PCV66 GUIDELINE-DIRECTED MEDICAL THERAPY AMONG HEART FAILURE PATIENTS FACILITATED BY NATURAL LANGUAGE PROCESSING OF AMBULATORY ELECTRONIC HEALTH RECORDS

Guideline-directed medical therapy (GDMT) reduces morbidity and mortality in stage C heart failure (HF) with reduced ejection fraction (HFrEF, EF ≤40%) and manages blood pressure, volume overload, and comorbidities in stage C heart failure with preserved ejection fraction (HFpEF, EF ≥50%). Despite these benefits, many patients with HF are not receiving GDMT. This retrospective analysis evaluated whether characteristics differ between GDMT+ and GDMT- patients. Real-world evidence was generated from de-identified real-world data sourced from the Veradigm electronic health record (EHR) Practice Fusion. Eligible patients had at least one HF symptom and at least one EF measurement between 01/01/2018 and 06/30/2019. HF symptoms and EF were extracted using natural language processing (NLP). Patients were stratified into GDMT+ (at least two Class of Evidence I GDMT treatments) and GDMT- cohorts and characterized according to demographics, clinical laboratory findings, and comorbidities. Comparisons of categorical and continuous variables were performed using Chi-square tests and simple linear regression (analysis of variance), respectively. Of 13,753 eligible patients, 42.1% had HFrEF and 57.9% had HFpEF. GDMT+ (n=7,224) and GDMT- (n=6,529) cohorts had comparable median ages (74 yr); 76% of patients were older than 65 yr; 45% were female. Univariate testing indicated significant differences between the GDMT cohorts for symptoms, EF, weight, body mass index, serum creatinine, hemoglobin A1c, and comorbidities including atherosclerotic cardiovascular disease, hypertension, atrial fibrillation, depression/anxiety, renal disease, diabetes, and metabolic disorder but not chronic obstructive pulmonary disease (COPD) (all p<0.05). Real-world evidence from this retrospective analysis using EHR data and NLP extraction demonstrates most variables, including comorbidities with the single exception of COPD, differentiated GDMT+ from GDMT- patients. Further evaluation of these variables as potential predictors of recommended therapy is warranted for population health insights into patients with HF.