Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2024 Jul 2;3(8):101064.
doi: 10.1016/j.jacadv.2024.101064. eCollection 2024 Aug.

Race- and Ethnicity-Related Differences in Heart Failure With Preserved Ejection Fraction Using Natural Language Processing

Sam Brown  1   2 , Dhruva Biswas  1   2 , Jack Wu  2 , Matthew Ryan  1   2 , Brett S Bernstein  1   2 , Natalie Fairhurst  1 , George Kaye  1 , Ranu Baral  1 , Antonio Cannata  2 , Tom Searle  3 , Narbeh Melikian  1   2 , Daniel Sado  1   2 , Thomas F Lüscher  4 , James Teo  1   3 , Richard Dobson  3 , Daniel I Bromage  1   2 , Theresa A McDonagh  1   2 , Ali Vazir  4 , Ajay M Shah  1   2 , Kevin O'Gallagher  1   2
Affiliations

Race- and Ethnicity-Related Differences in Heart Failure With Preserved Ejection Fraction Using Natural Language Processing

Sam Brown et al. JACC Adv. .

Abstract

Background: Heart failure with preserved ejection fraction (HFpEF) is the predominant form of HF in older adults. It represents a heterogenous clinical syndrome that is less well understood across different ethnicities.

Objectives: This study aimed to compare the clinical presentation and assess the diagnostic performance of existing HFpEF diagnostic tools between ethnic groups.

Methods: A validated Natural Language Processing (NLP) algorithm was applied to the electronic health records of a large London hospital to identify patients meeting the European Society of Cardiology criteria for a diagnosis of HFpEF. NLP extracted patient demographics (including self-reported ethnicity and socioeconomic status), comorbidities, investigation results (N-terminal pro-B-type natriuretic peptide, H2FPEF scores, and echocardiogram reports), and mortality. Analyses were stratified by ethnicity and adjusted for socioeconomic status.

Results: Our cohort consisted of 1,261 (64%) White, 578 (29%) Black, and 134 (7%) Asian patients meeting the European Society of Cardiology HFpEF diagnostic criteria. Compared to White patients, Black patients were younger at diagnosis and more likely to have metabolic comorbidities (obesity, diabetes, and hypertension) but less likely to have atrial fibrillation (30% vs 13%; P < 0.001). Black patients had lower N-terminal pro-B-type natriuretic peptide levels and a lower frequency of H2FPEF scores ≥6, indicative of likely HFpEF (26% vs 44%; P < 0.0001).

Conclusions: Leveraging an NLP-based artificial intelligence approach to quantify health inequities in HFpEF diagnosis, we discovered that established markers systematically underdiagnose HFpEF in Black patients, possibly due to differences in the underlying comorbidity patterns. Clinicians should be aware of these limitations and its implications for treatment and trial recruitment.

Keywords: AI (artificial intelligence); Natural Language Processing; health equity; heart failure; preserved ejection fraction.

PubMed Disclaimer

Conflict of interest statement

Professor Shah has served as an advisor to Forcefield Therapeutics and CYTE–Global Network for Clinical Research. Professor McDonagh has received speaker fees or advisory board fees from Abbott, Edwards, Boehringer Ingelheim, and AstraZeneca. This work was supported by grants from the 10.13039/501100000274British Heart Foundation (CH/1999001/11735, RG/20/3/34823, and RE/18/2/34213 to Professor Shah; CC/22/250022 to Dr Dobson, Professor Shah, Dr Teo, and Dr Gallagher) and King’s College Hospital Charity (D3003/122022/Shah/1188 to Dr Shah). Dr Gallagher and Dr Bromage are each supported by MRC Clinician Scientist Fellowships (MR/Y001311/1 to Dr Gallagher, MR/X001881/1 to Dr Bromage). All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.

Figures

None
Graphical abstract
Figure 1
Figure 1
Cohort Overview CONSORT (Consolidated Standards of Reporting Trails) diagram of patient inclusion and exclusion criteria. Simple criteria HFpEF includes patients with clinical signs and symptoms of HF and LVEF ≥50% but do not meet the additional structural and/or biochemical ESC HFpEF criteria for inclusion. ESC = European Society of Cardiology; HF = heart failure; HFpEF = heart failure with preserved ejection fraction; LVEF = left ventricular ejection fraction.
Figure 2
Figure 2
Comorbidities at HFpEF Diagnosis Radar plots showing the percentage of patients with each reported comorbidity mentioned prior to first HF mention, plotted separately by ethnicity. AF = atrial fibrillation; CAD = coronary artery disease; CKD = chronic kidney disease; DM = diabetes mellitus type 2; HF = heart failure; HTN = hypertension.
Figure 3
Figure 3
HF Markers (A) The H2FPEF score displayed as the percentage of patients meeting each of the score components; BMI >30 kg/m2, atrial fibrillation (AF), hypertension, pulmonary artery systolic pressure (PASP) > 35 mm Hg, setage >60 Years, and E/e’ ratio >9 (ratio of early diastolic mitral inflow velocity to mitral annulus relaxation velocity). Plotted as individual bars per ethnicity. (B) Pie chart depicting the percentage of patients with high (≥6), intermediate (2-5), and low (<1) H2FPEF scores as defined by the original publication, plotted separately by ethnicity. (C) Bar chart for median NT-proBNP stratified by ethnicity. P values calculated by Kruskal-Wallis Test. BMI = body mass index; HF = heart failure; NT-proBNP = N-terminal pro-B-type natriuretic peptide.
Figure 4
Figure 4
Survival Outcomes (A) Kaplan-Meier plot showing overall survival outcomes from first HF mention stratified by ethnicity and compared via log-rank test. (B) Forest plot showing adjusted hazard ratios for multivariate Cox regression analysis of overall survival from first HF mention by ethnicity, age, sex, socioeconomic deprivation (percentage in most deprived indices of multiple deprivation [IMD] quintile), NT-proBNP ≥125 pg/mL (375 pg/mL in atrial fibrillation) status and diagnostic group (clinician-assigned HFpEF or ESC criteria diagnosed). ESC = European Society of Cardiology; HF = heart failure; HFpEF = heart failure with preserved ejection fraction; NT-proBNP = N-terminal pro-B-type natriuretic peptide.
Figure 5
Figure 5
Cox Regression Analysis of HF Markers (A) Forest plot showing unadjusted hazard ratios for univariate cox regression analysis of overall survival from first HF mention by H2FPEF score ≥6 status stratified by ethnicity subgroup. (B) Forest plot showing unadjusted hazard ratios for univariate cox regression analysis of overall survival from first HF mention by NT-proBNP ≥125 pg/mL (375 pg/mL in atrial fibrillation) status stratified by ethnicity subgroup. HF = heart failure; NT-proBNP = N-terminal pro-B-type natriuretic peptide.
Central Illustration
Central Illustration
Race and Ethnicity-Related Differences in HFpEF Diagnosed by Natural Language Processing Piechart (central) shows the number of patients diagnosed with HFpEF according to the european society of cardiology diagnostic criteria stratified by self-reported ethnicity. (A) Shows a bar chart for the number of patients with H2FPEF score ≥6 by ethnicity, with P values. (B) Bar chart depicting the median NT-proBNP values for each ethnic group, along with their IQRs. ESC = European Society of Cardiology; NT-proBNP = N-terminal pro-B-type natriuretic peptide.
See this image and copyright information in PMC

References

    1. Borlaug B.A., Sharma K., Shah S.J., Ho J.E. Heart failure with preserved ejection fraction. J Am Coll Cardiol. 2023;81(18):1810–1834. - PubMed
    1. Owan T.E., Hodge D.O., Herges R.M., Jacobsen S.J., Roger V.L., Redfield M.M. Trends in prevalence and outcome of heart failure with preserved ejection fraction. N Engl J Med. 2006;355(3):251–259. - PubMed
    1. Chang P.P., Wruck L.M., Shahar E., et al. Trends in Hospitalizations and survival of acute Decompensated heart failure in Four US Communities (2005–2014) Circulation. 2018;138(1):12–24. - PMC - PubMed
    1. Sueta D., Yamamoto E., Nishihara T., et al. H2FPEF score as a prognostic value in HFpEF patients. Am J Hypertens. 2019;32(11):1082–1090. - PubMed
    1. Verbrugge F.H., Reddy Y.N.V., Sorimachi H., Omote K., Carter R.E., Borlaug B.A. Diagnostic scores predict morbidity and mortality in patients hospitalized for heart failure with preserved ejection fraction. Eur J Heart Fail. 2021;23(6):954–963. - PubMed

LinkOut - more resources