Malnutrition Risk and Depressive Symptoms in Community-Dwelling Older Adults

Objectives: To describe nutritional status in a volunteer cohort of older adults attending community sarcopenia screening and to explore factors associated with risk of malnutrition, with secondary interest in whether nutritional risk coincides with, rather than excludes, higher adiposity. Design: Cross-sectional, single-centre, exploratory analysis. Setting: A community sarcopenia-screening programme in Vojvodina, Serbia, open to volunteers recruited through pensioners’ associations. Participants: Self-selected, community-dwelling volunteers aged 60 years and older. The analytic sample comprised 388 participants with a complete Mini Nutritional Assessment (MNA). Measurements: Nutritional status was assessed with the full MNA. Depressive symptoms (GDS-15), cognition (MoCA), education, handgrip strength, gait speed, and body composition, including visceral adipose tissue (VAT) by dual-energy X-ray absorptiometry, were recorded. The primary analysis was a multivariable logistic regression for risk of malnutrition (at-risk versus normal) using predictors assessed separately from the MNA; the continuous MNA score and body composition were secondary or exploratory. Results: No participant met the criterion for established malnutrition; 52 of 388 (13.4%, 95% CI 10.2 to 17.2) were at risk of malnutrition. Depressive symptoms were the dominant and most robust correlate, in both the binary model (odds ratio 1.29 per GDS-15 point, 95% CI 1.18 to 1.41) and the continuous-score model. Because the MNA includes a depression/dementia item, this association may partly reflect construct overlap and should not be interpreted as wholly external to the nutritional assessment. Lower education was suggestive in the binary model (odds ratio 2.08, 95% CI 0.95 to 4.56) and significant only for the continuous score. Nutritional risk was not concentrated in the leanest participants: after adjustment for age and sex, higher visceral adipose tissue was associated with at-risk status (odds ratio 1.46 per kg, 95% CI 1.02 to 2.10), but this attenuated after adjustment for depressive symptoms and showed no formal effect modification by sex. Comorbidity and marital status were not associated with nutritional risk. Conclusions: In this volunteer, relatively preserved cohort, risk of malnutrition was most strongly associated with depressive symptoms, although the MNA itself includes a depression/dementia item, with a weaker educational gradient, and was not a condition of low body mass; the relationship with visceral adiposity was exploratory and non-independent. These cross-sectional associations support nutritional screening coupled with evaluation of mood, and caution against excluding older adults with normal or high body mass.