tools primarily use clinical and laboratory variables, with physical activity and social problems not considered in these scores. Therefore, using data from the Kitakawachi Clinical Background and Outcome of Heart Failure (KICKOFF) registry and a machine learning model, we developed a clinical score (the KICKOFF score) incorporating physical and social factors, in addition to clinical factors, to predict long-term mortality in patients with AHF. 11 The model's performance was evaluated with temporal validation in a previous study. 11 However, the KICKOFF score has not been validated externally in different geographic areas of Japan.The KICKOFF registry and the Kochi Registry of
Because of its increased prevalence in the aging population, acute heart failure (AHF) is an important and common cause of hospitalization, morbidity, and mortality in cardiology. 1 Some studies have revealed that the median age of patients with AHF is 80 years, and that these patients have various comorbidities, physical disorders, and social problems. 2,3 A decline in physical activity is a valuable predictor of adverse outcomes in patients with heart failure (HF), 4,5 and social problems are associated with HF. 6 Some mortality prediction models and scoring systems for AHF risk stratification have been reported and validated in Japanese patients with AHF. 7-10 Most of these