OBJECTIVE Recent public health recommendations emphasize adopting a healthful dietary pattern, but evidence is scarce on whether incremental diet quality changes have an impact on long-term diabetes prevention. We aim to evaluate diet quality changes during a 4-year period and subsequent 4-year type 2 diabetes incidence.

RESEARCH DESIGN AND METHODS Participants of prospective cohorts, the Nurses’ Health Study (NHS), NHS II, and the Health Professionals Follow-up Study, who were free of diabetes at baseline (n = 124,607), were observed for ≥20 years. Diet quality, reflected by the Alternate Healthy Eating Index (AHEI) score, was assessed every 4 years to calculate changes.

RESULTS We documented 9,361 cases of type 2 diabetes during 2,093,416 person-years of follow-up. A >10% decrease in AHEI score over 4 years was associated with a higher subsequent diabetes risk (pooled hazard ratio 1.34 [95% CI 1.23–1.46]) with multiple adjustment, whereas a >10% increase in AHEI score was associated with a lower risk (0.84 [0.78–0.90]). Greater improvement in diet quality was associated with lower diabetes risk across baseline diet quality status (P for trend ≤ 0.001 for low, medium, or high initial diet quality) and baseline BMI (P for trend ≤ 0.01 for BMI <25, 25–29, or 30 kg/m²). Changes in body weight explained 32% (95% CI 24–41) of the association between AHEI changes (per 10% increase) and diabetes risk.

CONCLUSIONS Improvement in overall diet quality is associated with a lower risk of type 2 diabetes, whereas deterioration in diet quality is associated with a higher risk. The association between diet quality changes and diabetes risk is only partly explained by body weight changes.