Phenotyping to Facilitate Accrual for a Cardiovascular Intervention

Abstract

Background The conventional approach for clinical studies is to identify a cohort of potentially eligible patients and then screen for enrollment. In an effort to reduce the cost and manual effort involved in the screening process, several studies have leveraged electronic health records (EHR) to refine cohorts to better match the eligibility criteria, which is referred to as phenotyping. We extend this approach to dynamically identify a cohort by repeating phenotyping in alternation with manual screening. Methods Our approach consists of multiple screen cycles. At the start of each cycle, the phenotyping algorithm is used to identify eligible patients from the EHR, creating an ordered list such that patients that are most likely eligible are listed first. This list is then manually screened, and the results are analyzed to improve the phenotyping for the next cycle. We describe the preliminary results and challenges in the implementation of this approach for an intervention study on heart failure. Results A total of 1,022 patients were screened, with 223 (23%) of patients being found eligible for enrollment into the intervention study. The iterative approach improved the phenotyping in each screening cycle. Without an iterative approach, the positive screening rate (PSR) was expected to dip below the 20% measured in the first cycle; however, the cyclical approach increased the PSR to 23%. Conclusions Our study demonstrates that dynamic phenotyping can facilitate recruitment for prospective clinical study. Future directions include improved informatics infrastructure and governance policies to enable real-time updates to research repositories, tooling for EHR annotation, and methodologies to reduce human annotation.

Remote Optimization of Guideline-Directed Medical Therapy in Patients With Heart Failure With Reduced Ejection Fraction

“…The remote optimization program was organized as a quality improvement initiative, with design details as previously reported. 5 Patients with a chronic HF and left ventricular ejection fraction of 40% or less who were receiving longitudinal cardiovascular follow-up at our center were identified through an electronic health record-based search algorithm 6,7 and contacted via telephone to verify eligibility and willingness to participate in the remote optimization program. For those whose clinicians granted permission, navigators (T.M., J.B.-H., and nonauthors) and pharmacists (J.D.…”

Evaluation of the Usage and Dosing of Guideline-Directed Medical Therapy for Heart Failure With Reduced Ejection Fraction Patients in Clinical Practice

“…We utilized an electronic health record-based algorithm to identify a cohort of patients with chronic heart failure and most recent left ventricular ejection fraction ≤ 40% as part of an ongoing population health-level quality improvement initiative. 12 Eligible patients were those age > 18 years who had established longitudinal care with specialty cardiology providers at our center (at least 2 clinic visits, including one within the last 18 months). Patients with end-stage HF requiring inotropic or mechanical circulatory support, those with prior cardiac transplantation or actively listed for transplantation, and those enrolled in hospice or palliative care were excluded.…”

Cloud Services for Patient Cohort Identification Using the Informatics for Integrating Biology and the Bedside Platform

“…A limitation of our evaluation is that we did not evaluate the “local-concept-map” Logic, as the MIMIC-III dataset already has standard codes for diagnosis and laboratory result, i.e., ICD and LOINC, respectively. Second, we limited our derived concepts to simple rules although machine learning models can be modeled using our current implementation mechanism [ 14 ]: our system supports machine learning via Pyspark-based algorithms that can run within a data pipeline. Our systems provide the mechanism to execute a programming script that results into a derived fact for each patient.…”