BACKGROUND: Nutritional assessment is a fundamental part of the treatment of patients hospitalized in the ICU, allowing the implementation of interventions appropriate to the identified requirements. Since the risk of malnutrition is a modifiable factor, its correct management can positively influence hospital evolution. This study aims to test the impact of the incorporation of nutritional screening and assessment on mortality and length of stay in patients hospitalized in an Intensive Care Unit in Cali, Colombia, during the years 2019 and 2021-2022. METHODS: This is a historical cohort epidemiological study where one cohort consisted of 114 patients who received a standard nutritional screening (interpretation of body mass index and its clinical impression). The other cohort of 630 patients was those exposed to screening with the Malnutrition Universal Screening Tool (MUST) scale. Hematological, clinical, and nutritional variables were considered and their relationship with adverse events, length of hospital stay, and discharge status. RESULTS: There were significant differences between the two cohorts ( CONCLUSION: Implementing the MUST screening method and specific nutritional interventions resulted in a significant improvement in patient mortality figures. In addition, the predictive mortality model revealed that emesis and leukopenia increased the probability of death.