This study aimed to evaluate the association between methionine intake and diabetes prevalence in Chinese adults and explore whether the association was source-specific. Data from 12,849 adults aged ≥20 years old were used from the China Health and Nutrition Survey during 1997-2011. Diabetes was diagnosed as self-reported and/or when blood tests results met the diagnostic criteria. A 3-day, 24-h recall was used to assess different sources of methionine. Multivariable mixed linear regression was used to examine the associations. Across the quartiles of total methionine intake, the odds ratio (ORs, 95% CI) of diabetes were 1.00, 1.49 (1.21 to 1.82), 1.72 (1.37 to 2.15), and 2.53 (1.97 to 3.23). In the subgroup analysis, similar trends were observed in both animal and plant methionine. There was a significant interaction between urbanization and diabetes. The positive association was only significant in those who lived in low or medium urbanization areas. The ORs (95% CI) were 1.00, 1.27 (0.85 to 1.88), 1.56 (1.01 to 2.39), and 1.79 (1.09 to 2.95) for medium urbanization, respectively. No interaction was identified when stratified by different methionine sources. In conclusion, methionine intake was positively associated with diabetes independent of food source, and it was modified by urbanization levels.
Keywords: animal methionine; diabetes; interactive effect; plant methionine; urbanization.