Agreement between Graz Malnutrition Screening (GMS) with subjective nutritional assessment instruments in hospitalized patients

Background: it is essential for an early nutritional intervention that utilizes effective and practical nutritional screening and evaluation tools to diagnose nutritional status, increasing the patient's survival. Objective: to evaluate the agreement of the Graz Malnutrition Screening (GMS) with subjective methods of nutritional evaluation in hospitalized patients. Methods: descriptive cross-sectional study with adults and elderly of both sexes evaluated within 48 hours of hospital admission. Nutritional status in cancer patients was identified by the Patient-Generated Subjective Global Assessment (PG-SGA (R)) and in the elderly by the Mini Nutritional Assessment Short-Form (MNA-SF (R)). GMS was applied in both groups and its efficiency was compared with that of PG-SGA (R) and MNA-SF (R). The agreement between the methods was evaluated by the kappa test, followed by assessment of diagnostic performance and correlation test. Results: of the 87 patients evaluated, 64.4% (56) presented nutritional risk according to GMS, while 49.4% (43) and 47.1% (41) indicated nutritional risk and malnutrition according to MNA-SF (R) and PG-SGA (R), respectively. GMS presented moderate agreement with PG-SGA (R) (p < 0.001) and MNA-SF (R) (p < 0.001), with high sensitivity, specificity, positive predictive value, and negative predictive value. Correlations were observed between the GMS score and both the PG-SGA (R) (p < 0.001) and MNA-SF (R) scores (p < 0.001). Conclusion: GMS was effective in detecting nutritional risk in hospitalized patients when compared to classic tools in the evaluation of nutritional status in hospitalized patients.