9%) with 7.1% (6.6%-7.6%) of children underwent HSCTs. Patients in higher-risk demographic groups had higher relapse and HSCT rates. Our analysis also identified differences in incidences of relapse and HSCT by race, ethnicity, and insurance status. CONCLUSIONS: Administrative data can be used to identify relapse and HSCT accurately in children with ALL whether they occur on- or off-therapy, in contrast with published approaches. This method has wide potential applicability for estimating these incidences in pediatric ALL, including patients not enrolled on clinical trials.

Mortality, Resource Utilization, and Inpatient Costs Vary Among Pediatric Heart Transplant Indications: A Merged Data Set Analysis From the United Network for Organ Sharing and Pediatric Health Information Systems Databases.

BACKGROUND: Merging United Network for Organ Sharing (UNOS) and Pediatric Health Information Systems databases has enabled a more granular analysis of pediatric heart transplant outcomes and resource utilization. We evaluated whether transplant indication at time of transplantation was associated with mortality, resource utilization, and inpatient costs during the first year after transplantation. METHODS AND RESULTS: We analyzed transplant outcomes and resource utilization from 2004 to 2015. Patients were categorized as congenital (CHD), myocarditis, or cardiomyopathy based on UNOS-defined primary indication. CHD complexity subgroup analyses (single-ventricle, complex, and simple biventricular CHD) were also performed. Of 2251 transplants (49% CHD, 5% myocarditis, 46% cardiomyopathy), CHD recipients were younger (2 [IQR 0-10], 6 [IQR 0-12], and 7 [IQR 1-14] years, respectively; P < .001) and less likely to have a ventricular assist device (VAD) at transplantation (3%, 27%, and 13%, respectively; P < .001). Patients with single-ventricle CHD had the longest time on the waitlist and were least likely to receive a VAD before transplantation. After adjusting for patient-level factors, transplant recipients with single-ventricle CHD had the greatest mortality during transplantation admission and within 1 year (odds ratio [OR] 11.8 [95% confidence interval (CI) 5.9-23.6] and OR 6.0 [95% CI 3.6-10.2], respectively, vs cardiomyopathy). Mortality was similar between patients with myocarditis and cardiomyopathy. Post-transplantation length of stay (LOS) was longer in transplant recipients with CHD than myocarditis or cardiomyopathy (25 [interquartile range [IQR] 15-45] vs 21 [IQR 12-35] vs 16 [IQR 12-25] days; P < .001), related in part to longer duration of intensive care unit-level care (ICU LOS 8 [IQR 4-20] vs 6 [IQR 4-13] vs 5 [IQR 3-8] days; P < .001). Similarly, patients with CHD had higher median post-transplantation costs than myocarditis or cardiomyopathy ($415K [IQR $201K-503K] vs $354K [IQR $179K-390K] vs $284K [IQR $145K-319K]; P < .001) that persisted after adjusting for patient-level factors (adjusted cost ratio 1.4 [95% CI 1.4-1.5], CHD vs cardiomyopathy) and was primarily driven by longer LOS. More than 50% were readmitted during the first year after transplantation, although readmission rates were similar across transplant indications (P = .42). CONCLUSIONS: Children with CHD, particularly single-ventricle patients, require substantially greater hospital resource utilization and have significantly worse outcomes during the first year after heart transplantation compared with other indications. Further work is aimed at identifying modifiable pre-transplantation risk factors, such as pre-transplantation conditioning with VAD support and cardiac rehabilitation, to improve post-transplantation outcomes and reduce resource utilization in this complex population.

Comparable on-therapy mortality and supportive care requirements in Black and White patients following initial induction for pediatric acute myeloid leukemia.

BACKGROUND: Black patients with acute myeloid leukemia (AML) are more likely to present with high acuity and consequently experience higher rates of induction mortality than white patients. Given the consistently identified racial disparities in overall survival (OS) among patients with AML, we aimed to evaluate whether there were sustained on-therapy racial differences in inpatient mortality, intensive care unit (ICU) requirements, or supportive care beyond initial induction. PROCEDURE: Within a retrospective cohort of 1239 children diagnosed with AML between 2004 and 2014 in the Pediatric Health Information System (PHIS) database who survived their initial course of induction chemotherapy, we compared on-therapy inpatient mortality, ICU-level care requirements, treatment course duration, cumulative length of hospital stay (LOS), and resource utilization after induction I by race. RESULTS: Over the period from the start of induction II through completion of frontline chemotherapy, there were no significant differences in mortality (adjusted odds ratios [OR], 1.01; 95% confidence intervals [CI], 0.41-2.48), ICU-level care requirements (adjusted OR, 0.93; 95% CI, 0.69-1.26), LOS (adjusted mean difference, 3.2 days; 95% CI, -2.3-9.6), or supportive care resource utilization for black patients relative to white patients. Course-specific analyses also demonstrated no differences by race. CONCLUSION: Although black patients have higher acuity at presentation and higher induction mortality, such disparities do not persist over subsequent frontline chemotherapy treatment. This finding allows interventions aimed at reducing disparities to be directed at presentation and induction.

Complete Versus Staged Repair for Neonates With Tetralogy of Fallot: Establishment and Validation of a Cohort of 2235 Patients Using Detailed Surgery Sequence Review of Health Care Administrative Data.

BACKGROUND: The surgical strategy for neonates with tetralogy of Fallot (TOF) consists of complete or staged repair. Assessing the comparative effectiveness of these approaches is facilitated by a large multicenter cohort. We propose a novel process for cohort assembly using the Pediatric Health Information System (PHIS), an administrative database that contains clinical and billing data for inpatient and emergency department stays from tertiary children's hospitals. METHODS: A 4-step process was used to identify neonates with TOF: (1) screen neonates in PHIS with International Classification of Diseases-9 (ICD-9) diagnosis or procedure codes for TOF; (2) include patients with TOF procedures before 30 days of age; (3) exclude patients with missing 2-year follow-up data; (4) analyze patients' 2-year surgery sequence patterns, exclude patients inconsistent with a treatment strategy for TOF, and designate patients as complete or staged repair. Manual chart review at 1 PHIS center was performed to validate this process. RESULTS: Between January 2004 and March 2015, 5862 patients were identified in step 1. Step 2 of cohort assembly excluded 3425 patients (58%); step 3 excluded 148 patients (3%); and step 4 excluded 54 patients (1%). The final cohort consisted of 2235 neonates with TOF from 45 hospitals. Manual chart review of 336 patients showed a positive predictive value for accurate PHIS identification of 44% after step 1 and 97% after step 4. CONCLUSIONS: This systematic cohort identification algorithm resulted in a high positive predictive value to appropriately categorize patients. This carefully assembled cohort offers a unique opportunity for future studies in neonatal TOF outcomes.

Successful merging of data from the United Network for Organ Sharing and the Pediatric Health Information System databases.

Data routinely collected through United Network for Organ Sharing (UNOS) lack the detailed information on medical resource utilization and treatment costs required to accomplish for center-level comparisons of quality of care and cost for pediatric heart transplantation. We aimed to overcome this limitation by merging UNOS with the Pediatric Health Information System (PHIS) database, an administrative database containing inpatient, emergency department, ambulatory surgery, and observation unit information from over 40 not-for-profit, tertiary care pediatric hospitals. Utilizing a probabilistic match based on center, date of birth, recipient gender, and transplant date within ±2 days, more than 90% of eligible UNOS patients (N = 2264) were successfully merged to their corresponding PHIS records. Thirty-day and 1-year mortality rates observed for the merged cohort (3.2% and 9.0%, respectively) were compared with those previously reported for pediatric heart transplants, as were the significant predictors of increased mortality. These results demonstrate that the established UNOS-PHIS cohort will provide a valid platform for subsequent research aimed at identifying center-level differences that could be exploited to optimize quality of care while minimizing cost across institutions.

Hospital Variation in Intensive Care Resource Utilization and Mortality in Newly Diagnosed Pediatric Leukemia.

OBJECTIVES: To evaluate hospital-level variability in resource utilization and mortality in children with new leukemia who require ICU support, and identify factors associated with variation. DESIGN: Retrospective cohort study. SETTING: Children's hospitals contributing to the Pediatric Health Information Systems administrative database from 1999 to 2011. PATIENTS: Inpatients less than 25 years old with newly diagnosed acute lymphocytic leukemia or acute myeloid leukemia requiring ICU support (n = 1,754). INTERVENTIONS, MEASUREMENTS, AND MAIN RESULTS: Evaluated exposures included leukemia type, year of diagnosis, and hospital-wide proportion of patients with public insurance. The main outcome was hospital mortality. Wide variability existed in the ICU resources used across hospitals. Combined acute lymphocytic leukemia and acute myeloid leukemia mortality varied by hospital from 0% (95% CI, 0-14.8%) to 42.9% (95% CI, 17.7-71.1%). A mixed-effects model with a hospital-level random effect suggests significant variation across hospitals in mortality (p = 0.007). When including patient and hospital factors as fixed effects into the model, younger age, acute myeloid leukemia versus acute lymphocytic leukemia diagnosis, leukemia diagnosis prior to 2005, hospital-wide proportion of public insurance patients, and hospital-level proportion of leukemia patients receiving ICU care are significantly associated with mortality. The variation across hospitals remains significant with all patient factors included (p = 0.021) but is no longer significant after adjusting for the hospital-level factors proportion of public insurance and proportion receiving ICU care (p = 0.48). CONCLUSIONS: Wide hospital-level variability in ICU resource utilization and mortality exists in the care of children with leukemia requiring ICU support. Hospital payer mix is associated with some mortality variability. Additional study into how ICU support could be standardized through clinical practice guidelines, impact of payer mix on hospital resources allocation to the ICU, and subsequent impact on patient outcomes is warranted.

The role of acuity of illness at presentation in early mortality in black children with acute myeloid leukemia.

Black patients with acute myeloid leukemia (AML) experience higher mortality than White patients. We compared induction mortality, acuity of illness prior to chemotherapy, and insurance type between Black and White patients to assess whether acuity of presentation mediates the disparity. Within a retrospective cohort of 1,122 children with AML treated with two courses of standard induction chemotherapy between 2004 and 2014 in the Pediatric Health Information System (PHIS) database, the association between race (Black versus White) and inpatient mortality during induction was examined. Intensive Care Unit (ICU)-level resource utilization during the first 72 hours following admission for initial AML chemotherapy was evaluated as a potential mediator. The total effect of race on mortality during Induction I revealed a strong association (unadjusted HR 2.75, CI: 1.18, 6.41). Black patients had a significantly higher unadjusted risk of requiring ICU-level resources within the first 72 hours after initial presentation (17% versus 11%; RR 1.52, CI: 1.04, 2.24). Mediation analyses revealed the indirect effect of race through acuity accounted for 61% of the relative excess mortality during Induction I. Publicly insured patients experienced greater induction mortality than privately insured patients regardless of race. Black patients with AML have significantly greater risk of induction mortality and are at increased risk for requiring ICU-level resources soon after presentation. Higher acuity amongst Black patients accounts for a substantial portion of the relative excess mortality during Induction I. Targeting factors affecting acuity of illness at presentation may lessen racial disparities in AML induction mortality.

Complications preceding early deaths in Black and White children with acute myeloid leukemia.

Black patients have a twofold increased risk of induction mortality compared to White patients with acute myeloid leukemia (AML). We reviewed diagnosis and billing data from Pediatric Health Information System for 28 AML Induction I deaths to investigate conditions preceding death in White and Black patients. Half of deaths occurred within 10 days of initial diagnostic admission. Respiratory, cardiac, renal, and infectious complications were common prior to both White and Black deaths. Deaths in White patients were more commonly preceded by intracranial hemorrhage compared to deaths in Black patients. Future studies should assess management approaches of complications by race to identify modifiable processes that reduce mortality.

Low rates of pregnancy screening in adolescents before teratogenic exposures in a national sample of children's hospitals.

BACKGROUND: Adolescents with cancer engage in sexual behaviors and are exposed to teratogenic chemotherapy. There are no data regarding pregnancy screening patterns for adolescents before chemotherapy exposure. METHODS: A cross-sectional study of leukemia and emergency room (ER) admissions in the Pediatric Health Information System from 1999 to 2011 was conducted. Females who were 10 to 18 years old and 1) had newly diagnosed acute lymphoblastic leukemia (ALL) or acute myeloid leukemia (AML) or 2) had ER visits with computed tomography (CT) of the abdomen/pelvis were included. The exposure was a hospital visit with either chemotherapy or an abdominal/pelvic CT scan. The main outcome was a pregnancy test billed on the same day or before the teratogenic exposure within the same index admission. Log-binomial regressions were used to compute prevalence ratios and 95% confidence intervals comparing pregnancy screening in the leukemia and ER cohorts. RESULTS: A total of 35,650 admissions were identified. The proportion of visits with an appropriately timed pregnancy test was 35%, 64%, and 58% in the ALL (n = 889), AML (n = 127), and ER cohorts (n = 34,634), respectively. Patients with ALL were significantly less likely to have a pregnancy test than the ER cohort (adjusted prevalence ratio, 0.71; 95% confidence interval, 0.65-0.78), but there was no significant difference between the AML and ER cohorts (adjusted prevalence ratio, 1.12; 95% confidence interval, 0.99-1.27). There was substantial hospital-level variation in pregnancy screening patterns. CONCLUSIONS: Adolescents with acute leukemia and ER visits have low rates of pregnancy screening before teratogenic exposures. Standardized practice guidelines for pregnancy screening among adolescents may improve screening rates. Cancer 2016;122:3394-3400. © 2016 American Cancer Society.

Supportive care utilization and treatment toxicity in children with Down syndrome and acute lymphoid leukaemia at free-standing paediatric hospitals in the United States.

Although inferior outcomes of children with Down syndrome (DS) and acute lymphoid leukaemia (ALL) are established, national supportive care patterns for these patients are unknown. A validated retrospective cohort of paediatric patients diagnosed with ALL from 1999 to 2011 was assembled from the US Pediatric Health Information System (PHIS) database to examine organ toxicity, sepsis, and resource utilization in children with and without DS. Among 10699 ALL patients, 298 had DS-ALL (2·8%). In a multivariate model, DS was associated with increased risk of cardiovascular (odds ratio [OR] 2·0, 95% confidence interval [CI] 1·6-2·7), respiratory (OR 2·1, 95% CI: 1·6-2·9), neurologic (OR 3·4, 95% CI 1·9-6·2), and hepatic (OR 1·4, 95% CI 1·0-1·9) dysfunction and sepsis (OR 1·8, 95% CI: 1·4-2·4). Children with DS-ALL used significantly more respiratory support, insulin, and anti-infectives, including broad-spectrum Gram-positive agents, quinolones, and azoles. They used significantly fewer analgesics and antiemetics compared to non-DS-ALL children. Ultimately, this study confirms the increased risk of infectious and end-organ toxicity in children with DS-ALL and quantifies important differences in resource utilization between children with DS and non-DS ALL. These findings highlight the importance of investigating the impact of these care variations and developing specific supportive care guidelines for this population.