AimsHeart failure (HF) is common in geriatric patients. Clinicians face diagnostic challenges primarily due to comorbidity and limited access to echocardiography. The purpose of this study was to identify independent determinants of the presence of HF in geriatric outpatients and to determine the optimal diagnostic strategy. Methods and resultsGeriatric outpatients [mean age 82 (+6) years, 30% men] with suspected HF underwent an extensive standardized diagnostic work-up. An expert consensus panel determined the presence of HF. Heart failure was present in 94 of 206 participants (46%). Male sex [odds ratio (OR) 2.0], age per 10 years (OR 1.6), nocturnal dyspnoea (OR 1.7), absence of wheezing (OR 2.1), loss of appetite (OR 1.7), and lower body mass index (BMI; OR 0.9) were independently associated with the presence of HF: the c-statistic of the model containing these items was 0.75. Of all additional tests, N-terminal pro-B-type natriuretic peptide (NT-proBNP) improved the diagnostic accuracy the most (OR ln NT-proBNP 2.8; c-statistic 0.92). A diagnostic rule, consisting of six clinical variables and NT-proBNP, showed good negative and positive predictive values. ConclusionHalf of geriatric patients suspected of HF actually have HF. Apart from age, gender, and nocturnal dyspnoea, absence of wheezing, loss of appetite, and lower BMI were independently associated with the presence of HF. Symptoms and signs in combination with NT-proBNP reliably identified the presence or absence of HF in the vast majority of patients. Additional diagnostic tests, in particular echocardiography, can be targeted at those in whom the presence of HF remains uncertain and to ascertain the cause of HF.--