Developing core indicators for identifying people at risk of delayed heart failure diagnosis.

BACKGROUND: Heart failure (HF) is frequently diagnosed during hospital admission, often after symptoms have been present for some time. Those diagnosed in hospital typically experience higher mortality reflecting not only possible diagnostic delay but also greater illness severity at presentation. The reasons behind delayed HF diagnosis are multifaceted and complex. This study aimed to achieve consensus on a priority list of patient, clinical, and service-level factors associated with delayed HF diagnosis, and to identify indicators that could support earlier detection of undiagnosed HF in primary care. METHODS: A three-round modified e-Delphi process involved patients and clinicians from primary and specialist care. Participants rated sociodemographic and clinical factors for their importance in delayed HF diagnosis and clinicians also rated service-level factors and identified indicators of undiagnosed HF. Consensus was defined as two-thirds agreement with stable opinions across rounds (McNemar p ≥ 0.05). Indicators of undiagnosed HF required additional ranking in the top 5 by > 50% of clinicians. RESULTS: The first survey was completed by 18 patients (67% women, median age 61) and 27 clinicians (67% nurses/allied health professionals, 33% doctors). Consensus was achieved, comprising 15 factors and 5 indicators. Key sociodemographic factors were patients lacking HF knowledge, lack of access to GP/cardiologist appointments, symptom confusion, younger age (< 50), and learning difficulties. Clinical factors included multimorbidity, respiratory/mental health conditions, obesity, and depression. Service-level factors included poor HF knowledge, limited N-terminal pro-B-type Natriuretic Peptide (NT-proBNP) testing and echocardiogram access in primary care, and fragmented care. The top 5 indicators of undiagnosed HF included elevated NT-proBNP without referral, loop diuretic use, and overlapping cardiac and respiratory histories. CONCLUSIONS: This study identifies critical factors and indicators that can aid earlier HF diagnosis in primary care. These indicators could be embedded into electronic health record-based alerts and used to support decision-making in primary care.