Moving Pediatric Complex Care Forward: Big Data and National Research Collaborations.

Abstract

The growing population of children with medical complexity (CMC), especially those with neurologic impairment (NI), is increasingly the focus of researchers, clinician leaders, and policy makers.1 In this issue of Pediatrics, Mpody et al2 add to this emerging literature base by clearly demonstrating that NI, a common health condition among CMC, is strongly associated with postoperative pneumonia. Using data from the National Surgical Quality Improvement Program (NSQIP), the authors confirm what most pediatricians learn during training: compared with children without underlying neurologic comorbidities, children with NI are at significantly higher risk for developing pneumonia and experiencing prolonged hospitalizations after a surgical procedure.3This study raises important questions, some of which are noted by the authors. For example, which specific clinical characteristics put children with cerebral palsy or neuromuscular disorders at increased risk for postoperative pneumonia, and which ones are protective? Among children with NI, which surgical procedures are associated with the highest risk of postoperative complications? How do other postoperative outcome metrics (eg, 30-day rehospitalization rate) differ by NI status? Answering these questions will improve clinical decision-making and clinicians’ ability to engage in shared decision-making with families of CMC and NI.Importantly, this study demonstrates how big data can improve care and highlights the need to increase efforts in organizing national data collection collaborations. In addition to the NSQIP, other large datasets, such as PEDSnet or PediQUEST, that draw from electronic health records and patient- and family-generated data sources across multiple locations enable targeted examinations of pediatric subpopulations often limited by small sample sizes at any one institution.4,5 However, despite significant advancements in informational technology over the last 2 decades, data collection infrastructure is not fully integrated into health systems serving CMC and NI.6 Additionally, challenges in appropriately classifying children by diagnoses in large datasets remain. As the authors note, many children in the NSQIP database have conditions in multiple NI categories (eg, a child may have both cerebral palsy and seizure disorder because of a remote hypoxic-ischemic event), making it difficult to ascertain which is truly driving increased risk of postoperative pneumonia. Even with expert chart extractors, this is a limitation of such datasets and points to the need for significant investment to optimize their utility.7Mpody et al2 discuss identifying effective postoperative pneumonia prevention strategies, which illuminates potential future research directions yet also has broader implications. The authors review the topic of perioperative oral care, a low-cost and low-burden intervention that clearly deserves further evaluation. Alternatively, examining pre- and perioperative respiratory physiotherapy may be considered given successes with this resource in adult populations across a wide range of surgical procedures.8 One could also justify studying adaptation of Enhanced Recovery After Surgery protocols (multimodal evidence-based approaches for the care of adult surgical patients), not to mention comparing different anesthesia regimens (eg, intravenous versus inhalation) given the risk of interactions between anesthetic agents and the multitude of central nervous system–acting medications children with NI are often prescribed.9,10There is a paucity of clinically relevant interventional research focused on the CMC and NI population.11 Unlike the body of literature supporting the patient- and family-centered medical home and pediatric complex care centers, the evidence guiding day-to-day clinical management of CMC with NI is woefully insufficient. In part, this is due to challenges inherent to all pediatric research, but it also reflects barriers unique to children with complex chronic health conditions. Not only are population sizes small, but for any particular condition, there exists high levels of heterogeneity regarding disease severity, stability, duration of an illness experience, development, and functioning. Multicenter complex care research collaborations to address these barriers are rare and need to be fostered to improve clinical management.As we recognize these challenges, the pipeline for developing pediatric complex care investigators must be expanded. Early research opportunities should be available to students and trainees in all relevant disciplines. Academic institutions and national funding agencies should increase early career development support (eg, postdoctoral research mechanisms, career development awards) of investigators whose scholarship is focused on topics directly relevant and impactful to the health and well-being of CMC and NI. Secondly, pediatric health care systems (ie, children’s hospitals) should ensure clinicians have the time and tools to collaborate in and contribute to research and quality initiatives.12 Lastly, current investigators should familiarize themselves and employ advanced methodologies in clinical informatics (eg, incorporation of machine learning into clinical decision-making) and interventional research (eg, Bayesian decision theory, adaptive trial designs, propensity score analyses).13–16Although significant progress in pediatric complex care has been made over the past decade, the field still faces enormous questions and challenges in its quest to improve the health and well-being of all CMC and NI. We hope studies such as this one by Mpody et al2 inspire more investigators and clinicians to focus their talent and expertise on pediatric complex care medicine.