Abstract
BACKGROUND: Dietary habits throughout life significantly influence health in old age; yet, little is known about the relationship between meal pattern and mortality among older adults. OBJECTIVE: This study aimed to prospectively investigate the association between meal pattern variables and 15-year all-cause mortality in a cohort of 70-year-olds from the Gothenburg H70 study, considering relevant covariates. DESIGN: A total of 551 individuals (321 women and 230 men) were included. Dietary intake was assessed using the diet history method, reflecting intake over the preceding 3 months. Meal patterns were described by the usual daily frequency of main meals, light meals, snacks, beverages, and total intake occasions (IO). Statistical analyses included Cox proportional hazards regression, Student's t-test, and Chi-square test. RESULTS: Subjects who were deceased at follow-up had a higher prevalence of undernutrition risk indicators (based on low body mass index [BMI], weight/appetite change, and eating difficulties) at baseline compared to those living 15 years later (P = 0.02). In the fully adjusted Cox model, individuals with high total intake frequencies (>5 per day) showed a significantly increased hazard ratio (1.51) for mortality compared to those with medium frequencies. Additionally, medium-high snack frequency (>2-3 snacks/day) was associated with an elevated mortality risk, independent of total energy intake and other covariates. DISCUSSION: These findings suggest a potential association between frequent daily IO, particularly snacks, and increased mortality risk, which is not fully explained by total energy consumption or other covariates. CONCLUSIONS: The 15-year follow-up provides a long-term view of meal patterns' impact on longevity, indicating that higher daily consumption frequencies may be associated with increased mortality risk between ages 70 and 85. Further research should examine the nutritional composition of various meal patterns to clarify these associations.