This career development application describes targeted coursework and mentored research for clinical evidence generation from electronic health records for precision medicine. Although randomized controlled trials (RCTs) are the gold standard of clinical evidence, either due to RCTs not matching the patient seen in the clinic, the rarity of a condition, or lack of equipoise to justify randomization, these data are often unavailable to guide a number of treatment decisions. A clinical example is the case of management of unruptured intracranial aneurysms (UIAs). Although an RCT was attempted to determine the benefit of preventive surgical treatment compared to careful observation, this study failed due to patient and provider concerns about randomization. It is clear that new methods of evidence generation are needed.
The research aims of this proposal describe a novel systematic approach to clinical evidence generation from EHRs under the hypothesis that domain expert provided insights into the ?on- the-ground? work of clinical assessment will enable the generation of expert-informed hypotheses, which coupled with data quality assessment and newly developed ?target trial? causal inference models will allow for robust precision evidence to support clinical decision making. This hypothesis will be tested by:
Aim 1 - Identify factors guiding clinical management recommendations for UIAs through in-depth, semi-structured interviews of neurovascular expert physicians;
Aim 2 - Extract factors guiding clinical recommendations and clinical outcomes from the EHR, characterize the quality and extractability of these variables, and assess their fitness for use in clinical evidence generation;
and Aim 3 - Assess the efficacy of preventive surgical treatment vs observation for UIAs through a target-trial causal inference approach. This project has significant potential to influence the clinical care provided for patients with UIAs, where there is currently a paucity of evidence guiding care. The creation of a robust, and reproducible framework for clinical evidence generation is both innovative and impactful on the field of biomedical informatics and clinical data science. To accomplish these research aims, Dr. Wiley will be mentored by a team of experts in clinical evidence generation (Dr. Lisa Schilling), qualitative methodologies (Dr. Jennifer Reich), data quality assessments (Dr. Michael Kahn), and causal inference methodologies (Dr. Debashis Ghosh). This mentorship coupled with Dr. Wiley?s existing record of research performance, dedication to local and national education and service, will ensure Dr. Wiley becomes an independent investigator and international leader in clinical evidence generation from EHRs for precision medicine

Public Health Relevance

This proposal seeks to create a novel systematic approach to clinical evidence generation from electronic health records for precision medicine. It will generate clinician expert informed hypotheses, which coupled with assessment of EHR data quality and causal inference statistical techniques, will create robust clinical evidence. These methods are applied to identify the optimal clinical management strategy of unruptured intracranial aneurysms and could significantly impact clinical care in the future.

Agency
National Institute of Health (NIH)
Institute
National Library of Medicine (NLM)
Type
Research Scientist Development Award - Research & Training (K01)
Project #
5K01LM013088-02
Application #
9931272
Study Section
Special Emphasis Panel (ZLM1)
Program Officer
Sim, Hua-Chuan
Project Start
2019-05-17
Project End
2022-04-30
Budget Start
2020-05-01
Budget End
2021-04-30
Support Year
2
Fiscal Year
2020
Total Cost
Indirect Cost
Name
University of Colorado Denver
Department
Internal Medicine/Medicine
Type
Schools of Medicine
DUNS #
041096314
City
Aurora
State
CO
Country
United States
Zip Code
80045