Reply to Frencken et al

To the Editor—We thank Dr Frencken and his colleagues [1] for their interesting comments regarding a recently issued article where we analyzed the rate of cardiovascular events (CVEs) in a population affected by community-acquired pneumonia (CAP). In particular, we found that, in a population of 1182 patients hospitalized with CAP, CVEs including myocardial infarction, stroke, heart failure, atrial fibrillation, and cardiovascular death occurred in approximately 30% of patients within 30 days of their admission and that such complications were associated with 30-day mortality [2]. Frencken et al [1] question that we treated CVEs as time-fixed exposure and did not consider the time-dependency of CVEs, which could inflate the risk in the CVE group. Accordingly, we reanalyzed CVEs as a time-dependent covariate and recalculated their impact on mortality in a new Cox model analysis, which confirmed the association between CVEs and mortality (hazard ratio [HR], 4.76; 95% confidence interval [CI], 3.14–7.21; P < .001). The association remained significant when the model was adjusted for age, pneumonia severity score, and comorbidities (HR, 2.52; 95% CI,1.61–3.97; P < .001).