Dietary pattern of non-frail very old people and relationship with underweight, muscle mass and strength and gait speed test

Abstract Objectives To identify the dietary pattern of non-frail very old people and to evaluate the association of this pattern with body composition, strength, and gait speed. Methods Cross-sectional study with a sample of 96 old people aged 80 or over, non-frail. Food consumption was analyzed using three non-consecutive food records. Muscle mass was assessed using arm muscular area. Muscle strength was measured by grip strength. Cluster analysis was used to distinguish dietary patterns. Bivariate analysis and multivariate Poisson Regression were performed, exploring the relationship between dietary patterns and specific independent variables. Results Two dietary patterns were identified, labeled healthy and traditional. The prevalence of underweight among the old people of the traditional pattern was 10% (95%CI 1.01-1.20) higher than among the old people of the healthy pattern and this prevalence was practically maintained (PR 1.09; 95%CI 1.00-1.18) in the model adjusted by functionality. Regarding the classification of the arm muscular area, the old people of the traditional pattern presented 15% (95%CI 1.00-1.32) more prevalence of low muscle mass, when compared with the old people of the healthy pattern. This prevalence in the functionality-adjusted model was no longer statistically significant. No association was found between dietary pattern and strength and gait speed. Conclusions The findings showed that old people who adhere to healthy dietary patterns have a lower risk of underweight and that low muscle mass is probably more associated with functionality than with dietary patterns.