Up-to-date information about non-fatal overdose emergency department (ED) encounters can provide critical information about the evolution of the opioid epidemic and response of the healthcare system and is an essential to planning of clinical trials to address the problems underlying this epidemic. Establishing a platform that delivers high positive predictive value for opioid-related overdose (OD) using a combination of coded and natural language terms in electronic health records (EHRs) is an essential step toward the large-scale surveillance necessary to evaluate the pragmatic effectiveness of numerous systemic and policy-based efforts and to create the infrastructure for large scale trials to reduce drug-related mortality and morbidity. However, to date, localized and federal efforts have been largely based on discrete ICD-10 code data or have had time lags of one-to-three years for more detailed data. Perhaps more Importantly, they have not had the capacity for planning and feasibility assessment for clinical and translational research at specific sites. We propose foundational work to create an inter-institutional research database and network focused on patients presenting to EDs with opioid-related OD. To create this network, we will extend previous work to develop: (1) an e-phenotype for case identification in the ED based on EHR data, and (2) combine this with a data dictionary and coded data extraction tools, and natural language processing (NLP) algorithms, to obtain additional data from EHRs; tools for primary capture of data during clinical care; and tools for integration of data on social determinants of health. This will allow for a more thorough characterization of individuals presenting to EDs with opioid-related OD including: demographics, comorbidities, OD agent and source, intentionality of the OD, ED treatment and discharge disposition. Through refinement and automation, the data extraction process will be extended to a set of pilot CTSA Accrual to Clinical Trials (ACT) Network sites and then potentially to other CTSA's nation-wide. This new functionality focuses on providing a platform to accelerate research. To accomplish this work, our Specific Aims are to: 1) Demonstrate the feasibility of extending the ACT Network data model and infrastructure to monitor the opioid epidemic using ED data, 2) Create a prototype opioid overdose monitoring and response network across participating institutions and a toolkit for other CTSA sites to join the network, and 3) Demonstrate potential usefulness of the network in monitoring the opioid epidemic and in planning clinical trials. This proposal is innovative as it aims to develop a feasible and effective near real time means of monitoring opioid-related OD presentation in EDs across the country to inform point-of-care service delivery, prevention and treatment intervention development and evaluation. Downstream, the application of project deliverables include dissemination of the replication toolkit to leverage the platform of the CTSAs and NIDA Clinical Trials Network to build capacity for timely surveillance of opioid-related OD to facilitate prevention and treatment research and intervention.
Patients presenting with opioid-related overdose to hospital emergency departments represent a high-risk group for morbidity and mortality and a potential target for interventions to combat the opioid epidemic. This proposal is innovative as it aims to develop a feasible and effective inter-institutional research database and network focused on these patients. These tools will inform point-of-care service delivery, as well as enhance nationwide prevention and treatment interventions.
