Nutritional status in older persons according to healthcare setting: A systematic review and meta-analysis of prevalence data using MNA(®)

BACKGROUND & AIMS: Old persons are more likely to suffer from malnutrition, which may result in higher dependency in activities of daily living. We aimed to provide a quantitative synthesis of prevalence data on malnutrition and its risk as assessed by the Mini Nutritional Assessment across different healthcare settings. The association between nutritional status and setting-related level of dependence was also investigated.

METHODS: Non-interventional studies published as full-text articles in English up to 31th December 2014 were searched for in PubMed and by reviewing references of eligible articles. Meta-analysis and meta-regression of potential sources of heterogeneity were conducted.

RESULTS: A total of 240 studies/795 citations - providing 258 setting-specific prevalence estimates (113,967 subjects) - fulfilled inclusion criteria for meta-analysis. Prevalence of malnutrition differed significantly across the healthcare settings considered: community, 3.1% (95%CI, 2.3-3.8); outpatients, 6.0% (95%CI, 4.6-7.5); home-care services, 8.7% (95%CI, 5.8-11.7); hospital, 22.0% (95%CI, 18.9-22.5); nursing homes, 17.5% (95%CI, 14.3-20.6); long-term care, 28.7% (95%CI, 21.4-36.0); rehabilitation/sub-acute care, 29.4% (95%CI, 21.7-36.9). For every setting significant heterogeneity in individual study results was observed (I(2) ≥80%, P < 0.001) and meta-regression showed that study quality was the most important determinant. Finally, meta-regression of all the studies included showed that both malnutrition and its risk were directly associated with the setting-related level of dependence (P < 0.001). However, despite multiple adjustments, residual heterogeneity remained high.

CONCLUSION: We provided updated estimates of malnutrition and its risk in different healthcare settings. Although the level of dependence appears to be an important determinant, heterogeneity in individual study results remained substantially unexplained. The cause-effect relationship between nutritional status and level of dependence deserves further investigation.