The overarching object of our BD2K parent award: The overarching goal of this proposal is to develop, test, and deploy Office of the National Coordinator (ONC) national standards to support EHR data repurposing and interoperability for Alzheimer's Disease (AD) and related dementias (ADRD). AD, previously characterized by clinically observed neurodegenerative changes, is now defined as a biological diagnosis ??with or without clinically observed syndromic symptoms. The recent updates to the National Institute on Aging-Alzheimer's Association (NIA-AA) describe a research framework on which to base future research into disease etiology and intervention. The NIA-AA notes the need for a common language on which to share and harmonize disparate research endeavors into AD. In response, this supplemental proposal will provide the computational, semantic medical terminology infrastructure to support the representation of clinical-pathologic data collected in vivo and post-mortem on patients on the AD spectrum. These goals will be achieved through three Specific Aims (SA): 1) data element inventory and terminology development; 2) chart abstraction and terminology binding; and 3) database populations and query validation. In SA1, data elements currently collected and/or to be collected for AD care and diagnosis will be identified and inventoried. Data elements will include pathology data recorded during autopsy, laboratory data used during patient diagnoses/care, and PET-scan data pertaining to protein expression. Neurologists, neuropathologists and expert terminologists will collaborate to define each data element in the level of detail necessary and sufficiently to represent the clinical concepts in SNOMED CT. The terminology will then be created in the Nebraska SNOMED CT terminology development environment. In SA2, the terminology developed in SA1 will be applied to data contained in deceased patient records including autopsy reports, laboratory data and imaging reports. Abstracted encoded data elements will be transferred to the UNMC CDW and paraffin-embedded biorepository registry. In SA3, these databases will be interrogated with plausible, scientifically useful queries to validate the utility and fitness for purpose of the encoded data. The success of this supplemental project will advance the ability to capture and represent clinical data about AD and expose data to computational methods for research and discovery. Such efforts meet the goals of our parent award, U01 Big Data to Knowledge award: Deploying ONC national standards in support of metadata for research warehouse management of repurposed EHR clinical data, and extend our findings to ADRD. Successful completion of each SA will incrementally advance the ability to repurpose and reuse clinical data for AD research. The computable reference terminologies, founded on ONC data standards will enable the on-going, reusable links between EHR data and research data for AD patient research.
| Campbell, Walter S; Karlsson, Daniel; Vreeman, Daniel J et al. (2017) A computable pathology report for precision medicine: extending an observables ontology unifying SNOMED CT and LOINC. J Am Med Inform Assoc : |
| Campbell, James R; Talmon, Geoffrey; Cushman-Vokoun, Allison et al. (2016) An Extended SNOMED CT Concept Model for Observations in Molecular Genetics. AMIA Annu Symp Proc 2016:352-360 |
