New Prediction Tool Supports Confident Prescribing of Heart Failure Medicines

Clinicians often delay or withhold therapy because of concerns about adverse effects. This gap undermines outcomes despite strong recommendations for combination treatment. In response, researchers have developed an online prediction tool to support safer and more confident initiation of heart failure medicines.

The Heart Failure Treatment Effect Calculator, developed by The George Institute for Global Health in collaboration with Brigham and Women’s Hospital and Harvard Medical School, is designed to forecast short‑term clinical changes when starting therapy. The research underpinning the tool was presented at the European Society of Cardiology Congress and published in Nature Medicine on August 31, 2026. The model was derived from individual patient data pooled across major randomized heart failure trials. It is freely accessible for clinical use.

The calculator generates personalized estimates using routinely available patient characteristics, including age, sex, body mass index, baseline blood pressure, kidney function, serum potassium, and prior heart failure hospitalization. It projects the short-term effects of starting combinations of five treatment classes: angiotensin receptor neprilysin inhibitors, angiotensin-converting enzyme inhibitors or angiotensin receptor blockers, sodium-glucose cotransporter 2 inhibitors, steroidal mineralocorticoid receptor antagonists, and non-steroidal mineralocorticoid receptor antagonists. Outputs focus on blood pressure, estimated glomerular filtration rate, and potassium, helping inform bedside risk-benefit discussions and reduce uncertainty at the time of prescribing.

Model development incorporated data from 38,753 participants across nine major trials. Recommended combination regimens were associated with modest blood pressure reductions, early declines in kidney function, and small to modest increases in potassium, but these changes were accompanied by substantial reductions in worsening heart failure events. Compared with standard care, risk reductions ranged from 31% to 61%. External validation in 1,016 participants across four additional trials showed close alignment between predicted and observed effects.

Guidelines recommend a four-pillar regimen for heart failure with reduced ejection fraction, consisting of a beta-blocker, an angiotensin receptor neprilysin inhibitor, a mineralocorticoid receptor antagonist, and a sodium-glucose cotransporter 2 inhibitor. This approach can reduce all-cause mortality by up to 60%, yet registry data suggest only 2% of eligible patients receive such therapy. With heart failure affecting an estimated 64 million people globally and one-year mortality risk reaching up to 30%, the calculator could help support broader use of evidence-based treatment, although the authors note that model estimates may be less generalizable outside trial-like populations.

“Over the last 30 years, we have seen significant treatment advances for patients with heart failure, but adoption of these therapies has been poor. Fear of adverse effects, particularly low blood pressure, worsening kidney function or high potassium levels are the major reasons clinicians do not start treatment,” said Dr. Nelson Wang, Cardiologist and Senior Research Fellow at The George Institute for Global Health.

“By providing a personalised estimate of what to expect when initiating combination therapy, the calculator can reduce uncertainty and give clinicians greater confidence to prescribe these treatments simultaneously, ultimately helping more patients receive therapies that improve long-term outcomes,” Wang added.

Related Links
The George Institute for Global Health