The subjective global assessment (SGA) defines nutritional and functional status of patients with the aim of identifying who could benefit from a nutritional intervention. The SGA was designed to evaluate the nutritional risk of infectious complications among the surgical patients. Later on, it became largely utilized to evaluate the nutritional status of other groups of patients, because it is reliable and easy to use. However, it is important to verify that the SGA is an adequate test also in other clinical situations. This review analyses recent literature regarding the actual utilization of the SGA in clinical situations frequently associated to malnutrition: kidney diseases, AIDS, cancer, aging. The reproducibility of SGA depends on the experience of the caregiver administering it. Its validation compared to more objective parameters like visceral proteins or body composition is not always obvious. In several studies the SGA does not seem enough sensitive to detect a beginning of malnutrition. Its specificity and predictive validity were demonstrated in all of the clinical situations analyzed. In conclusion, the SGA is a tool adequate to identify patients with a nutritional risk and who would benefit of a nutritional intervention aiming at preventing associated complications.