Developing and Validating a Kaiser Permanente Electronic Health Record Data Transfer to External Research Partner Using Software Designed for Health Data Interoperability
technology adoption and diffusion, information technology, virtual data warehouse, technology assessment
Background: Chronic obstructive pulmonary disease (COPD) affects over 12 million Americans, is responsible for 800,000 hospitalizations each year and is the third leading cause of death in the United States. The continuing lag between clinical practice and treatment makes COPD ripe for patient-centered outcomes research. The COPD Foundation’s PCORnet-funded COPD Patient Powered Research Network (COPDPPRN) created a COPD research repository of > 5,000 participants; 346 of these are represented in Kaiser Permanente’s PCORnet PORTAL database and in the COPD Outcomes-based Network for Clinical and Effectiveness Translation (CONCERT). Our objective is to develop and validate a scalable methodology for linking and sharing consented patients’ comprehensive electronic health record (EHR) data across research data partners.
Methods: A subdatabase containing the EHR data for the subset of COPDPPRN-consented PORTAL patients can be accessed via a RESTful application program interface (API) developed by Kaiser Permanente. Utilizing Corepoint Integration Engine to manage authenticated communication, linkage is verified and data securely transferred to a DatStat shell at COPDPPRN. To validate, the dataset will be generated at DatStat to compare to the CONCERT EHR data as gold standard. Validation includes data accuracy/integrity and identifies potential organizational/systematic/technical errors made during transfer.
Results: Using PCORnet’s Common Data Model structure, we populated a subdatabase with COPDPPRN-consented CONCERT patients’ EHR data. Kaiser Permanente and DatStat built APIs to provide secure access to the data using the Corepoint technology. The API, called via a secure HTTPS web request, utilizes multifactor authentication. Upon validation of the client credentials and existence of the requested data, the API returns the participant’s data in encrypted XML format. Corepoint Health maintains the private key to decrypt the XML data and transfer to DatStat. Data transfers are scheduled, and validation will be completed by early 2017. Report on the preliminary results is planned for presentation at the Health Care Systems Research Network 2017 Conference.
Conclusion: This work is the first step toward creating a fully functional, scalable system for linking patient data across sources in the PCORnet enterprise to enhance PPRN data for research. Once the API is tested and validated, it can be adapted to access EHR data across PCORnet partners.
Gillespie S, McBurnie MA, Malanga V, Kirk K, Thornton M, Crawford P, Hu W, Petrie R, Stockemer M, Davis K, Mularski R. Developing and validating a Kaiser Permanente electronic health record data transfer to external research partner using software designed for health data interoperability, J Patient Cent Res Rev. 2017;4:173-4.