Abstract
AIM: Selecting optimal energy intake during diet therapy for older patients with diabetes mellitus is difficult because of the large differences in physical function and comorbid diseases. In Japan, although requirements for total energy intake are calculated by multiplying a person's standard bodyweight (BW) by the amount of physical activity, evidence supporting the application of this method among older people is limited. Therefore, we aimed to assess optimal energy intake by evaluating the relationship between energy intake and mortality in older patients. METHODS: We evaluated data from a 6-year prospective follow up of 756 older patients with diabetes mellitus, and the association between baseline nutrient intake and mortality. Total energy intake and nutrients were evaluated, and energy intake per actual BW was categorized into quartiles (Q). Cox regression analysis was used for statistical analyses. Energy intake per standard BW or age-related target BW was statistically analyzed using the same protocol. RESULTS: Analysis of energy intake per actual BW showed that hazard ratios for mortality was significantly higher in Q1 and Q4. Similar associations were found for energy intake per standard or target BW. Subgroup analysis showed that mortality rate was the lowest in Q2 in the young-old population and in Q3 in the old-old population. CONCLUSIONS: A U-shaped relationship was observed between energy intake per BW and mortality in older patients with diabetes mellitus, which suggests that the optimal energy intake per actual or target BW should encompass a wide range to prevent malnutrition and excessive nutrition in these patients. Geriatr Gerontol Int 2020; 20: 59-65.