Introduction: Despite the well-established association between diet and chronic disease risk, there is limited evidence on the association between diet change and long-term risk of diabetes in a community-based setting.
Hypothesis: Decline in diet quality is associated with an increased risk for incidence of diabetes in the Atherosclerosis Risk in Communities (ARIC) Study.
Methods: We included 8285 ARIC participants who had complete dietary intake assessments by food frequency questionnaire (FFQ) at both visits 1 (1987-1989) and 3 (1993-1995) and had no prevalent diabetes, cardiovascular disease, or cancer prior to visit 3. Change in diet quality was assessed using the Healthy Eating Index (HEI)-2015 and the Alternative HEI (AHEI)-2010 scores calculated at the two study visits. Change in scores was classified into five categories from moderate/large decrease to moderate/large increase. We used adjusted Cox models to estimate hazard ratios (HR) and 95% confidence intervals (CI) for the association between HEI-2015/AHEI-2010 score changes and incidence of diabetes, which is defined using doctor diagnosis, medication use, or elevated glucose.
Results: There were 2,136 cases of incident diabetes during a median of 18.05 years of follow-up (124,639 person-years). For HEI-2015, there were no statistically significant association between dietary score change and diabetes risk, comparing the five categories of changes (adjusted HRs and 95% CIs for the five categories from decrease to increase were 1.06 [0.90-1.25],1.04 [0.90-1.21],1.00 (ref.), 0.91 [0.79-1.05], 1.04 [0.91-1.20]). Similar trends were found in AHEI-2010 scores changes, but for the moderate to large decrease group, the decrease in dietary scores was associated with a higher diabetes risk (HR 1.23, 95%CI 1.03 to 1.46), compared to the stable group.
Conclusion: The results suggest that deterioration of diet quality is associated with a higher risk of diabetes.