Clinician-assessed NYHA and EHRA symptom classifications only moderately reflect patient-reported quality of life in heart failure and atrial fibrillation

Authors:

Abstract
Background: Clinician-assessed symptoms are expected to agree with patient-perceived symptoms. However, patients may value the impact of symptoms on their health-related quality of life (HRQoL) differently from clinicians. This study aimed to evaluate the relationship between patient-perceived HRQoL and clinician-assessed symptom severity in patients with Heart Failure (HF) and Atrial Fibrillation (AF). Methods: Newly diagnosed HF and AF patients were prospectively included. Information on symptoms and HRQoL was collected at diagnosis and after 12 months during routine HF- and AF-outpatient clinic visits. Symptom severity was assessed by clinicians using the NYHA classification for HF and EHRA score for AF. HRQoL was assessed using the CaReQol-CHF (HF) and AFEQT (AF). Results: Complete baseline and 12-month follow-up information was available for 254 HF patients and 765 AF patients. Moderate correlations were observed between NYHA classification and CaReQoL-CHF at diagnosis and follow-up (Spearman’s ρ: HF social : 0.202 and 0.426, HF physical : 0.317 and 0.421, respectively), and between EHRA score and AFEQT (Spearman’s ρ: −0.363 and −0.305). Clinician-assessed improvement in HF or AF symptoms from diagnosis to 12 months was associated with patient-reported improvement in HRQoL (HF social : F(250.2) = 3.11, p < 0.046; HF physical : F(251.2) = 11.19, p < 0.001; and AF: F(762.2) = 25.86, p < 0.001). When clinicians reported unchanged or worsening symptoms, no significant HRQoL changes were reported by patients. Conclusion: Clinician-assessed symptoms correlate moderately with patient-reported HRQoL, with large inter-individual variation in HRQoL. NYHA and EHRA classifications may not accurately reflect patient-perceived burden of disease over time. Incorporating patient-reported outcomes into routine care may better guide treatment, improve the identification of patients’ needs, and support informed clinical and shared decision-making.

This publication has 36 references indexed in Scilit: