ISPOR 2020: Veradigm Examines Care Gaps in Heart Failure

Thought Leadership  |  09 June 2020  |  By John M. Farah, PhD

Veradigm recently presented four posters in association with the International Society for Pharmacoeconomics and Outcomes Research (ISPOR) 2020 conference, held online May 18 20 because of the ongoing COVID-19 pandemic. The presentations delivered on the conference theme, “Health Economics and Outcomes Research (HEOR): Advancing Evidence to Action.”

The studies leveraged real-world data from ambulatory patients with heart failure (HF), sourced from an electronic health record (EHR) dataset in the Health Insights database and from the PINNACLE Registry®, a cardiovascular registry operated in partnership between Veradigm and the American College of Cardiology.

Veradigm researchers applied natural language processing (NLP) and machine learning (ML) to successfully enrich capture of HF symptoms and ejection fraction (EF), which is a measurement, expressed as a percentage, of how much blood the left ventricle pumps out during each contraction. These important indicators of heart function in at-risk HF patients are typically available only in free-text fields. The extracted data were de-identified and used to generate real-world evidence (RWE) critical to understanding and addressing care gaps in HF management.

Veradigm Research Posters Accepted by ISPOR

  • Mining for Heart Failure Symptoms in Clinical Notes (PCV81) Using Natural Language Processing (NLP) methods to enrich data from structured fields, signs and symptoms of HF were successfully extracted from unstructured clinical notes available in a large cloud-based, ambulatory patient EHR platform.
  • Development and Validation of a Method to Extract Left Ventricular Ejection Fraction Data from EHR Physician Notes (PCV82) A machine learning (ML) process developed by Veradigm data scientists delivered greater accuracy in extracting EF values from clinical notes than a rule-based NLP algorithm for patients with HF.
  • Guideline-Directed Medical Therapy Among Heart Failure Patients Facilitated by Natural Language Processing of Ambulatory Electronic Health Records (PCV66) In a retrospective study of patients with HF, application of Veradigm digital technologies (ML and NLP) revealed EF and signs and symptoms were among several variables that differentiated patients receiving guideline-directed medical therapy from those who did not.
  • Documentation Rate of Left Ventricular Ejection Fraction in a Cardiovascular Registry and in Electronic Healthcare Record (EHR) Systems (PCV14) In the Veradigm PINNACLE cardiovascular registry, documentation of EF was higher than that recorded in structured fields of three general purpose EHRs.

Only Veradigm has these custom-built analytic tools designed specifically to extract and de-identify patient data from unstructured clinical notes in Veradigm EHRs and registries. These data enrichment solutions hold tremendous potential value for healthcare stakeholders, including leaders in HEOR and RWE, population health, and epidemiology, along with others in the life sciences who are interested in gaining timely, in-depth insights into real-world patient populations, for purposes of addressing challenging research questions, optimizing clinical outcomes, and advancing patient care.

Registrants for the virtual conference may access all of the nearly 60 hours of content as on-demand recordings through June 30th.

For more information regarding Veradigm EHR and registries datasets, analytic capability, and data enrichment services please contact us.

Download Guideline-directed Medical Therapy among Heart Failure Patients Poster

Download Method to Extract Left Ventricular Ejection Fraction Data Poster

Download Mining for Heart Failure Symptoms in Clinical Notes Poster

Download Documentation Rate of Left Ventricular Ejection Fraction Poster

Spread the word

Related insights