Malnutrition risk and its association with adverse outcomes in a Belgian cohort of community-dwelling adults aged 80 years and over

Objectives: To investigate the prevalence of malnutrition risk and its association with adverse outcomes in a Belgian cohort of community-dwelling adults aged >= 80 years, a worldwide growing age-group. Methods: In the BELFRAIL cohort, malnutrition risk was evaluated with the Mini Nutritional Assessment (MNA total score <24) and prealbumin levels (<20 mg/dl). Agreement between them was assessed with Kohen's kappa coefficient. Association with first unplanned hospitalization (3.0 +/- 0.25 years follow-up) and mortality (5.1 +/- 0.25 years follow-up) was investigated with survival analysis and Cox multivariate regression. Results: Out of 567 BELFRAIL participants, 556 (98.1%) had MNA and 545 (96.1%) prealbumin levels. Sixty-eight (12.2%) were at risk of malnutrition based on MNA and 69 (12.7%) based on prealbumin, with very poor agreement between them (Kappa = 0.024, 95% CI -0.064, 0.112). For both MNA and prealbumin, participants with malnutrition risk had lower physical and cognitive performance tests' scores. They had no higher risk for first hospitalization compared to those without malnutrition risk, but higher risk for all-cause mortality even after adjustment for multimorbidity, inflammation, physical and mental functioning (HR 1.35 95%CI 0.92-1.97 for MNA; HR 1.46; 95%CI 1.01-2.12 for prealbumin). Conclusion: Malnutrition risk based on MNA or prealbumin was low in a Belgian cohort of community-dwelling adults aged >= 80 years. Physical and cognitive performance was lower in those with malnutrition risk, but malnutrition risk was not independently associated with hospitalization and mortality (except for malnutrition risk by prealbumin). Further research needs to investigate the best tool to assess malnutrition risk in this age group.