Abstract
PURPOSE: The purpose of this study was to analyze the validity of the nutrition subscale from the Braden Scale for Predicting Pressure Sore Risk in hospitalized patients.
DESIGN: A prospective, quasi-experimental, repeated-measures design guided data collection and analysis.
SUBJECTS AND SETTINGS: One hundred seventy adult patients from 2 private hospitals located in urban areas in Southeastern Brazil, with a Braden Scale score of 18 or less, and who agreed to participate in the study were assessed between January and August 2006. Participants were primarily male (57.0%) and had a mean age of 67.0 +/- 15.4 years (mean +/- SD).
METHODS: Objective assessment and subjective global assessment of nutritional status were performed on admission. Every 2 days, patients deemed at potential risk for pressure ulcer development underwent evaluation of protein-energy intake, skin assessment, and repeated risk assessment for pressure ulcer development via the Braden Scale. Univariate and multivariate logistic regression analyses were used to assess the predictive power of nutritional variables related to risk for pressure ulcer development.
RESULTS: The mean length of stay among patients was 17.8 +/- 16.8 days. Multivariate regression analysis revealed that serum albumin levels (odds ratio = 5.226; P < .001) and subjective global nutritional assessment (odds ratio = 3.246; P < .001) were the best nutritional predictors of pressure ulcer development.
CONCLUSION: We did not find the Braden nutrition subscale score to be predictive for pressure ulcer development in hospitalized patients. Serum albumin levels and subjective global nutritional assessment were the best nutritional predictors of pressure ulcer development.