Electronic Medical Record (EMR) data represents a promising, but under-utilized, resource for evaluating interventions that focus on devastating pediatric health events such as neurologic injuries and illnesses. However, a limitation to using existing data is the lack of universally used, standardized, meaningful outcome measures within the datasets. The overall objective of this project is to demonstrate feasibility and proof of concept for the development of a Pediatric Functional Status Score (PFSS) that uses clinically relevant claims data from the EMR (i.e., diagnoses, procedures, pharmaceuticals, and durable goods) to accurately represent a child?s functional mobility, self-care, and cognitive/communication status. The hypothesis is that the constellation of claims data associated with a particular point in a child?s clinical course is an accurate reflection of the child?s functional status at that time. The rationale for this research is that a functional status score based on readily available, high-quality big data sources will allow researchers to compare the effectiveness of pediatric rehabilitation interventions. Such a tool is necessary to give researchers the ability to quickly and accurately assess functional status at any time that includes child-specific claims data. Using consensus methodology (Delphi method and nominal group technique), this study will engage ten expert pediatric neurology, trauma, and rehabilitation clinicians to determine which billing codes are clinically relevant to a child?s functional mobility, self-care, and cognitive/communication status (Aim 1). Through the creation of predictive models, researchers will then determine the feasibility of using identified billing codes to develop a PFSS that accurately represents a child?s functional status in these domains at the time of discharge from inpatient rehabilitation (Aim 2). The PFSS will be calibrated against the Functional Independence Measure for Children (WeeFIM?), a validated, but not universally, used outcome measure.
Aim 2 will incorporate retrospective data from the EMR for the 1,797 children (6 months to 18 years) with acute neurologic injuries or illnesses admitted to the Inpatient Rehabilitation Unit at large pediatric quaternary center between 2000 and 2019. The results of this work will provide the necessary proof of concept for a subsequent R01 submission wherein we will fully develop and validate the PFSS in a multi-site proposal. Ultimately, a PFSS that relies on highly conserved, child-specific claims data is significant because it provides a transformative tool that researchers can use to harness big data to determine the effectiveness of interventions for acute neurologic injury or illness at any time that has associated claims data and without additional burden to patients or clinicians. The innovation of our design mitigates current challenges to the use of big data in pediatric rehabilitation research by creating a clinically meaningful PFSS based on child-specific claims data that investigators can use immediately to study interventions for children with acute neurologic injuries and illnesses.
Our current ability to use big data in pediatric rehabilitation research is limited by the lack of a universal outcome measure within existing datasets. This study seeks to use claims data to create a child-specific functional status score for mobility, self-care, and cognition/communication, filling a critical gap in our ability to compare interventions for children with neurologic injuries or illnesses. The creation of such a tool would enable a scalable and cost-effective method to use large, widely available data sources to study the functional outcomes of these children, thereby opening new horizons for comparative effectiveness research to identify interventions that improve long-term outcomes and quality-of-life.
