Performance of Malnutrition Screening Tools on People With Chronic Diseases: A Bivariate Meta-Analysis

Vollständiger Abstract

Worum geht es in dieser Arbeit?

Background: Malnutrition significantly impacts mortality and morbidity in patients with chronic diseases. Accurate screening tools are necessary for the early identification and management of this condition. However, limited meta-analyses on screening tools designed to assess malnutrition in patients with chronic diseases exist. Purpose: This study was designed to evaluate the clinical efficacy of several screening tools widely used to detect malnutrition in patients with chronic diseases in clinical settings. Three tools were addressed in this meta-analysis, including the Malnutrition Screening Tool (MST), Malnutrition Universal Screening Tools (MUST), and Nutritional Risk Screening 2002 (NRS-2002). Methods: Seven electronic databases, including CINAHL-EBSCO, Cochrane, Embase, OVID-MEDLINE, PubMed, Scopus, and Web of Science, were searched systematically from their respective inception dates to October 29, 2023. Studies designed to evaluate the sensitivity and specificity of malnutrition using the Patient-Generated Subjective Global Assessment as the reference standard were included. Bivariate and random effects were used to summarize the following outcome variables: sensitivity, specificity, positive likelihood ratio, negative likelihood ratio, diagnostic odds ratio, and area under the curve. All of the statistical analyses were conducted using STATA software. Results: Twenty-six studies were included in the analysis. The sensitivity of MST in detecting malnutrition (.78, 95% confidence interval [CI] [.62, .88]) was higher than that of either MUST (.74, 95% CI [.68, .80]) or NRS-2002 (.67, 95% CI [.61, .71]). The respective specificities of NRS-2002, MST, and MUST were comparable (.88, 95% CI [.83, .92] vs. .82, 95% CI [.64, .90] vs. .80, 95% CI [.73, .89]). Similarly, in terms of malnutrition detection accuracy, the three tools had similar areas under curve (.82, 95% CI [.79, .85] vs. .87, 95% CI [.84, .90] vs. .79, 95% CI [.75, .82], respectively). However, Fagan nomograms showing positive and negative likelihood ratios of 0.1 indicate all three screening tools have a minimal effect rating with regard to detecting malnutrition. Conclusions: MST, MUST, and NRS-2002 all present a good level of accuracy in detecting malnutrition in patients with chronic diseases. Thus, it is recommended that nurses, physicians, dieticians, and other health care workers use these tools in daily practice. Further investigations are warranted to validate these findings.