Response to: Correspondence on “Electrocardiographic findings and prognostic values in patients hospitalised with COVID-19 in the World Heart Federation Global Study” by Marateb et al

A second general issue is that Marateb and Mañanas considered that we intended to build or validate a clinical prediction model, defined in the TRIPOD statement as a tool ‘that combines multiple predictors by assigning relative weights to each predictor to obtain a risk or probability’ for diagnosis or prognostication, to aid healthcare providers in the decision-making.6 However, we just aimed to ‘prospectively evaluate the prognostic value of major ECG abnormalities in patients hospitalised with COVID-19 in a comprehensive pancontinental study’ in ‘a multicentric cohort study’. [...]we followed the guidelines for reporting observational studies, the Strengthening the Reporting of Observational Studies in Epidemiology statement.7 The corresponding checklist was filled out during the submission, and a statistical reviewer revised the paper as a routine in all Heart original articles. Risk factors used in the multivariate model were all prevalent conditions. [...]their contribution to the outcomes does not depend on the follow-up period. [...]unfortunately, we do not have data on when ECG acquisition was performed after admission. There is no clear benefit of computing a few extra days free of outcomes (from a clinical perspective) between groups. [...]treating the outcome dichotomously for this short follow-up period makes more sense, and for a dichotomous variable, the MLR is the appropriate choice.