Background: Malnutrition is an often-overlooked yet potentially crucial factor influencing COVID-19 outcomes. Poor nutritional status weakens immune function, increases infection susceptibility, and worsens prognoses in hospitalized patients. However, its specific role in COVID-19 mortality remains insufficiently characterized. The aim of the study was to assess the impact of malnutrition, as determined by the Nutritional Risk Score (NRS-2002), on in-hospital mortality. Methods: This retrospective, single-center study analyzed 222 patients hospitalized with COVID-19 during the Delta variant predominance. Thirty-one patients died during hospitalization. Malnutrition (NRS ≥ 3) emerged as a strong predictor of in-hospital mortality in univariate Cox proportional hazard models, both before and after adjustment for potential confounders. Adjusted analyses used 10 different sets of three out of five mortality-related variables. Results: Hazard ratios for malnutrition ranged from 3.19 to 5.88 (p < 0.01 for all models), highlighting its substantial impact on mortality risk. The high Nagelkerke’s R² values (0.66–0.77) indicate that the models explained a significant proportion of mortality variance. Nutritional status plays a critical role in COVID-19 survival among hospitalized patients. Conclusions: Given its simplicity and effectiveness, integrating the NRS-2002 into routine clinical assessments may help identify high-risk patients early. Future research should explore whether early nutritional interventions can mitigate the mortality risks associated with malnutrition in severe COVID-19 cases or patients with other infectious diseases or acute inflammation.

Activities of Daily Living (ADL) are fundamental tasks for individuals to manage their basic needs. Our study aims to examine ADL at admission (adADL) and the Pre-COVID-19 to Admission ADL Difference (ADL-change) as potential predictors of in-hospital mortality. This is a retrospective analysis of clinical data (including the Katz index for ADL) from 141 older patients aged at least 65 years hospitalized in a COVID-19-dedicated unit (not requiring ICU) from September 2021 until January 2022 in Poznań, Poland. Thirty patients (21.3% of all) died during hospitalization. Non-survivors were older than survivors, exhibited lower oxygen saturation, more severe inflammation, higher D-dimer concentrations, and were more commonly prescribed antibiotics. The AUC for in-hospital mortality was for adADL: 0.7417 (95% CI: 0.6478–0.8357; p < 0.0001) and for ADL-change: 0.6869 (95% CI: 0.579–0.7928; p = 0.0018). The corresponding cut-offs were 0 for adADL and 3 for ADL-change. Cox proportional hazard models yielded hazard ratios of 3.57 (95% CI 1.57–8.10; p = 0.0024) for adADL and 3.78 (95% CI 1.49–9.54; p = 0.005) for ADL-change. ADL assessment offers valuable insights into in-hospital mortality among older COVID-19 patients. Monitoring ADL in these patients indicates high-risk individuals for in-hospital death. Integrating ADL into routine clinical practice might enhance care for older patients.