US Pediatric Population Research Articles

Economic Burden of Growth Hormone Deficiency in a US Pediatric Population

Pediatric GHD is a rare disorder of short stature that is currently treated with daily injections of GH. In addition to short stature, GHD is associated with other comorbidities such as impaired musculoskeletal development, cardiovascular disease, and decreased quality of life. The objective of this study was to analyze GH utilization, adherence, and healthcare costs among children with GHD who had either Medicaid or commercial health insurance. Children (age <18 years) with a diagnosis of GHD between January 1, 2007, and December 31, 2017 were identified in the IBM Marketscan Commercial and Medicaid Databases. Children with GHD were direct matched (1:3) to controls without GHD (or other short stature-related disorder) on age, gender, plan type, region, and race (Medicaid only). The index date was set as the date of the first GHD diagnosis during the selection window for GHD patients and using random assignment for controls. Patients were followed for the 12 months prior to the index date until the end of continuous database enrollment or December 31, 2018. Baseline comorbidities and medications were measured during the 12 months pre-index. Treatment patterns and all-cause and GHD-related healthcare costs were measured during the variable follow-up period. Multivariable modeling was used to compare costs between GHD patients and controls and between GH treated and untreated GHD patients while adjusting for baseline characteristics. There were 6,820 Medicaid and 14,070 commercial patients with GHD who met the study inclusion criteria. Mean (SD) age at index was 9.5 (4.5) years for Medicaid patients and 11.1 (3.7) years for commercial patients. A majority of patients were male (>65%) and followed for at least 3.7 years. Overall, 63.2% of Medicaid and 68.4% of commercial patients were treated with GH during follow-up. Among Medicaid patients, the treatment rate was highest among white males and lowest among black females. Adherence, as measured by proportion of days covered, was low, with 18.4% of Medicaid patients and 32.3% of commercial patients considered “adherent” (PDC ≥ 0.8). Nearly half (49.1%) of treated Medicaid patients and 24.3% of commercial patients discontinued GH therapy before age 13. After adjusting for baseline characteristics, all-cause non-GH costs were 5.7times higher (Δ$19,309) for Medicaid patients and 5.5 times higher (Δ$12,305) for commercial patients than matched non-GHD controls. Adjusted all-cause non-GH costs were 0.6 times lower (Δ$14,416) for treated Medicaid patients and 0.7 times lower (Δ$7,650) for treated commercial patients than for untreated patients. Pediatric GHD presents a significant healthcare burden, and many patients remain untreated or undertreated. Untreated GHD was associated with higher non-GH healthcare costs than treated GHD. Strategies to improve adherence may reduce the healthcare burden faced by these patients.

Predicting the natural development of peanut tolerance using longitudinal trajectories of peanut-specific serum IgE

Peanut allergy is one of the most common food-specific allergies and is estimated to affect 2.2% of the US pediatric population.1 Tolerance to peanut can develop in 20% to 25% of children previously diagnosed with peanut allergy.2,3 Earlier recognition of children more likely to develop tolerance is useful for counseling on prognosis. Currently, the primary means of diagnosing and trending the natural history of peanut allergy are skin prick tests and peanut-specific serum IgE (sIgE) levels.4 Ideally, in practice, the goal is to optimally use less-invasive testing to identify the likelihood of a favorable outcome before performing food challenges.

Economic burden of growth hormone deficiency in a US pediatric population.

BACKGROUND: Pediatric growth hormone deficiency (GHD) is a rare disorder of short stature that is currently treated with daily injections of somatropin. In addition to short stature, GHD is associated with other comorbidities such as impaired musculoskeletal development, cardiovascular disease, and decreased quality of life. OBJECTIVE: To analyze somatropin utilization, adherence, and health care costs among children with GHD who had either Medicaid or commercial health insurance. METHODS: Children (aged < 18 years) with a GHD diagnosis between January 1, 2008, and December 31, 2017, were identified in the IBM MarketScan Commercial and Medicaid databases. Patients with at least 12- and 6-month continuous enrollment pre- and postdiagnosis were eligible. Children with GHD were direct matched (1:3) to controls without GHD (or other short stature-related disorders) on age, gender, plan type, region, and race (Medicaid only). Index date was the date of the first GHD diagnosis during the selection window for GHD patients and using random assignment for controls. Patients were followed until the end of continuous database enrollment or December 31, 2018. Baseline comorbidities and medications were measured during the 12 months pre-index, whereas somatropin treatment patterns along with all-cause and GHD-related health care costs were measured during the variable follow-up period. Multivariable modeling was used to compare costs between GHD patients and controls and between somatropin-treated and -untreated GHD patients while adjusting for baseline characteristics. RESULTS: There were 6,820 Medicaid and 14,070 commercial patients with GHD who met the study inclusion criteria. Mean (SD) age at index was 9.5 (4.5) years for Medicaid patients and 11.1 (3.7) years for commercial patients. The majority of patients were male (> 65%), and mean follow-up time for all cases and controls was 3-4 years. Overall, 63.2% of Medicaid and 68.4% of commercial GHD patients were treated with somatropin at some point during follow-up. Among Medicaid GHD patients, the treatment rate was highest among White males and lowest among Black females. Adherence was low as the proportion of days covered was ≥ 80% for only 18.4% of Medicaid patients and 32.3% of commercial patients and 49.1% of treated Medicaid and 24.3% of treated commercial patients discontinued before turning age 13. After adjusting for baseline characteristics, all-cause non-somatropin annualized costs were 5.67 times higher (Δ$19,309) for Medicaid GHD patients and 5.46 times higher (Δ$12,305) for commercial GHD patients than matched non-GHD controls. Adjusted all-cause non-somatropin annualized costs were 0.59 times lower (Δ$14,416) for treated Medicaid patients and 0.69 times lower (Δ$7,650) for treated commercial patients than for untreated patients. CONCLUSIONS: Pediatric GHD presents a significant health care burden, and many patients remain untreated or undertreated. Untreated GHD was associated with higher non-somatropin health care costs than treated GHD. Strategies to optimize treatment and improve adherence may reduce the health care burden faced by these patients. DISCLOSURES: This study was funded by Ascendis Pharma, Inc. Smith and Pitukcheewanont are employed by Ascendis Pharma, Inc. Manjelievskaia, Lopez-Gonzalez, and Morrow are employed by IBM Watson Health, which received funding from Ascendis Pharma, Inc., to conduct this study. Kaplowitz is a paid consultant of Ascendis Pharma, Inc.

1395. Influenza B-Associated Pediatric Mortality in the US Between 2010 and 2019

BackgroundTo assess the contribution of influenza B to mortality in the US pediatric population, we analyzed the proportion of influenza-associated pediatric mortality attributed to influenza A and B over nine influenza seasons using national surveillance data. The effectiveness of influenza vaccines against influenza B in the pediatric population was also assessed.MethodsThe study period was the 2010/11 to 2018/19 influenza seasons. Proportions of circulating strains in the general population and influenza-associated pediatric mortality for each season were obtained from annual Centers for Disease Control and Prevention Morbidity and Mortality Weekly Reports on influenza. A Chi-squared test with Yates’ correction was used to assess the contribution of influenza B to pediatric mortality relative to its circulation among influenza viruses. Consolidated vaccine effectiveness (VE) against influenza B for inactivated influenza vaccine (IIV) in the 2010/11 to 2017/18 seasons and live attenuated influenza vaccine (LAIV) in the 2010/11 to 2015/16 seasons were obtained from a published meta-analysis and annual US Flu VE Network studies. There were no US data on LAIV VE for 2016/17 and 2017/18.ResultsDuring the 2010/11 to 2018/19 seasons, influenza B accounted for 4.0%–29.2% of all circulating influenza strains. A/H3N2 viruses were the predominant circulating strain in most seasons. In the same period, influenza B accounted for 7.0%–54.1% of pediatric influenza-associated mortality (Figure). The proportion of influenza B-related deaths was significantly higher (p< 0.01) than what would have been expected based on the proportion of circulating influenza B strains in the general population, overall and in the 2010/11, 2012/13, 2016/17, and 2017/18 seasons. Point estimates of VE against influenza B for children aged 2–17 years ranged from 33%–70% for IIV between 2010/11 and 2017/18, and from 53%–82% for LAIV between 2010/11 and 2015/16.Proportion of circulating influenza B strains compared with influenza B-associated pediatric mortality in the US between the 2010/11 and 2018/19 seasons ConclusionDuring the study period, influenza B accounted for a disproportionate percentage of pediatric mortality in the US relative to its overall circulation. These data counter the perception that influenza B is less severe than influenza A in children and highlight the importance of influenza vaccination to prevent influenza and its complications.DisclosuresAllyn Bandell, PharmD, AstraZeneca (Employee, Shareholder) Pedro Piedra, MD, Roche (Consultant)Shionogi (Grant/Research Support) Christopher S. Ambrose, MD, MBA, AstraZeneca (Employee, Shareholder) Ravi Jhaveri, MD, AbbVie (Grant/Research Support)Alios (Research Grant or Support)Gilead (Grant/Research Support)Medimmune/AstraZeneca (Employee)Merck (Grant/Research Support)Saol Therapeutics (Consultant)Seqirus (Consultant)

Neighborhood risk: Socioeconomic status and hospital admission for pediatric burn patients

BackgroundBurn injury continues to cause significant morbidity and mortality in the US pediatric population. Many studies using inpatient samples have found a relationship between low socioeconomic status (SES) and burn injury. The purpose of our study was to evaluate the association between SES and the likelihood of admission for Emergency Department (ED) visits for pediatric burn injury. Study designA retrospective database review of pediatric ED visits for burn injury from a statewide hospital system, from January 1, 2005 to December 31, 2014. SES was assigned using an eight factor Neighborhood Risk Index (NRI) created from census block group data, with a higher score indicative of lower SES. The outcome measure was ED visits admitted to inpatient care. ResultsWe analyzed a sample of 1845 pediatric ED visits for burn injuries. Most visits were discharged from the ED (88.4%) while 10.5% were admitted to inpatient care and 1.0% were transferred to another hospital. In a multivariable logistic regression model, patients from high risk areas (>75th percentile NRI) had 1.58 higher odds of inpatient admission compared to patients from low risk areas (<75th percentile NRI; 95% CI: 1.08–2.30), after adjusting for age, gender, ethnicity, distance to the hospital, and previous ED visit for burn injury in the past 30 days. In addition, for every 1-mile increase in distance, a child’s likelihood of admission increased by 6% (95% CI: 4–9%). ConclusionsChildren with a burn injury from the highest risk socioeconomic areas in Rhode Island had a higher likelihood of inpatient admission. Further research is needed to determine what factors associated with socioeconomic status impact this finding.

A045 PHENOTYPIC DIFFERENCES IN ALLERGY TO MULTIPLE FOODS WITHIN A COHORT OF BLACK AND WHITE CHILDREN

IgE-mediated food allergy (FA) affects 8% of the US pediatric population, with greater disease burden observed among Black children, who previous survey data indicate may be more likely to be allergic to multiple foods. However, little work has investigated putative racial differences in multi-food allergy phenotypes within clinical FA populations.

Racial differences in prediabetes prevalence by test type for the US pediatric and adult population: NHANES 1999-2016.

Previous studies have shown that US estimates of prediabetes or diabetes differ depending on test type, fasting plasma glucose (FPG) vs hemoglobin A1c (HbA1c). Given age, race, and test differences reported in the literature, we sought to further examine these differences in prediabetes detection using a nationally representative sample. Using the National Health and Nutrition Examination Survey (NHANES) 1999-2016, individuals were identified as having prediabetes with an HbA1c of 5.7% to 6.4% or a FPG of 100 to 125 mg/dL. We excluded individuals with measurements in the diabetic range. We ran generalized estimating equation logistic regressions to examine the relationship between age, race, and test type with interactions, controlling for sex and body mass index. We compared the difference in predicted prediabetes prevalence detected by impaired fasting glycemia (IFG) vs HbA1c by race/ethnicity among children and adults separately using adjusted Wald tests. The absolute difference in predicted prediabetes detected by IFG vs HbA1c was 19.9% for white adolescents, 0% for black adolescents, and 20.1% for Hispanic adolescents; 21.4% for white adults, -1.2% for black adults, and 19.2% for Hispanic adults. Using adjusted Wald tests, we found the absolute differences between black vs white and black vs Hispanic individuals to be significant, but, not between Hispanic and white individuals among children and adults separately. These observations highlight differences in test performance among racial/ethnic groups. Our findings corroborate the need for further studies to determine appropriate HbA1c cutoff levels for diagnosis of prediabetes by age group and race.

Typhoid Fever in the US Pediatric Population, 1999-2015: Opportunities for Improvement.

Typhoid fever in the United States is acquired primarily through international travel by unvaccinated travelers. There is currently no typhoid vaccine licensed in the United States for use in children <2 years. We reviewed Salmonella enterica serotype Typhi infections reported to the Centers for Disease Control and Prevention (CDC) and antimicrobial-resistance data on Typhi isolates in CDC's National Antimicrobial Resistance Monitoring System from 1999 through 2015. 5131 cases of typhoid fever were diagnosed and 5004 Typhi isolates tested for antimicrobial susceptibility. Among 1992 pediatric typhoid fever patients, 1616 (81%) had traveled internationally within 30 days of illness onset, 1544 (81%) of 1906 were hospitalized (median duration, 6 days; range, 0-50), and none died. Forty percent (799) were <6 years old; 12% were <2 years old. Based on age and travel destination, 1435 (83%) of 1722 pediatric patients were vaccine-eligible; only 68 (5%) of 1361 were known to be vaccinated. Of 2003 isolates tested for antimicrobial susceptibility, 1216 (61%) were fluoroquinolone-nonsusceptible, of which 272 (22%) were also resistant to ampicillin, chloramphenicol, and trimethoprim-sulfamethoxazole (multidrug-resistant [MDR]). All were susceptible to ceftriaxone and azithromycin. MDR and fluoroquinolone-nonsusceptible isolates were more common in children than adults (16% vs 9%, P < .001, and 61% vs 54%, P < .001, respectively). Fluoroquinolone nonsusceptibility was more common among travel-associated than domestically acquired cases (70% vs 17%, P < .001). Approximately 95% of currently vaccine-eligible pediatric travelers were unvaccinated, and antimicrobial-resistant infections were common. New public health strategies are needed to improve coverage with currently licensed vaccines. Introduction of an effective pretravel typhoid vaccine for children <2 years could reduce disease burden and prevent drug-resistant infections.

Motor Vehicle Collision Injuries: An Analysis of Facial Fractures in the Urban Pediatric Population.

Motor vehicle collisions (MVC) are a leading cause of unintentional death and injury in the US pediatric population. Compliance with prevention measures such as seatbelts and child safety seats varies considerably with patient demographics. In this study, the authors examine facial fracture secondary to MVC in an urban pediatric population. A retrospective chart review was performed of all facial fractures as a result of MVC in the pediatric population in a level 1 trauma center in an urban environment (University Hospital in Newark, NJ). Patient demographics were collected, as well as location of fractures, concomitant injuries, and surgical management strategies. Seventy-five patients met inclusion criteria for the authors' study. Mean age was 14 years old. Common fracture sites included orbital, mandible, nasal bone, and frontal sinus. Patients were more likely to incur fracture of the zygoma, orbit, nasal bone, frontal sinus, and nasoorbitoethmoid (NOE) if involved in an MVC compared to all other etiologies. Common concomitant injuries included traumatic brain injury, intracranial hemorrhage, and skull and long bone fractures. Open reduction and internal fixation with titanium plates was the most common surgical procedure indicated. Motor vehicle collisions related injury has significant health implications in the urban pediatric population. Orbital, zygoma, and nasal facial fractures and TBI are injuries commonly associated with MVC. The facial fractures are likely due to lack of proper utilization of safety equipment and airbags. Development of effective prevention techniques relies heavily on analysis of injury patterns and management strategies.

Trends in Sodium Intake in Children and Adolescents in the US and the Impact of US Department of Agriculture Guidelines: NHANES 2003-2016

To examine trends in sodium intake and the impact of nutritional guidelines in the US pediatric population. Sodium intake data collected between 2003 and 2016 in the US National Health and Nutrition Examination Surveys (NHANES) were analyzed. Trends in intake for individuals aged 4-17years and subgroups based on age, sex, and race and ethnicity were examined. Adherence to US Department of Agriculture guidelines was assessed. A total of 16 013 individuals (50.6% male) were included in the analysis. The median sodium intake was2840mg/day (95% CI, 2805-2875mg/day), decreasing from 2912mg/day (95% CI 2848-2961mg/day) in 2003-2004 to 2787mg/day (95% CI, 2677-2867mg/day) in 2015-2016 (P=.005). Intake increased with age (2507mg/day for individuals aged 4-8, 2934mg/day for those aged 9-13 years, and 3124mg/day for those aged 14-17 years; P<.001) and was greater in males than in females (3053mg/day vs 2624mg/day; P<.001). Caucasians, Hispanics, and African Americans consumed 2860, 2733, and 2880mg/day, respectively (P<.001). Population adherence to US Department of Agriculture recommendations was 25.0% in 2003-2010 and 25.5% in 2011-2016 (P=.677). No age, sex, or racial/ethnicity subgroup had an adherence rate >30% after implementation of pediatric guidelines in2010. Sodium intake remains elevated in all pediatric population segments, and guideline adherence ispoor. A greater effort to reduce sodium consumption is needed to mitigate future cardiovascular disease risk.

Egg Allergy in US Children

Egg allergy is common in young children (<5years) and has significant negative impacts on quality of life. The objective of this study was to characterize egg allergy prevalence, severity, baked egg tolerance, and other associated factors in a large US cohort. A national cross-sectional survey was administered from October 2015 to September 2016, resulting in complete parent-proxy responses for 38,408 children. Weighted proportions were estimated to compare egg allergy prevalence and characteristics between key subpopulations. The overall prevalence of current, convincingly egg allergy was 0.9% among all children and 1.3% among children <5 years. Black children were over-represented among children with egg allergy, accounting for 23.4% (95% confidence interval: 13.1-38.4) of egg-allergic children despite comprising 13.2% (12.3-14.2) of the US pediatric population. Among children with egg allergy, 64.2% reported baked egg tolerance and 60.2% had allergy to other foods, with 29.3% having peanut allergy. Asthma was more prevalent in children with an egg allergy than children with other top 8 food allergies (46.5% [35.8-57.4] vs 33.2% [29.6-37.0], P < .05). Among children with current egg allergy, those with baked egg tolerance reported that their food allergy resulted in significantly reduced psychosocial burden, relative to their baked egg-allergic counterparts (M= 3.1 [2.9-3.3] vs M= 3.7 [3.5-3.9]). Egg allergy is common amongst young children. Nearly two-thirds of children with egg allergy reported baked egg tolerance. Increased efforts are needed to ensure that children with egg allergy are appropriately evaluated as many have comorbid allergic disease and determination of baked egg tolerance may improve quality of life.

PMD29 HOSPITAL VARIATIONS IN THE USE OF ROBOTIC-ASSISTED SURGERIES IN THE U.S. PEDIATRIC POPULATION

Robotic surgery has been widely recognized by the medical field due to its advancement in minimally invasive surgical techniques. This retrospective cohort study evaluated the role of hospital-level characteristics in the use of robotic-assisted surgeries in the US pediatric population.

Utilization of Healthcare Resources by HIV-Positive Children in the United States: A National Perspective.

We characterized and estimated the cost of inpatient hospital utilization by US pediatric patients who tested positive for the human immunodeficiency virus (HIV). The 2012 Kids' Inpatient Database was analyzed to provide a descriptive assessment of national inpatient hospital utilization. We analyzed a stratified probability sampling of 3.2 million pediatric hospital discharges weighted to 6.7 million national discharges. Descriptive statistics for hospital and patient characteristics were identified and binary variables were analyzed using the Student t test. The Kids' Inpatient Database is the largest available all-payer pediatric (20 years old and younger) inpatient care database in the United States, yielding national estimates of hospital inpatient stays. Children aged 17 years and younger were included in the study and conditions related to pregnancy and delivery. We estimated that 1344 pediatric discharges were associated with an HIV diagnosis, totaling 10,704 inpatient days at a cost of $91 million. Among pediatric patients with HIV, 55% were African American, 20% were white, 15% were Asian/Pacific Islander, 8% were other races (including Hispanics and Native Americans), and 51% were female. Children who were HIV positive were more likely to have longer mean hospital stays, have higher mean hospital charges, be of a higher median age (8 years and older), have Medicaid insurance, come from lower-income families, be treated in urban teaching hospitals, and be more likely to die during hospitalization (P < 0.01 for all). Among non-HIV-related pediatric discharges, 20% occurred in households with a mean annual income >$63,000 compared with only 12% for children who were HIV positive. During hospitalization, at least one procedure was performed in 56.6% of children with HIV compared with 45.65% of hospitalized children without HIV. The most frequently observed diagnoses associated with children infected with HIV were gastrointestinal disorders, mental disorders, and bacterial infections and sepsis. The results suggest that pediatric patients who were HIV positive were significantly older, from lower-income areas, and members of minority groups. They underwent more procedures during hospitalization, incurred more than twice the total cost, stayed in the hospital twice as long, and had statistically higher in-hospital mortality than children who were HIV negative. As we continue to explore effective and judicious treatment options for patients who are HIV positive, our national estimates of resource utilization can be used to conduct a more detailed examination of current medical practices and specific patterns of diagnoses associated with HIV infection in the US pediatric population.

Prevalence and Characteristics of Shellfish Allergy in the Pediatric Population of the United States

Shellfish allergy (SA) is one of the most common food allergies causing anaphylaxis in adults and children. There are limited data showing the prevalence of SA in US children. To determine the prevalence and reaction characteristics of SA in the US pediatric population. A cross-sectional food allergy prevalence survey was administered via phone and the Web by the National Opinion Research Center at the University of Chicago from 2015 to 2016. Point prevalence SA estimates, complex survey weighted proportions, and 95% CIs were determined. Relative proportions of demographic characteristics were compared using weighted Pearson χ2 statistics. The prevalence of SA was 1.3% (95% CI, 1.1-1.5), with more children allergic to crustaceans (1.2%; 95% CI, 1.0-1.3) than to mollusks (0.5%; 95% CI, 0.4-0.6). Mean ages of shellfish, crustacean, and mollusk allergy diagnoses were 5.0 (95% CI, 4.4-5.6), 5.1 (95% CI, 4.6-5.6), and 7.7 (95% CI, 5.7-9.7) years, respectively. More than half (54.9%; 95% CI, 48.1-61.4) of pediatric patients with SA had more than 1 lifetime food allergy-related emergency room visit, but only 45.7% (95% CI, 39.2-52.4) carried an epinephrine autoinjector. Children with SA were more likely to be black/Hispanic/Latino and have comorbid asthma, allergic rhinitis, or a parental history of asthma, environmental, or other food allergies (P < .001). The epidemiology of SA in the US pediatric population shows that crustacean allergy is more common than mollusk allergy. A disparity in SA children and epinephrine autoinjector carriage exists. Results from this study will lead to increased awareness of the need for detailed histories, specific diagnostic tests, and rescue epinephrine for anaphylaxis in US children with SA.

P300 A COMPARISON OF CRUSTACEAN AND MOLLUSK ALLERGY IN THE PEDIATRIC POPULATION OF THE UNITED STATES

Shellfish allergy (SA) is one of the most common food allergies causing severe reactions, however, there have been limited data showing trends surrounding its subtypes, crustacean and mollusk allergy, in the US pediatric population.

1493. Creation and External Validation of a Clinical Prediction Rule for Diarrheal Etiology Using Natural Language Processing

BackgroundInfectious diarrheal illness is a significant contributor to healthcare costs in the US pediatric population. New multi-pathogen PCR-based panels have shown increased sensitivity over previous methods; however, they are costly and clinical utility may be limited in many cases. Clinical Prediction Rules (CPRs) may help optimize the appropriate use of these tests. Furthermore, Natural Language Processing (NLP) is an emerging tool to extract clinical history for decision support. Here, we examine NLP for the validation of a CPR for pediatric diarrhea.MethodsUsing data from a prospective clinical trial at 5 US pediatric hospitals, 961 diarrheal cases were assessed for etiology and relevant clinical variables. Of 65 variables collected in that study, 42 were excluded in our models based on a scarcity of documentation in reviewed clinical charts. The remaining 23 variables were ranked by random forest (RF) variable importance and utilized in both an RF and stepwise logistic regression (LR) model for viral-only etiology. We investigated whether NLP could accurately extract data from clinical notes comparable to study questionnaires. We used the eHOST abstraction software to abstract 6 clinical variables from patient charts that were useful in our CPR. These data will be used to train an NLP algorithm to extract the same variables from additional charts, and be combined with data from 2 other variables coded in the EMR to externally validate our model.ResultsBoth RF and LR models achieved cross-validated area under the receiver operating characteristic curves of 0.74 using the top 5 variables (season, age, bloody diarrhea, vomiting/nausea, and fever), which did not improve significantly with the addition of more variables. Of 270 charts abstracted for NLP training, there were 41 occurrences of bloody diarrhea annotated, 339 occurrences of vomiting, and 145 occurrences of fever. Inter-annotator agreement over 9 training sets ranged between 0.63 and 0.83.ConclusionWe have constructed a parsimonious CPR involving only 5 inputs for the prediction of a viral-only etiology for pediatric diarrheal illness using prospectively collected data. With the training of an NLP algorithm for automated chart abstraction we will validate the CPR. NLP could allow a CPR to run without manual data entry to improve care. Disclosures All authors: No reported disclosures.

Variability In States' Coverage Of Children With Medical Complexity Through Home And Community-Based Services Waivers.

Even though children with medical complexity represent less than 1percent of the US pediatric population, they are among the costliest users of the health care system. Much of the care for these children is delivered in home and community-based settings and covered by Medicaid waivers under Section 1915(c). Expenditures related to these waivers have been steadily increasing, with most recent estimates showing spending that exceeds $48billion per year. Little is known about these waivers' economic impact or effectiveness, because their components and coverage have not previously been well defined. Our study addressed this paucity of data by analyzing the scope of coverage offered by the waivers and evaluating how states are using them to cover children with medical complexity. We found great variability in how states choose to interpret scope of coverage and services offered, and this variability may have an impact on child and family outcomes.

Prevalence of type 1 and type 2 diabetes among US pediatric population in the MarketScan Multi-State Database, 2002 to 2016.

To estimate the prevalence of type 1 (T1DM) and type 2 diabetes mellitus (T2DM) among U.S. Medicaid pediatric population aged <18 years 2002 to 2016 by age, sex, and race/ethnicity. Participants aged <18 years old from 2002 to 2016 were identified from the MarketScan Multi-State Medicaid Database. Diabetes was defined as having (a) ≥1 claims for an outpatient or inpatient diabetes diagnosis and ≥2 prescriptions for any anti-diabetes medications or (b) records of ≥2 claims for an outpatient or inpatient diabetes diagnosis that were at least 30 days apart. Annual prevalence of diabetes and 95% confidence intervals (CIs) were calculated. Age-, sex-, and race-stratified prevalence were also assessed. The annual prevalence of T1DM increased from 1.29 to 2.34/1000 pediatric persons from 2002 to 2016. The prevalence of T2DM rose from 0.70 in 2002 to 2.76/1000 in 2011, but then dropped to 2.12/1000 pediatric persons in 2016 in the Medicaid population. Prevalence of both T1DM and T2DM increased with age. While the prevalence of T1DM was similar in both sexes, and was most prevalent in Whites, prevalence of T2DM was higher in girls and was most prevalent in Blacks. While the annual prevalence of T1DM in pediatric persons enrolled in Medicaid increased continuously from 2002 to 2016, the annual prevalence of T2DM increased from 2002 to 2011, with a subsequent decrease in 2016, possibly because of the increase of relatively healthier participants with the expanded eligibility through the ACA between 2011 and 2016.

Epidemiology of Children With Multiple Complex Chronic Conditions in a Mixed Urban-Rural US Community.

Children with multiple complex chronic conditions (MCCs) represent a small fraction of our communities but a disproportionate amount of health care cost and mortality. Because the temporal trends of children with MCCs within a geographically well-defined US pediatric population has not been previously assessed, health care planning and policy for this vulnerable population is limited. In this population-based, repeated cross-sectional study, we identified and enrolled all eligible children residing in Olmsted County, Minnesota, through the Rochester Epidemiology Project, a medical record linkage system of Olmsted County residents. The pediatric complex chronic conditions classification system version 2 was used to identify children with MCCs. Five-year period prevalence and incidence rates were calculated during the study period (1999-2014) and characterized by age, sex, ethnicity, and socioeconomic status (SES) by using the housing-based index of socioeconomic status, a validated individual housing-based SES index. Age-, sex-, and ethnicity-adjusted prevalence and incidence rates were calculated, adjusting to the 2010 US total pediatric population. Five-year prevalence and incidence rates of children with MCCs in Olmsted County increased from 1200 to 1938 per 100 000 persons and from 256 to 335 per 100 000 person-years, respectively, during the study period. MCCs tend to be slightly more prevalent among children with a lower SES and with a racial minority background. Both 5-year prevalence and incidence rates of children with MCCs have significantly increased over time, and health disparities are present among these children. The clinical and financial outcomes of children with MCCs need to be assessed for formulating suitable health care planning given limited resources.

Cardiovascular Health in Pediatric Heart Transplant Patients

Purpose Ideal “cardiovascular health” (CVH) - optimal levels of 7 metrics: diet, physical activity (PA), nonsmoking, BMI, BP, lipids, & glucose - is associated with greater longevity and quality of life in adults. Pediatric heart transplant (HT) patients (pts) may be at risk for suboptimal CVH due to immunosuppression and comorbidities. Hypotheses HT pts’ CVH scores will be lower compared to the general US pediatric population (GP), will diminish over time, and will correlate with complications. Methods Single-center retrospective study of HT pts from 2003-2014 that survived 1 year post-transplant (PT). Five CVH metrics were collected at listing, 1, 3 and 5 years PT; PA & diet were not available. CVH score was calculated as the sum of individual metrics: ideal=2; intermediate=1 and poor=0; CVH score 8-10 points = high (favorable). We compared the latest PT year with the NHANES weighted dataset representing the GP, adjusted for age, sex & race/ethnicity. Cumulative AUC CVH scores were calculated PT and dichotomized as high (favorable) > 33 and low Results We included 110 HT pts; 56% male, median age 6 years (range 0.1-21). NHANES 2007-2016 had 19,081 participants. Most HT pts had high CVH scores at listing, 1, 3 and 5 years PT: 73, 74, 87 and 75%, respectively. Figure 1A shows differences in the distribution of CVH metrics between HT pts and the GP. Figure 1B shows the most recent CVH score distribution; there were differences in the distribution of CVH scores in age groups (p=0.02). HT pts with low (vs. high) cumulative CVH scores had a HR of 2.95 (p=0.002) for the composite adverse outcome of chronic kidney disease, coronary allograft vasculopathy, relisting or death. Conclusion CVH scores in pediatric HT pts were stable over time with younger pts more likely to have high CVH. Compared to the GP, HT pts had lower ideal glucose but higher ideal lipids. Low cumulative CVH scores were associated with ∼3-fold higher risks of adverse outcome. Future interventions to improve CVH in HT pts, such as targeted counseling, should be considered.