<b>Objective:</b> To assess the performance of CORB score in CAP subjetcts for predicting in-hospital death, death within 30 days of admission, the requirement for mechanical ventilation (MV) and vasopressor support. <b>Methods:</b> A retrospective cohort study. An alternative CORB score that replaced SpO2 ≤ 90% with the SpO2/FiO2 ratio was estimated. Crude and adjusted odds ratios were calculated for each variable. The AUROCs were calculated for each score and outcome. AUROCs were compared with the DeLong test, considering a p&lt;0.05 statistically significant. <b>Results:</b> 1811 subjects entered the analysis, 15.1% died in-hospital, 16.4% died within 30 days, 8.7% required MV and 9.7% vasopressor support. CORB had an AUROC 0.660 (95%CI: 0.623-0.697; p&lt;0.001) for in-hospital mortality, AUROC 0.657 (95%CI: 0.621-0.692; p&lt;0.001) for 30-day mortality, AUROC 0.637 (CI95%: 0.589-0.685; p&lt;0.001) for MV requirement and AUROC 0.635 (95%CI: 0.589-0.681; p&lt;0.001) for vasopressor support. CORB performance increases when SpO2/FiO2 &lt;300 is used as the oxygenation criteria in the prediction of requirement for MV and vasopressor support, AUROC of 0.700 (95%CI: 0.654-0.746; p&lt;0.001) and AUROC 0.702 (95%CI: 0.66-0.745; p&lt;0.001), respectively. CURB-65 score performs better than CORB for in-hospital mortality AUROC 0.727 (95%CI: 0.695-0.759; p&lt;0.001) and 30-day mortality AUROC 0.726 (95%CI: 0.695-0.756; p&lt;0.001). <b>Conclusions:</b> CORB score has a better performance than CURB-65 to predicting the need for MV and vasopressor support in CAP. Performance improves when SpO2/FiO2 &lt;300 is used as the oxygenation parameter in the CORB. CURB-65 score is superior in the prediction of mortality