Loading...
Publication
GLIM criteria for undernutrition diagnosis in hospitalized patients: a validation study
| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 40152 | 1.79 MB | Adobe PDF |
Authors
Advisor(s)
Abstract(s)
Objetivo: A desnutrição é motivo de preocupação em ambientes hospitalares, devido à sua elevada prevalência e ao impacto desta condição nos resultados do tratamento. Os critérios da Global Leadership Initiative on Malnutrition (GLIM) foram recentemente propostos com o objetivo de padronizar o diagnóstico da desnutrição. O objetivo deste estudo foi avaliar a validade concorrente e preditiva do GLIM, através da associação com o tempo de internamento, numa amostra de adultos e idosos hospitalizados.
Metodologia: Estudo observacional prospetivo realizado numa amostra de adultos e idosos internados num Hospital Universitário do Porto. O NRS-2002 foi aplicado para rastreio da desnutrição. A validade concorrente entre o GLIM e o PG-SGA foi avaliada pelo cálculo da concordância, Kappa de Cohen (k), sensibilidade e especificidade. A validade preditiva foi avaliada através da associação independente do GLIM, com o tempo de internamento, utilizando a Regressão Proporcional de Cox.
Resultados: A sensibilidade e especificidade do GLIM foram de 70,1% e 79,4%, respectivamente. A concordância entre GLIM e PG-SGA foi de 58,4% e Kappa 0.484. Os desnutridos, segundo o GLIM, têm maior probabilidade de um tempo de internamento prolongado, uma vez que desnutrição moderada 0,564 (IC 95%: 0,431; 0,738) e desnutrição grave, 0,639 (IC 95% 0,505- 0,809), associam-se a menor probabilidade de alta para domicílio.
Conclusões: O GLIM apresentou concordância moderada com o PG-SGA e especificidade satisfatória. O presente estudo mostrou que apresentar desnutrição moderade ou grave diminui a probabilidade de ter alta para domicílio e, portanto, a desnutrição de acordo com o GLIM está associada a um tempo de internamento mais longo. O GLIM pode ser utilizado na prática clínica para diagnosticar desnutrição em doentes hospitalizados.
Aim: Undernutrition is a cause for concern in hospital environments due to its high prevalence and the impact of this condition on treatment outcomes. The Global Leadership Initiative on Malnutrition (GLIM) criteria were recently proposed, with the aim of standardising the diagnosis of undernutrition. The present study aims to determine the concurrent validity of GLIM and its predictive validity, through the association with LOS, in a sample of adults and older adults hospitalized patients. Methods: Prospective observational study, in a sample of adults admitted to a University Hospital in Porto. NRS-2002 was applied for undernutrition screening. Concurrent validity between the GLIM and the PG-SGA was assessed by calculating agreement, Cohen´s Kappa, sensitivity and specificity. The predictive validity was studied through the independent association of the GLIM criteria, with the length of stay, using Cox Proportional Regression. Results: GLIM had a sensitivity of 70.1% and a specificity of 79.4%; agreement between GLIM and PG-SGA was 58.4 %, Kappa 0.484. Undernutrition according to GLIM decreased the probability of discharge to usual residence: 0.564 (95% CI: 0.431; 0.738) for moderate undernutrition and 0.639 (95% CI 0.505- 0.809), for severe undernutrition. Conclusions: GLIM showed a moderate agreement with PG-SGA and a satisfying specificity. The findings of the present study also showed that being undernourished according to the GLIM criteria, both moderately and severely, decreases the probability of being discharged home or to usual residence, and thus undernutrition according to GLIM is associated with longer LOS. The GLIM criteria can be used in clinical practice to diagnose undernutrition in hospitalized patients.
Aim: Undernutrition is a cause for concern in hospital environments due to its high prevalence and the impact of this condition on treatment outcomes. The Global Leadership Initiative on Malnutrition (GLIM) criteria were recently proposed, with the aim of standardising the diagnosis of undernutrition. The present study aims to determine the concurrent validity of GLIM and its predictive validity, through the association with LOS, in a sample of adults and older adults hospitalized patients. Methods: Prospective observational study, in a sample of adults admitted to a University Hospital in Porto. NRS-2002 was applied for undernutrition screening. Concurrent validity between the GLIM and the PG-SGA was assessed by calculating agreement, Cohen´s Kappa, sensitivity and specificity. The predictive validity was studied through the independent association of the GLIM criteria, with the length of stay, using Cox Proportional Regression. Results: GLIM had a sensitivity of 70.1% and a specificity of 79.4%; agreement between GLIM and PG-SGA was 58.4 %, Kappa 0.484. Undernutrition according to GLIM decreased the probability of discharge to usual residence: 0.564 (95% CI: 0.431; 0.738) for moderate undernutrition and 0.639 (95% CI 0.505- 0.809), for severe undernutrition. Conclusions: GLIM showed a moderate agreement with PG-SGA and a satisfying specificity. The findings of the present study also showed that being undernourished according to the GLIM criteria, both moderately and severely, decreases the probability of being discharged home or to usual residence, and thus undernutrition according to GLIM is associated with longer LOS. The GLIM criteria can be used in clinical practice to diagnose undernutrition in hospitalized patients.
Description
Trabalho Complementar apresentado à Universidade Fernando Pessoa como parte dos requisitos para obtenção do grau de Licenciada em Ciências da Nutrição
Keywords
Desnutrição Tempo de internamento hospitalar GLIM Validade concorrente Validade preditiva Undernutrition Length of hospital stay Concurrent vailidity Predictive validity