Verdu-Rotellar JM, Abellana R, Vaillant-Roussel H, et al. - Based on variables easily accessible in a community setting, the HEFESTOS score was developed in this study as well as validated in an external European cohort. This score enabled proper prediction of the risk of death/hospitalization during the first 30 days following a heart failure (HF) decompensation episode.

• In this multinational study (HEFESTOS), a derivation cohort comprising HF patients recruited in 14 primary healthcare centers in Barcelona as well as a validation cohort from primary healthcare in 9 other European countries were included.

• In the derivation and validation cohorts, there were 561 and 250 patients, respectively, and mortality/hospitalization during the first 30 days after a decompensation episode was 30.5% and 26%, respectively.

• A score of risk was created using multivariable logistic regression models, and identified predictors included worsening of dyspnea, orthopnea, paroxysmal nocturnal dyspnea, crackles, New York Heart Association functional class III/IV, oxygen saturation ≤ 90%, heart rate > 100 b.p.m., and previous hospitalization due to HF.

• In the derivation and validation cohorts, an area under the curve of 0.807 and 0.73, respectively, was generated by the model.

• With respect to the likelihood of hospitalization/death, three risk groups were described: low <5%, medium 5–20%, and high >20%.

• In the derivation cohort, outcome incidence for the low-risk, medium risk, and high risk groups was 2.7%, 12.8%, and 46.2%, respectively; in the validation cohort, it was 9.1%, 12.9%, and 39.6%.