Introduction Autistic youth have higher rates of hyperlipidemia and diabetes than non-autistic youth and thus need age-appropriate monitoring for hyperlipidemia and diabetes. Little is known about how frequently autistic youth are monitored for these conditions. Methods This analysis assessed monitoring for hyperlipidemia and diabetes among 230 autistic youth ages 16 to 30 years (113 were prescribed anti-psychotics, 117 were not) between January 2011 and May 2020. Outcomes assessed included the proportion of patients who had ANY testing for hyperlipidemia and diabetes in both groups, proportion of prescriptions monitored for hyperlipidemia and diabetes in the last year, and identification of patient factors associated with monitoring. Results A significantly higher proportion of autistic youth prescribed anti-psychotics had testing for both hyperlipidemia and diabetes during the study period than autistic youth who were not (73% vs. 49%, p< 0.001). While most autistic youth who were prescribed anti-psychotics had some monitoring done, of the 1538 prescriptions for anti-psychotics (new and renewal) identified, 847 (55%) were considered unmonitored. Having other bloodwork done was significantly associated with higher odds of testing for hyperlipidemia or diabetes (OR 1.45, 95% CI [1.37, 1.56]), but not other factors assessed. Conclusions Most autistic youth prescribed anti-psychotics in this cohort underwent some monitoring for hyperlipidemia and diabetes, but not at a level consistent with guidelines. Many autistic youth not prescribed anti-psychotics are not getting testing for hyperlipidemia or diabetes. Providers should consider adding lipid and diabetes testing to other bloodwork, as this was positively associated with monitoring in this analysis.

BackgroundRecurrent urinary tract infection (rUTI) is a major diagnostic and clinical challenge. Dysregulated innate immune responses, including antimicrobial peptides and cytokines, may underlie UTI susceptibility. This study investigates whether urinary concentrations of antimicrobial peptides and cytokines differ in children and adolescents with a history of rUTI and whether they can accurately classify rUTI status.MethodsIn this single-center cross-sectional study, urine samples were collected from asymptomatic girls and adolescent females with a history of rUTI and age-matched controls recruited at Nationwide Children’s Hospital (Columbus, Ohio, USA). Concentrations of antimicrobial peptides (alpha-defensins 1-3, beta-defensin 1, cathelicidin, secretory leukocyte protease inhibitor, lipocalin 2, and ribonuclease 7) and cytokines (interleukin-1 beta, interleukin-6, interleukin-8, and tumor necrosis factor alpha) were quantified using enzyme-linked immunosorbent assays. A logistic regression model with variable selection was developed to classify rUTI participants based on urinary biomarkers and clinical factors.FindingsBetween 2019 and 2022, urine samples were analyzed from 42 participants with rUTI and 37 healthy controls without UTI. Compared to controls, participants with rUTI had lower concentrations of beta-defensin 1, cathelicidin, and ribonuclease 7, and higher concentrations of alpha-defensins 1-3, lipocalin 2, and secretory leukocyte protease inhibitor. Cytokine concentrations, including interleukin-1 beta, interleukin-6, interleukin-8, and tumor necrosis factor alpha, were elevated in the rUTI group. A multivariable classification model integrating urinary biomarkers with clinical features demonstrated high discriminatory performance with an area under the receiver operating curve of 0.97 and a prevalence-adjusted area under the precision-recall curve of 0.94.InterpretationGirls and adolescent females with rUTI exhibit a distinct urinary immune profile characterized by dysregulated antimicrobial peptides and elevated proinflammatory cytokines. A model integrating these biomarkers with clinical features accurately classified rUTI status, supporting their potential utility as biomarkers for identifying youth with rUTI.

Background and Objectives: Early identification of cardiac dysfunction in multi-system inflammatory syndrome in children (MIS-C) is crucial for effective management. Our primary objective was to predict left ventricular systolic dysfunction (LVSD) through a multicenter collaborative assessing admission laboratory data and echocardiogram findings. Methods: Laboratory and clinical data were collected by retrospective chart review from a cohort of pediatric patients admitted and treated for MIS-C in our institutions. Laboratory data including absolute lymphocyte count, albumin, sedimentation rate, C-reactive protein, procalcitonin, d-dimer, fibrinogen, ferritin, interleukin-6 level, and lymphocyte subsets (T, B and NK quantitation, TBNK) were collected. We built a LASSO logistic regression model to predict which MIS-C patients would have left ventricular systolic dysfunction LVSD using only laboratory data obtained within the first 24 h of admission. Results: Of the 1474 MIS-C patients evaluated, 297 had LVSD. The linear kinetic analysis found differences in albumin, lymphocyte count, C-reactive proteins and fibrinogen for systolic dysfunction patients, and of these C-reactive proteins, fibrinogen and procalcitonin were more predictive earlier. The best model for coronary artery abnormalities (CAAs) performed poorly, with a mean cross-validated AUC of 0.57. The model performed well with a cross-validated AUC of 0.845. Conclusions: This model identified widely available biomarkers to successfully predict systolic dysfunction in MIS-C patients. Those at high risk of systolic dysfunction had higher peak laboratory values for C-reactive protein, fibrinogen, and procalcitonin early on. A regularized logistic regression model was validated to provide excellent discrimination for LVSD.

Autistic adolescents and young adults (AYAs) are more likely to be obese than nonautistic peers. They are also more likely to be on psychiatric medications that can influence weight. The transition to adult health care is a challenge for autistic AYAs. The extent to which medications and transfer to adult care influence the weight of autistic youth is not well described. We assessed changes in body mass index (BMI) over time among a cohort of 216 autistic AYAs who had transferred from a pediatric hospital to an adult primary care clinic for autistic AYAs. We compared the mean BMI before and after transfer and used linear regression with main effects and age-related interaction terms to evaluate the contribution of transfer and psychiatric medications on BMI over time. We found that mean BMI increased over time and that age and transfer to adult care both appeared to be drivers of BMI increases. We also found that psychiatric medications had complex effects on weight over time. For example, medications for attention-deficit hyperactivity disorder were associated with a lower BMI at age 16, but showed interaction with age such that each year of increased age beyond 16 mitigated the BMI decrease at age 16. These findings highlight the importance of evaluating for changes in side effects of medicines over time, particularly during the transition to adult care and support other calls to address healthy eating and activity habits with autistic AYAs in adolescence.

Objective Autistic adolescents and young adults (AYAs) are more likely to be obese than nonautistic peers. They are also more likely to be on psychiatric medications that can influence weight. The transition to adult health care is a challenge for autistic AYAs. The extent to which medications and transfer to adult care influence the weight of autistic youth is not well described. Study design We assessed changes in body mass index (BMI) over time among a cohort of 216 autistic AYAs who had transferred from a pediatric hospital to an adult primary care clinic for autistic AYAs. We compared the mean BMI before and after transfer and used linear regression with main effects and age-related interaction terms to evaluate the contribution of transfer and psychiatric medications on BMI over time. Results We found that mean BMI increased over time and that age and transfer to adult care both appeared to be drivers of BMI increases. We also found that psychiatric medications had complex effects on weight over time. For example, medications for attention-deficit hyperactivity disorder were associated with a lower BMI at age 16, but showed interaction with age such that each year of increased age beyond 16 mitigated the BMI decrease at age 16. Conclusions These findings highlight the importance of evaluating for changes in side effects of medicines over time, particularly during the transition to adult care and support other calls to address healthy eating and activity habits with autistic AYAs in adolescence.

BACKGROUND:International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) codes have been shown to underestimate physical abuse prevalence. Machine learning models are capable of efficiently processing a wide variety of data and may provide better estimates of abuse.OBJECTIVE:To achieve proof of concept applying machine learning to identify codes associated with abuse.PARTICIPANTS AND SETTING:Children <5 years, presenting to the emergency department with an injury or abuse-specific ICD-10-CM code and evaluated by the child protection team (CPT) from 2016 to 2020 at a large Midwestern children's hospital.METHODS:The Pediatric Health Information System (PHIS) and the CPT administrative databases were used to identify the study sample and injury and abuse-specific ICD-10-CM codes. Subjects were divided into abused and non-abused groups based on the CPT's evaluation. A LASSO logistic regression model was constructed using ICD-10-CM codes and patient age to identify children likely to be diagnosed by the CPT as abused. Performance was evaluated using repeated cross-validation (CV) and Reciever Operator Characteristic curve.RESULTS:We identified 2028 patients evaluated by the CPT with 512 diagnosed as abused. Using diagnosis codes and patient age, our model was able to accurately identify patients with confirmed PA (mean CV AUC = 0.87). Performance was still weaker for patients without existing ICD codes for abuse (mean CV AUC = 0.81).CONCLUSIONS:We built a model that employs injury ICD-10-CM codes and age to improve accuracy of distinguishing abusive from non-abusive injuries. This pilot modelling endeavor is a steppingstone towards improving population-level estimates of abuse.

Purpose To develop a machine learning model to predict appointment no-shows among Orthopedic pediatric patients. Methods This retrospective cohort study utilized electronic health records from two large orthopedic centers in an urban pediatric hospital, spanning May 2021 to May 2023. A Lasso-regulated logistic regression model was developed to predict appointment no-shows and identify significant predictors. No-shows were defined as missed appointments or cancellations within 24 hours of the scheduled time. Candidate predictors included patient demographics, history of no-shows, and appointment characteristics. Model performance was evaluated using positive predictive value (PPV) and lift score. Adjusted odds ratios (AOR) with 95% confidence intervals (CI) for significant predictors were reported. Results The study included 54,467 appointment encounters from 21,455 patients, with a no-show rate of 17.4%. The Lasso model ranked appointments by no-show risk from the highest to the lowest. Interventions targeting the top 4% highest risk appointments can achieve a PPV of 80% and a lift score of 4.6. Significant predictors included private insurance (AOR: 0.08, 95% CI: 0.06-0.09), number of past no-shows (AOR: 1.06, 95% CI: 1.04-1.08), number of past late cancellations (AOR: 1.35, 95% CI: 1.31-1.38), and appointment sequential (AOR: 1.93, 95% CI: 1.70-2.20). Conclusions Targeted risk prediction using a machine learning model can help clinics focus limited intervention resources on high-risk patients to reduce appointment no-shows.

The transition from pediatric to adult health care is a vulnerable time period for autistic adolescents and young adults (AYA) and for some autistic AYA may include a period of receiving care in both the pediatric and adult health systems. We sought to assess the proportion of autistic AYA who continued to use pediatric health services after their first adult primary care appointment and to identify factors associated with continued pediatric contact. We analyzed electronic medical record (EMR) data from a cohort of autistic AYA seen in a primary-care-based program for autistic people. Using logistic and linear regression, we assessed the relationship between eight patient characteristics and (1) the odds of a patient having ANY pediatric visits after their first adult appointment and (2) the number of pediatric visits among those with at least one pediatric visit. The cohort included 230 autistic AYA, who were mostly white (68%), mostly male (82%), with a mean age of 19.4 years at the time of their last pediatric visit before entering adult care. The majority ( n = 149; 65%) had pediatric contact after the first adult visit. Younger age at the time of the first adult visit and more pediatric visits prior to the first adult visit were associated with continued pediatric contact. In this cohort of autistic AYA, most patients had contact with the pediatric system after their first adult primary care appointment.

Urinary tract infections (UTIs) commonly afflict people with diabetes. To better understand the mechanisms that predispose diabetics to UTIs, we employ diabetic mouse models and altered insulin signaling to show that insulin receptor (IR) shapes UTI defenses. Our findings are validated in human biosamples. We report that diabetic mice have suppressed IR expression and are more susceptible to UTIs caused by uropathogenic Escherichia coli (UPEC). Systemic IR inhibition increases UPEC susceptibility, while IR activation reduces UTIs. Localized IR deletion in bladder urothelium promotes UTI by increasing barrier permeability and suppressing antimicrobial peptides. Mechanistically, IR deletion reduces nuclear factor κB (NF-κB)-dependent programming that co-regulates urothelial tight junction integrity and antimicrobial peptides. Exfoliated urothelial cells or urine samples from diabetic youths show suppressed expression of IR, barrier genes, and antimicrobial peptides. These observations demonstrate that urothelial insulin signaling has a role in UTI prevention and link IR to urothelial barrier maintenance and antimicrobial peptide expression.

BACKGROUND Patient-generated health data (PGHD) captured via smart devices or digital health technologies can reflect an individual health journey. PGHD enables tracking and monitoring of personal health conditions, symptoms, and medications out of the clinic, which is crucial for self-care and shared clinical decisions. In addition to self-reported measures and structured PGHD (eg, self-screening, sensor-based biometric data), free-text and unstructured PGHD (eg, patient care note, medical diary) can provide a broader view of a patient’s journey and health condition. Natural language processing (NLP) is used to process and analyze unstructured data to create meaningful summaries and insights, showing promise to improve the utilization of PGHD. OBJECTIVE Our aim is to understand and demonstrate the feasibility of an NLP pipeline to extract medication and symptom information from real-world patient and caregiver data. METHODS We report a secondary data analysis, using a data set collected from 24 parents of children with special health care needs (CSHCN) who were recruited via a nonrandom sampling approach. Participants used a voice-interactive app for 2 weeks, generating free-text patient notes (audio transcription or text entry). We built an NLP pipeline using a zero-shot approach (adaptive to low-resource settings). We used named entity recognition (NER) and medical ontologies (RXNorm and SNOMED CT [Systematized Nomenclature of Medicine Clinical Terms]) to identify medication and symptoms. Sentence-level dependency parse trees and part-of-speech tags were used to extract additional entity information using the syntactic properties of a note. We assessed the data; evaluated the pipeline with the patient notes; and reported the precision, recall, and F1 scores. RESULTS In total, 87 patient notes are included (audio transcriptions n=78 and text entries n=9) from 24 parents who have at least one CSHCN. The participants were between the ages of 26 and 59 years. The majority were White (n=22, 92%), had more than one child (n=16, 67%), lived in Ohio (n=22, 92%), had mid- or upper-mid household income (n=15, 62.5%), and had higher level education (n=24, 58%). Out of 87 notes, 30 were drug and medication related, and 46 were symptom related. We captured medication instances (medication, unit, quantity, and date) and symptoms satisfactorily (precision >0.65, recall >0.77, F1>0.72). These results indicate the potential when using NER and dependency parsing through an NLP pipeline on information extraction from unstructured PGHD. CONCLUSIONS The proposed NLP pipeline was found to be feasible for use with real-world unstructured PGHD to accomplish medication and symptom extraction. Unstructured PGHD can be leveraged to inform clinical decision-making, remote monitoring, and self-care including medical adherence and chronic disease management. With customizable information extraction methods using NER and medical ontologies, NLP models can feasibly extract a broad range of clinical information from unstructured PGHD in low-resource settings (eg, a limited number of patient notes or training data).

Background The transition from pediatric to adult care is a challenge for autistic adolescents and young adults. Data on patient features associated with timely transfer between pediatric and adult health care are limited. Our objective was to describe the patient features associated with timely transfer to adult health care (defined as </ = 6 months between first adult visit and most recent prior pediatric visit) among a cohort of autistic adolescents and young adults. Methods and findings We analyzed pediatric and adult electronic medical record data from a cohort of adolescents and young adults who established with a primary-care based program for autistic adolescents and young adults after they transferred from a single children’s hospital. Using forward feature selection and logistic regression, we selected an optimal subset of patient characteristics or features via five repetitions of five-fold cross validation over varying time-frames prior to the first adult visit to identify patient features associated with a timely transfer to adult health care. A total of 224 autistic adolescents and young adults were included. Across all models, total outpatient encounters and total encounters, which are very correlated (r = 0.997), were selected as the first variable in 91.2% the models. These variables predicted timely transfer well, with an area under the receiver-operator curve ranging from 0.81 to 0.88. Conclusions Total outpatient encounters and total encounters in pediatric care showed good ability to predict timely transfer to adult health care in a population of autistic adolescents and young adults.

Objective: To develop and validate a predictive algorithm that identifies pediatric patients at risk of asthma-related emergencies, and to test whether algorithm performance can be improved in an external site via local retraining.Methods: In a retrospective cohort at the first site, data from 26 008 patients with asthma aged 2-18 years (2012-2017) were used to develop a lasso-regularized logistic regression model predicting emergency department visits for asthma within one year of a primary care encounter, known as the Asthma Emergency Risk (AER) score. Internal validation was conducted on 8634 patient encounters from 2018. External validation of the AER score was conducted using 1313 pediatric patient encounters from a second site during 2018. The AER score components were then reweighted using logistic regression using data from the second site to improve local model performance. Prediction intervals (PI) were constructed via 10 000 bootstrapped samples.Results: At the first site, the AER score had a cross-validated area under the receiver operating characteristic curve (AUROC) of 0.768 (95% PI: 0.745-0.790) during model training and an AUROC of 0.769 in the 2018 internal validation dataset (p = 0.959). When applied without modification to the second site, the AER score had an AUROC of 0.684 (95% PI: 0.624-0.742). After local refitting, the cross-validated AUROC improved to 0.737 (95% PI: 0.676-0.794; p = 0.037 as compared to initial AUROC).Conclusions: The AER score demonstrated strong internal validity, but external validity was dependent on reweighting model components to reflect local data characteristics at the external site.

OBJECTIVES:Develop and deploy a disease cohort-based machine learning algorithm for timely identification of hospitalized pediatric patients at risk for clinical deterioration that outperforms our existing situational awareness program. DESIGN:Retrospective cohort study. SETTING:Nationwide Children's Hospital, a freestanding, quaternary-care, academic children's hospital in Columbus, OH. PATIENTS:All patients admitted to inpatient units participating in the preexisting situational awareness program from October 20, 2015, to December 31, 2019, excluding patients over 18 years old at admission and those with a neonatal ICU stay during their hospitalization. INTERVENTIONS:We developed separate algorithms for cardiac, malignancy, and general cohorts via lasso-regularized logistic regression. Candidate model predictors included vital signs, supplemental oxygen, nursing assessments, early warning scores, diagnoses, lab results, and situational awareness criteria. Model performance was characterized in clinical terms and compared with our previous situational awareness program based on a novel retrospective validation approach. Simulations with frontline staff, prior to clinical implementation, informed user experience and refined interdisciplinary workflows. Model implementation was piloted on cardiology and hospital medicine units in early 2021. MEASUREMENTS AND MAIN RESULTS:The Deterioration Risk Index (DRI) was 2.4 times as sensitive as our existing situational awareness program (sensitivities of 53% and 22%, respectively; p < 0.001) and required 2.3 times fewer alarms per detected event (121 DRI alarms per detected event vs 276 for existing program). Notable improvements were a four-fold sensitivity gain for the cardiac diagnostic cohort (73% vs 18%; p < 0.001) and a three-fold gain (81% vs 27%; p < 0.001) for the malignancy diagnostic cohort. Postimplementation pilot results over 18 months revealed a 77% reduction in deterioration events (three events observed vs 13.1 expected, p = 0.001). CONCLUSIONS:The etiology of pediatric inpatient deterioration requires acknowledgement of the unique pathophysiology among cardiology and oncology patients. Selection and weighting of diverse candidate risk factors via machine learning can produce a more sensitive early warning system for clinical deterioration. Leveraging preexisting situational awareness platforms and accounting for operational impacts of model implementation are key aspects to successful bedside translation.

Background: Child physical abuse (PA) is a significant societal concern with limited research into predictors of re-reports.Objective: Our research explores correlations between sociodemographic variables and re-reported PA. Our aim was to characterize populations at higher risk and identify changes in presentation during the COVID-19 pandemic. Participants and setting: This retrospective descriptive study focused on 238 patients with re-reports of PA made by a pediatric hospital from January 2019 through April 2021.Methods: We analyzed sociodemographic information and details of reports made to child protective services (CPS) obtained from the electronic health record.Results: Females were 2.5 years older than males (mean 11.0 and 8.5 years, respectively) (p < .001, 95%CI 1.21-3.76). Males were more likely to have observable injuries (OR 2.61, p < .001) and a CPS response (OR = 2.70, p = .003). Patients categorized as "Other" races were less likely to have observable injuries (OR = 0.32, p = .006). Presentation changed during the pandemic: a quadrupling of re-reports by behavioral health clinicians caused the percentage of reports made by them to increase significantly (OR = 3.46, p < .001) and the mean age increased by 2.0 years (8.2 years before, 10.2 years during) (p = .009, 95%CI 0.5-3.5), though females remained approximately 2.2 years older than males (p = .003, 95%CI 0.8-3.7).Conclusions: Males experienced higher rates of re-reported PA and were younger at the time of re-report. Changes to presentation during the pandemic suggest an increase in PA among older children. Future research should further explore differences in sex/race, while current prevention efforts should focus on children receiving behavioral health care.

Background Many of the benefits of electronic health records (EHRs) have not been achieved at expected levels because of a variety of unintended negative consequences such as documentation burden. Previous studies have characterized EHR use during and outside work hours, with many reporting that physicians spend considerable time on documentation-related tasks. These studies characterized EHR use during and outside work hours using clock time versus actual physician clinic schedules to define the outside work time. Objective This study aimed to characterize EHR work outside scheduled clinic hours among primary care pediatricians using a retrospective descriptive task analysis of EHR access log data and actual physician clinic schedules to define work time. Methods We conducted a retrospective, exploratory, descriptive task analysis of EHR access log data from primary care pediatricians in September 2019 at a large Midwestern pediatric health center to quantify and identify actions completed outside scheduled clinic hours. Mixed-effects statistical modeling was used to investigate the effects of age, sex, clinical full-time equivalent status, and EHR work during scheduled clinic hours on the use of EHRs outside scheduled clinic hours. Results Primary care pediatricians (n=56) in this study generated 1,523,872 access log data points (across 1069 physician workdays) and spent an average of 4.4 (SD 2.0) hours and 0.8 (SD 0.8) hours per physician per workday engaged in EHRs during and outside scheduled clinic hours, respectively. Approximately three-quarters of the time working in EHR during or outside scheduled clinic hours was spent reviewing data and reports. Mixed-effects regression revealed no associations of age, sex, or clinical full-time equivalent status with EHR use during or outside scheduled clinic hours. Conclusions For every hour primary care pediatricians spent engaged with the EHR during scheduled clinic hours, they spent approximately 10 minutes interacting with the EHR outside scheduled clinic hours. Most of their time (during and outside scheduled clinic hours) was spent reviewing data, records, and other information in EHR.

Patient-generated health data (PGHD) is becoming a necessary component of remote monitoring and clinical decision-making as well as self-care and managing chronic conditions. Medical diaries and patient notes have been the primary source of unstructured PGHD, enabling patients to collect and communicate their health information remotely and in real-time to their providers. Yet there is no established mechanism or pipeline for processing free text patient notes (“unstructured PGHD”) or integrating these notes into clinical decision making in a low-resource setting (i.e., not depending on a large dataset or processing power). In the literature, there are a number of studies reporting NLP applications on clinical notes to identify symptoms and conditions, but these studies have been limited to public data (e.g. social media posts by patients). In this paper, we evaluate the performance of an hybrid (deep learning + rule-based) NLP pipeline on a low-resource PHGD dataset through an empirical evaluation of automatic component extraction where we measure the model’s ability to conduct automatic entity extraction using ontologies (medication-dose: RxNORM, symptoms: SNOMED) from patient notes.