Resistant hypertension (RH) doubles the risk for adverse cardiovascular outcomes compared to non-resistant hypertension. Defined as having uncontrolled high blood pressure despite the use of at least 3 antihypertensive medications or controlled blood pressure on at least 4 medications, RH is estimated to affect at least 20 million Americans. Importantly, true RH must be differentiated from pseudo-resistant hypertension (pseudo-RH), occurring when blood pressure remains elevated due to extrinsic factors such as suboptimal medication dosing, medication non-adherence, or white-coat effect. Inability to distinguish true RH from pseudo-RH, and tailor treatment accordingly, compounds the risks of overtreating pseudo-RH (e.g. syncope, falls, acute kidney injury) as well as undertreating true RH (e.g. stroke, myocardial infarction). Distinguishing true from pseudo-RH, however, is clinically difficult, in part given the complexities involved in capturing medication adherence patterns and confirming white-coat effect. Therefore, our overall objective is to determine whether electronic health record (EHR) based analytics and tools can be used to close persistent gaps in care for RH.
The specific aims of the research project are to: (1) develop and validate a computerized algorithm that uses EHR data to identify RH and distinguish between true and pseudo-RH including its subtypes; (2) develop and optimize a CDS tool for aiding clinicians in the identification and management of apparent RH; and, (3) pilot the implementation of a CDS tool for facilitating care of RH in addition to pseudo-RH and its subtypes. This research promises to enhance our understanding of how health information technology can be leveraged to inform scientific discovery, while also driving high-value care for RH. The proposed work will be conducted as part of a K23 award program, designed provide the advanced research skills and experience needed for the PI to successfully pursue an independent academic career focused on: (i) optimizing value of care (i.e. improved quality at decreased cost); (ii) leveraging health information technology and clinically generated data to gain new insights into disease states; and, (iii) promoting innovation in care delivery using implementation science principles. The PI will accomplish the proposed research and training aims with the support of his mentoring team: Dr. Teryl Nuckols (health services research and value of care), Dr. Susan Cheng (preventive cardiology, large data analytics, and population health) and, Dr. Joshua Pevnick (clinical informatics). These efforts will be supported by the outstanding research environment and infrastructure of Cedars-Sinai including the Smidt Heart Institute, the Biostatistics and Bioinformatics Research Center, and the Research Informatics and Scientific Computing Core. Given the strong mentoring, institutional, and infrastructure supports in place, the proposed K award program is ideally designed to provide the experience needed to launch the PI in his career as an independent investigator and future leader in cardiovascular outcomes research.
Hypertension remains a major contributor to morbidity and mortality in the United States, and a fifth of affected individuals have a form of hypertension that appears resistant to usual approaches. The ability to reliably distinguish true resistant hypertension from pseudo-resistant hypertension (due to under-prescribing, medication non-adherence, or white coat effect) is essential for minimizing diagnostic and therapeutic errors. We will use electronic health record based analytics and tools to develop a clinically feasible and pragmatic approach to improving the quality and value of care for patients with resistant hypertension.
