Evidence generation for diabetes prevention and treatment through the integration of clinical medicine, nutritional epidemiology, and behavioral science

Abstract Diet and other lifestyle factors are central to diabetes prevention and management. Effective dietary and lifestyle support requires consideration of differences in pathophysiology, food culture, and living circumstances. This narrative review summarizes our research integrating clinical medicine, nutritional epidemiology, and behavioral science, with a particular focus on diet quality. Analyses of large US cohorts showed that higher-quality plant-based diets were associated with lower risks of stroke and mortality. In longitudinal biomarker analyses, increases in healthful plant-based diet scores were associated with favorable changes in high-sensitivity C-reactive protein, whereas increases in unhealthful plant-based diet scores were associated with unfavorable changes in inflammatory biomarkers. Metabolite profiles related to healthful plant-based diets were also associated with lower type 2 diabetes risk. Building on earlier work on adiponectin and type 1 diabetes, our studies in Japan showed that even small weight changes were associated with type 2 diabetes risk, depending on age and baseline body mass index. Visceral fat correlated with ectopic fat, and fast eating and eating late at night were associated with greater pancreatic fat deposition. In older adults, lower sweet taste sensitivity was associated with frailty and poorer mental health-related quality of life. These findings suggest that diet quality and lifestyle should be assessed together with each person’s metabolic characteristics and living circumstances. Intervention and implementation studies based on this largely observational evidence are the next step toward individualized diabetes prevention and care.