Background: Congestive heart failure (HF) is a chronic, frequent and disabling condition but with a modifiable course and a large potential for improving. The aim of this project was to develop a clinical prediction model of biological and non biological factors in patients with first diagnosis of HF that facilitates the risk-stratification and decision-making process at the point of care.
Methods and results: Historical cohort analysis of 600 patients attended at three tertiary hospitals and diagnosed of a first episode of HF according Framingham criteria. There were followed 1 year. We analyzed sociodemographic, clinical and laboratory data with potential prognostic value. The modelling process concluded into a logistic regression multivariable analysis and a predictive rule: PREDICE SCORE. Age, dependency for daily basic activities, creatinine clearance, sodium levels at admission and systolic dysfunction diagnosis (HF with left ventricular ejection fraction 〈 40%) were the selected variables. The model showed a c-statistic of 0.763. PREDICE Score, has range of 22 points to stratifications of 1-year mortality.
Conclusions: The follow-up of 600 patients hospitalized by a first episode of congestive HF, allowed us to obtain a predictive 1 year mortality model from the combination of demographic data, routine biochemistry and easy handling social and functional variables at the point of care. The variables included were non-invasive, undemanding to collect, and widely available. It allows for risk stratification and therapeutical targeting and may help in the clinical decisions process in a sustainable way.