The association between dietary factors and the incident of type 2 diabetes: A case‐control study

Background The protective or harmful effects of dietary factors including soy products and other foods to diabetes are uncertain; results of researches of different ethnic groups worldwide are inconsistent. Dietary pattern analysis considers the interactions and cumulative effects among nutrients and foods, and can explore the relationship of nutrition and disease from dietary holistic perspective. Objective To explore the relationship between nutritional factors and the incidence of type 2 diabetes, clarify the potentional protective and harmful factors for diabetes and its impact strength. Methods Our study was a 1:2 (sex/age) match case‐control study. The cases subjects were diabetics who were diagnosed within 3 months and the controls were without disorder of glucose metabolism. New onset diabetes cases were recruited in the clinic and ward in the hospitals and community, and the controls were recruited in the community through the screen of fasting glucose. Face‐to‐face interviews by trained interviewers were conducted to collect information on their socio‐demographic characteristics, lifestyle, and diet using structured questionnaires. Factor analysis was used to construct dietary pattern. The study participants were divided into quartiles (Q1 (lowest), Q2, Q3, and Q4) by food and nutrient intake, factor scores separately. Logistic regression was used to investigate the association of foods, nutrients and dietary patterns with Diabetes. Trend tests were performed by treating quartiles variables as continuous variables. Results Our study enrolled 384 cases (male 162, female 222) and 768 controls ( male 324, female 444). Multiple logistic regression analysis indicated that the highest quartile intake of whole‐wheat bread, total fruits, total milk products, fresh beans and dry beans can reduce the risk of diabetes by 52% (95% CI: 23%–69%), 54% (95% CI: 28%–71%), 42% (95% CI: 9%–63%) and 63%(95% CI: 41%–77%) respectively, and the risk also lower with the intakes ( p <0.01). When adjusted by some potential confounding factors, the benefit effect of skim milk disappears but whole milk still shows a strong protective effect of diabetes. Moderate intake of total grain was inversed associated with diabetes, and reduce 40%–50% of diabetes risk in the second and the third quartile, but this health effect disappears when the grain intake reached the highest quartile. Processed meats, fishes could increase the risk of diabetes 54%–70% in their highest intake quartile compared with the lowest intake. Higher intake of total dietary fat increases the risk of diabetes by 33% and there has the linearly trend. Among the four dietary patterns derived from the factor analysis, the pattern of high fruit has the protective effect to diabetes and has the strong linear trend, the highest reduction is 58% (95%CI: 32–74%) in the highest quartile score. The pattern of high meat low grain may be harmful to diabetes, but the effect disappears after adjusted for the confounding factors. Conclusion Greater intake of whole grain, milk, fruits, fresh beans and other dry beans can reduce the risk of diabetes. Moderate total grain may benefit to diabetes but does not show linear trend. The pattern of greater fruit, moderate vegetable and fishes has beneficial effect on diabetes. Support or Funding Information Funding: This research was funded with grants from the Department of Health of Guangdong Province (No: A2008158); the Program for New Century Excellent Talents in University, Ministry of Education, China (NCET‐06‐0719); the Chinese Nutrition Society in 2006; the Danone Institute in 2009(DIC2009‐10); National Nature Science Foundation (NSFC‐81202197)