<b><i>Background/Aims: </i></b>In this corollary analysis of the EXCITE study, we looked at possible differences in baseline risk factors and mortality between subjects excluded from the trial because non-eligible (n=216) or because eligible but refusing to participate (n=116). <b><i>Methods: </i></b>Baseline characteristics and mortality data were recorded. Survival and independent predictors of mortality were assessed by Kaplan-Meier and Cox regression analyses. <b><i>Results: </i></b>The incidence rate of mortality was higher in non-eligible vs. eligible non-randomized patients (21.0 vs. 10.9 deaths/100 persons-year; P<0.001). The crude excess risk of death in non-eligible patients (HR 1.96; 95% CI 1.36 to 2.77; P<0.001) was reduced after adjustment for risk factors which differed in the two cohorts including age, blood pressure, phosphate, CRP, smoking, diabetes, triglycerides, cardiovascular comorbidities and history of neoplasia (HR 1.60; 95% CI 1.10 to 2.35; P=0.017) and almost nullified after including in the same model also information on deambulation impairment (HR 1.16; 95% CI 0.75 to 1.80; P=0.513). <b><i>Conclusions: </i></b>Deambulation ability mostly explains the difference in survival rate in non-eligible and eligible non-randomized patients in the EXCITE trial. Extending data analyses and outcome reporting also to subjects not taking part in a trial may be helpful to assess the representability of the study population.