Patient Perspectives Towards Artificial Intelligence in Heart Failure Care

Abstract

Background Artificial intelligence (AI) tools are increasingly being developed to support the management of chronic diseases, including heart failure (HF). Little is known about the views of patients with HF, a population typically older, with high comorbidity, and familiar with close clinician relationships. This study aims to investigate HF patients' perception of AI integration into HF care to help guide a patient-centred adoption. Methods We conducted a cross-sectional survey on consecutive patients with HF with reduced ejection fraction attending follow-up at a clinic. The questionnaire assessed satisfaction with current care, trust in cardiologists, and attitudes toward AI involvement in diagnosis and treatment decisions. Responses were collected on Likert scales and analysed using ordinal and binary logistic regression to test associations with age, sex, education level, smartphone ownership, symptom burden and prior experience with AI. Results A total of 110 patients completed the questionnaire. Attitudes towards AI were mixed. While 38.1% were happy for their doctor to use AI's help in treatment decisions, support fell significantly to 18.2% when AI acted without physician input and to 21.8% when AI remotely adjusted treatment with fewer in-person visits. Even when described as outperforming doctors, half of the patients remained uncomfortable. In direct comparisons, 80.9% preferred cardiologist diagnoses, 84.6% preferred cardiologist treatment plans, and 97.3% would trust their cardiologist over AI in cases of disagreement. No association was found with age, sex, or education. Smartphone ownership, however, predicted greater acceptance of remote AI adjustments. Conclusion HF patients report high satisfaction with care and strong trust in cardiologists, with more cautious attitudes toward AI than seen in general patient populations. Smartphone ownership, rather than demographics or specific experience with AI, predicted openness to AI. Preserving clinician oversight and designing accessible AI tools will be key to equitable adoption in HF care.

Ähnliche Arbeiten