Racial and ethnic disparities in common inpatient safety outcomes in a children's hospital cohort.

BACKGROUND: Emerging evidence has shown racial and ethnic disparities in rates of harm for hospitalised children. Previous work has also demonstrated how highly heterogeneous approaches to collection of race and ethnicity data pose challenges to population-level analyses. This work aims to both create an approach to aggregating safety data from multiple hospitals by race and ethnicity and apply the approach to the examination of potential disparities in high-frequency harm conditions. METHODS: In this cross-sectional, multicentre study, a cohort of hospitals from the Solutions for Patient Safety network with varying race and ethnicity data collection systems submitted validated central line-associated bloodstream infection (CLABSI) and unplanned extubation (UE) data stratified by patient race and ethnicity categories. Data were submitted using a crosswalk created by the study team that reconciled varying approaches to race and ethnicity data collection by participating hospitals. Harm rates for race and ethnicity categories were compared with reference values reflective of the cohort and broader children's hospital population. RESULTS: Racial and ethnic disparities were identified in both harm types. Multiracial Hispanic, Combined Hispanic and Native Hawaiian or other Pacific Islander patients had CLABSI rates of 2.6-3.6 SD above reference values. For Black or African American patients, UE rates were 3.2-4.4 SD higher. Rates of both events in White patients were significantly lower than reference values. CONCLUSIONS: The combination of harm data across hospitals with varying race and ethnicity collection systems was accomplished through iterative development of a race and ethnicity category framework. We identified racial and ethnic disparities in CLABSI and UE that can be addressed in future improvement work by identifying and modifying care delivery factors that contribute to safety disparities.

Reducing Patient Length of Stay After Surgical Correction for Neuromuscular Scoliosis.

BACKGROUND: Surgical correction remains the most effective treatment of patients with severe neuromuscular scoliosis (NMS). However, postoperative inpatient clinical care after spinal surgery for NMS patients is highly complex and variable, depending on the provider and/or institution. Our project used a quality improvement methodology to implement a standardized clinical care pathway for NMS patients during their primary spinal surgery, with the global aim of decreasing postoperative hospitalized length of stay (LOS). METHODS: A multidisciplinary team developed a postoperative pathway for nonambulatory patients with NMS after their primary, nonstaged corrective spinal surgery. The outcome measure was mean LOS, and the process measure was percentage compliance with the newly implemented NMS pathway. The team used statistical process control charts and measures were evaluated over 3 phases to assess the impact of the interventions over time: pretesting, testing, and sustain. RESULTS: The mean LOS decreased from 8.0 to 5.3 days; a statistically significant change based on statistical process control chart rules. Percentage compliance with the NMS pathway improved during the testing and sustain phases, compared with the pretesting phase. LOS variability decreased from pretesting to the combined testing and sustain phases. CONCLUSIONS: Using quality improvement methodology, we successfully standardized major components of clinical care for patients with NMS after spinal corrective surgery. Through enhanced care consistency, the mean postoperative LOS decreased by nearly 3 days and decreased LOS variability. Robust multidisciplinary involvement with key stakeholders ensured sustainability of these improvements.

Always-On Sub-Microwatt Spiking Neural Network Based on Spike-Driven Clock- and Power-Gating for an Ultra-Low-Power Intelligent Device.

This paper presents a novel spiking neural network (SNN) classifier architecture for enabling always-on artificial intelligent (AI) functions, such as keyword spotting (KWS) and visual wake-up, in ultra-low-power internet-of-things (IoT) devices. Such always-on hardware tends to dominate the power efficiency of an IoT device and therefore it is paramount to minimize its power dissipation. A key observation is that the input signal to always-on hardware is typically sparse in time. This is a great opportunity that a SNN classifier can leverage because the switching activity and the power consumption of SNN hardware can scale with spike rate. To leverage this scalability, the proposed SNN classifier architecture employs event-driven architecture, especially fine-grained clock generation and gating and fine-grained power gating, to obtain very low static power dissipation. The prototype is fabricated in 65 nm CMOS and occupies an area of 1.99 mm2. At 0.52 V supply voltage, it consumes 75 nW at no input activity and less than 300 nW at 100% input activity. It still maintains competitive inference accuracy for KWS and other always-on classification workloads. The prototype achieved a power consumption reduction of over three orders of magnitude compared to the state-of-the-art for SNN hardware and of about 2.3X compared to the state-of-the-art KWS hardware.

The effect of age, sex, race/ethnicity, health insurance, and food specific serum immunoglobulin E on outcomes of oral food challenges.

Background: Although oral food challenge (OFC) is an important clinical procedure for diagnosing food allergy, there is a paucity of literature on the outcome of the procedure and specifically the patients on whom the procedure is performed from the aspects of their age, sex, race/ethnicity, health insurance status, and serum specific IgE to the food tested. Objective: We aimed to review results of OFC and determine the impact of patient age, sex, race/ethnicity, insurance status, private or public, and food specific serum IgE on the outcome of OFC. Methods: A retrospective chart review was performed of patients undergoing OFCs at a children's hospital outpatient allergy clinic over a two-year period. The outcome of OFC was allergic or non-allergic based on determination and documentation by the treating physician. A logistic regression model was built to determine the association between the OFC outcomes, age, and symptoms at the time of OFC. A Chi-square analysis was performed to check for any significant relationship between the OFC outcome and age when stratified by insurance status. Results: Five hundred and eight children underwent 641 OFCs. Twenty nine percent of OFCs had an allergic outcome with the most commonly challenged foods being peanuts, eggs, and milk. Patient age and gender, when stratified by insurance status, did not have a significant effect on OFC outcomes. Serum IgE to peanuts and egg was significantly different between allergic OFC and non-allergic outcome. Vomiting and urticaria/angioedema correlated with an allergic OFC outcome. Conclusion: OFCs confirm the food allergy diagnosis in about one-third of patients tested, and they should continue to be used when possible for an accurate diagnosis. Age, sex, and insurance status do not have a significant association with the outcome of OFC and cannot be added as predictive factors.

Depression Screening of Adolescents With Diabetes: 5-Years of Implementation and Outcomes.

As part of a quality improvement (QI) initiative, Cincinnati Children's Hospital Diabetes Center established a model of reliable, feasible, routine depression screening in a subspecialty clinic for adolescents with diabetes.1 On the occasion of the special section of the journal focusing on screening for, detection of, and treatment of depression in diverse settings, the purpose of this correspondence is to describe the evolution of an in-clinic, pragmatic approach to depression and suicidal ideation (SI) screening, and to evaluate depression screening scores and glycemic control across age groups over time.

Integrating Psychological Screening Into Medical Care for Youth With Abdominal Pain.

BACKGROUND: Pediatric functional abdominal pain disorders are common, costly, and disabling. Clinical anxiety is highly prevalent and is associated with increased pain and functional disability. Thus, a psychological screening process is recommended but is infrequently used in current practice. METHODS: A screening process for patient-reported anxiety (Screen for Child Anxiety and Related Disorders), functional disability (Functional Disability Inventory), and pain levels was implemented in a large gastroenterology division within a major medical center. Quality improvement methods and traditional analytic approaches were used to test the feasibility and outcomes of routine screening in patients ages 8 to 18 with abdominal pain. RESULTS: Screening rates increased from <1% to >80%. A total of 1291 patients who reported having abdominal pain completed the screening during the first 6 months. Clinically significant anxiety (43.1%), at least moderate disability (45%), and elevated pain (61.5%) were common in children with abdominal pain. The presence of clinically significant anxiety corresponded with higher pain and pain-related disability. Twenty-one percent of youth had clinical elevations in all 3 areas. In such instances, medical providers received an automated prompt to tailor care, including to consider a psychological referral. After the project implementation, psychological referral rates increased from 8.3 per 1000 patients to 15.2 per 1000 patients. CONCLUSIONS: Systematic screening for anxiety, pain, and pain-related disability as a routine part of medical care can be reliably implemented with clinically meaningful results. Future directions include examining the role of anxiety over the long-term and reducing clinician burden.

Association of an Asthma Improvement Collaborative With Health Care Utilization in Medicaid-Insured Pediatric Patients in an Urban Community.

Importance: Asthma is the most common chronic condition of childhood. Hospitalizations and emergency department (ED) visits for asthma are more frequently experienced by minority children and adolescents and those with low socioeconomic status. Objective: To reduce asthma-related hospitalizations and ED visits for Medicaid-insured pediatric patients residing in Hamilton County, Ohio. Design, Setting, and Participants: From January 1, 2010, through December 31, 2015, a multidisciplinary team used quality-improvement methods and the chronic care model to conduct interventions in inpatient, outpatient, and community settings in a large, urban academic pediatric hospital in Hamilton County, Ohio. Children and adolescents aged 2 to 17 years who resided in Hamilton County, had a diagnosis of asthma, and were Medicaid insured were studied. Interventions: Interventions were implemented in 3 phases: hospital-based inpatient care redesign, outpatient-based care enhancements, and community-based supports. Plan-do-study-act cycles allowed for small-scale implementation of change concepts and rapid evaluation of how such tests affected processes and outcomes of interest. Main Outcomes and Measures: The study measured asthma-related hospitalizations and ED visits per 10 000 Medicaid-insured pediatric patients. Data were measured monthly on a rolling 12-month mean basis. Data from multiple previous years were used to establish a baseline. Data were tracked with annotated control charts and with interrupted time-series analysis. Results: Of the estimated 36 000 children and adolescents with asthma in Hamilton County (approximately 13 000 of whom are Medicaid insured and 6000 of whom are cared for in Cincinnati Children's Hospital primary care practices), asthma-related hospitalizations decreased from 8.1 (95% CI, 7.7-8.5) to 4.7 (95% CI, 4.3-5.1) per 10 000 Medicaid patients per month by June 30, 2014, a 41.8% (95% CI, 41.7%-42.0%) relative reduction. Emergency department visits decreased from 21.5 (95% CI, 20.6-22.3) to 12.4 (95% CI, 11.5-13.2) per 10 000 Medicaid patients per month by June 30, 2014, a 42.4% (95% CI, 42.2%-42.6%) relative reduction. Improvements were sustained for the subsequent 12 months. The proportion of patients who were rehospitalized or had a return ED visit for asthma within 30 days of an index hospitalization was reduced from 12% to 7%. The proportion of patients with documented well-controlled asthma in this study's primary care population increased from 48% to 54%. Conclusions and Relevance: An integrated, multilevel approach focused on enhancing availability and accessibility of treatments, removing barriers to adherence, mitigating multidomain risks, augmenting self-management, and creating a collaborative relationship between the family and the health care system was associated with improved asthma outcomes for a population of Medicaid-insured pediatric patients. Similar models used in accountable care organizations or across patient panels and with other chronic conditions could be feasible and warrant evaluation.

Psychosocial Patient-Reported Outcomes in Pediatric and Adolescent Diabetes: a Review and Case Example.

PURPOSE OF REVIEW: The purpose of this review is to define psychosocial patient-reported outcomes (PROs) relevant to pediatric and adolescent diabetes populations. Potential domains for PROs include a spectrum of emotional, behavioral, social, physical, overall health, and/or care management areas. A literature review of potential PRO measures, selection criteria, and implementation strategies including a case example will be presented. RECENT FINDINGS: Among the pediatric, adolescent, and emerging adult populations, research indicates a relative higher risk for distress, depression, anxiety, and eating disorders as compared to peers without diabetes. Use of PRO measures can expand providers' focus beyond glycemic control, or simply hemoglobin A1c, to better appreciate the impact of diabetes on the whole child/adolescent, and provide services that address patients' individually identified needs, which are most salient to them. Successful selection and implementation of psychosocial PRO measures should be designed to include pathways for real-time provider interaction with the patient and respective PRO data to guide clinical care.