VA HSR&D IIR 11-292 (PI Michael Matheny): Coronary Angiography population health management, including development and validation of real-time natural language processing tools, risk prediction and stratification incorporating both structured data and free text data, and the deployment and use of a population surveillance dashboard for acute kidney injury and institutional quality profiling and outlier detection.
VA HSR&D IIR 12-364 (PI Ruth Reeves, Co-PI Theodore Speroff): This study will develop a temporal reasoning NLP tool for ordering relevant events contained within the EHR, linking across multiple documents to construe sequences of PTSD symptom occurrences and treatments over time, importing and updating these events within a queryable event database.
VA HSR&D IIR 13-052 (PI: Michael Matheny, Co-PI (San Diego VA) Samuel Ho): Advanced liver disease population health management, including development and validation of real-time NLP processing tools for case identification and risk stratification augmentation to structured data, with the deployment and use of a patient dashboard to manage ALD patients (previously identified or unidentified) in the inpatient setting including the transition of care period to the outpatient PCP or gastroenterology clinic follow-up.
VA HSR&D IIR 14-XXX (PI: Edward Siew): This study will perform a series of studies within a National cohort of Veteran AKI survivors to characterize variations in the patterns of recovery, recurrent AKI, and their association with future ESRD. It will also identify clinical predictors of recovery and recurrent AKI that can improve risk stratification for AKI survivors, and then examine how AKI impacts the delivery of specific care strategies, focusing on RAASi therapy, and finally will characterize different patterns of RAASi use among AKI survivors, identify the clinical features that associate with RAASi use, and examine the potential association between RAASi use and recurrent AKI. These findings will help identify testable care strategies and inform individual risk/benefit assessments. In addition, the characterization of risks and benefits, methodological refinements, and quantification of potential effect sizes will set the stage for future studies needed to improve outcomes in AKI survivors.
AstraZeneca Unrestricted Grant (PI: Michael Matheny): The purpose of the study is 1) to determine within the national VA patient cohort if any clinical characteristics would predict an opioid-treated patient’s subsequent development of OIC and 2) to determine if any clinical characteristics would predict an OIC patient’s response or lack of a response to generic or over-the-counter (OTC) laxatives. This study is in support of AstraZeneca’s Naloxegol program obtain insights into the development of OIC among opioid-treated patients and OIC patients’ response to laxatives.
VA OAA Advanced Fellowship in Medical Informatics (PI: Steven Brown): A 2 year post-doctoral fellowship offered to train medical informatics fellows in close collaboration with VUMC Department of Biomedical Informatics. 2 slots for 2 years are offered, all previous requests for a 3rd year or a 3rd recruitment slot have been approved on an as-needed basis.
PCORI CDRN Phase 1 (PI: Ohno-Machado, Site PI: Michael Matheny): This project is designed to utilize a distributed privacy-preserving architecture to integrate data from three existing networks covering over 21 million patients in all 50 states: (1) VA Informatics and Computing Infrastructure (VINCI), with data from Veteran Health Administration’s (VHA); (2) the University of California Research eXchange (UC-ReX) network, with data from UC Davis, Irvine, Los Angeles, San Francisco, and San Diego; and (3) SCANNER, a consortium of UCSD, Tennessee VA, and three federally qualified health systems in the Los Angeles area supplemented with claims and health information exchange (HIE) data, led by the University of Southern California. Initial use cases will focus on three conditions: (1) congestive heart failure, (2) Kawasaki disease, and (3) obesity. Our site is responsible for adapting and executing the OMOP CDM for the national VA, harmonizing the PCORNet CDM across the CDRN with the OMOP CDM at the VA, deploying distributed analytic tools across the CDRN, ensuring data quality for the clinical use cases, and potentially participating in a pragmatic clinical trial.
NIH NHGRI U-01 XXX (PI: Denny, Site PI: Michael Matheny): This is a grant highlighting the excellent potential between leveraging VA and VUMC infrastructures and research teams in order to successfully compete for research funding. Dr. Denny and Levy are conducting a multi-site trial of real-time genetic testing for anti-coagulation and cancer directed therapy considerations, and our research group is currently managing the VA site portion of the trial for data infrastructure and informatics tool extension and loop-back to the coordinating center.
FDA U-01 XXX (PI: Frederic Resnic, Site PI: Michael Matheny): Establishing appropriate expectations for device performance is critical in monitoring for medical device failures, as well as for surveillance for potential risks from contamination or sabotage. We propose to develop methods and infrastructure for prospective medical device safety surveillance that address these key limitations, while establishing a national distributed cardiovascular device safety network to monitor newly released, high risk cardiovascular devices.
VA HSR&D CRE 12-037 (PI: Jennifer Garvin, Site PI: Michael Matheny): This study will develop an automated post-discharge communication aid that contains information needed to prompt beta blocker titration by the PACTs at the point of care. We operationally define point of care within the context of PACT practice as any setting where a provider evaluates clinical information and makes a care decision. This communication aid will contain key clinical information such as the patient’s ejection fraction, current beta blocker dose, target beta blocker dose per the guidelines, information on patient’s heart rate and blood pressure, and any beta blocker allergy. We will use informatics techniques, including information extraction and natural language processing (NLP), to accurately extract key clinical information and to identify candidate patients for beta blocker titration.
VA HSR&D IIR 12-064 (PI: Steven Luther, Site PI: Michael Matheny): This proposal focused on the management of patients with spinal cord injury and prevention of pressure ulcers in this population. The research includes NLP, risk stratification, identification of high risk patients and execution of tailored interventions in order to provide tailored care to these patients.
VA HSR&D VINCI Resource Center (PI: Jonathan Nebeker, Site PI: Michael Matheny): Development of OMOP CDM for national VA and development of visualization and analytic tools to be used by the National VA HSR&D community.