BMC Cardiovasc Disord. 2026 Jul 6;26(1):563. doi: 10.1186/s12872-026-06230-9.
ABSTRACT
BACKGROUND: Heart failure (HF) is a common disease among older individuals and is associated with poor quality of life and prognosis. Individuals at risk of developing HF are usually already patients in primary healthcare, but diagnosing HF at an early stage can be challenging. Identifying patients at risk of HF and initiating early treatment is crucial for their outcomes. Using the variables gender, age, multimorbidity (MM) level, and socioeconomic status (SES), we aimed to study the possibility of identifying individuals at high risk of HF diagnosis within two years.
METHODS: A longitudinal registry-based study, including 961,190 inhabitants aged from 20 years onwards without a HF diagnosis living in southern Sweden during 2015. Logistic regression was applied to estimate the OR of HF diagnosis within two years by adjusting for the variables gender, age, MM level, and SES. Linear predictions were made based on models by adding these variables in steps. Each model was compared with the previous model using a likelihood-ratio test. The optimal cutoff point for sensitivity and specificity was calculated using the Youden method.
RESULTS: Age had the highest OR of HF diagnosis within two years, followed by MM level, gender, and SES. ROC (Receiver Operating Characteristic) analysis, including these variables in steps, generated an increasing AUC (area under the curve), from 0.5144 to 0.9379. When all four variables were included in the model, an optimal cutoff point according to Youden was established at 1.15%, which predicted the probability with a sensitivity of 87.69% and specificity of 78.48%. The positive predictive value was 4.78%, and the negative predictive value was 99.81% for the whole adult population; for those aged 70 years and older, it was 21.02% and 98.99%; and for those aged 80 years and older, it was 33.62% and 98.09%, respectively.
CONCLUSIONS: Age was the most important factor for predicting the probability of HF diagnosis within two years in our study, followed by MM level, gender, and SES. These findings may help identify population groups at increased risk of HF in whom targeted case-finding strategies could be evaluated in future studies.
PMID:42402558 | DOI:10.1186/s12872-026-06230-9