To evaluate the ability of SMART-COP (systolic blood pressure, multilobar infiltrates, albumin, respiratory rate, tachycardia, confusion, oxygen and pH) score to predict the need for intensive care unit (ICU) admission and mortality among patients with non-ventilator-associated hospital-acquired pneumonia (NV-HAP) and to compare ICU-hospitalized patients with those followed-up in the clinic, as well as the patients who survived with those who died in the ICU, in terms of clinical and laboratory parameters. A total of 203 patients (aged > 65 years) who were diagnosed with NV-HAP while staying in the geriatric clinic were enrolled in this retrospective observational study. Patient information was retrieved from hospital files. In a total of 203 patients with NV-HAP, the rate of ICU admission was 77.3% and the rate of mortality was 40.9%. The SMART-COP score was significantly higher in those admitted to the ICU and those died in the ICU (ICU nonsurvivors). The rate of ICU mortality was 52.9%. The SMART-COP score had significantly poor to moderate ability to predict the need for ICU admission (area under the curve [AUC] = 0.583) and both in-hospital mortality (AUC = 0.633) and ICU mortality (AUC = 0.617) with low sensitivity. The regression analysis revealed that a one-unit increase in SMART-COP score resulted in a 1.2-fold increase in both the hospital and ICU mortality (P < 0.05 for both) and 1.1-fold increase in ICU admission (P = 0.154). The SMART-COP score has poor to moderate ability to predict the need for ICU admission, in-hospital mortality and ICU mortality, and a one-unit increase in the SMART-COP score significantly increases the risk of both hospital and ICU mortality. Geriatr Gerontol Int 2024; ••: ••-••.