中国2型糖尿病防治指南(2013年版)

  目的  了解浙江省德清县农村社区居民饮食偏好与2型糖尿病（T2DM）发病的关系，为采取相应的干预措施提供参考依据。  方法  采用动态前瞻性队列研究方法于2006 — 2008年、2011 — 2012年和2013 — 2014年随机抽取浙江省德清县2个、6个和7个农村社区未患糖尿病的28 233名居民作为研究对象进行问卷调查、体格检查和实验室检测；采用单因素和多因素Cox回归模型分析饮食偏好与T2DM发病的关系。  结果  至2015年11月，德清县28 233名农村社区居民中有2 773人失访（其中874人死亡），失访率为9.82 %。T2DM新发病例369例，发病密度为3.13/千人年；饮食均衡者、以素食为主者和以荤食为主者新发T2DM病例分别为227例、46例和96例，发病密度分别为2.59/千人年、2.88/千人年和6.67/千人年。单因素Cox回归分析结果显示，浙江省德清县以荤食为主的农村社区居民T2DM发病风险为饮食均衡居民的2.58倍（HR = 2.58，95 % CI = 2.03～3.27）；在控制了性别、年龄、文化程度、是否务农、是否吸烟、是否饮酒、是否饮茶、BMI、是否高血压、是否空腹血糖受损、糖尿病家族史等11个混杂因素后，多因素Cox回归分析结果显示，浙江省德清县以荤食为主的农村社区居民T2DM发病风险为饮食均衡居民的2.46倍（HR = 2.46，95 % CI = 1.94～3.13）。  结论  饮食偏好以荤食为主可能增加农村社区居民T2DM的发病风险，今后在T2DM的防治中应加以重视。

Association between dietary meat consumption and incident type 2 diabetes: the EPIC-InterAct study

  Objective  To explore the association between dietary behavior and the incidence of type 2 diabetes mellitus (T2DM) among rural adults in Deqing county, Zhejiang province and to provide evidences for implementing relevant interventions.  Methods  We conducted a dynamic prospective cohort study among a total of 28 233 diabetes-free permanent residents aged 18 – 64 years randomly selected from 2, 6, and 7 rural communities between 2006 – 2008, 2011 – 2012, and 2013 – 2014, respectively. Questionnaire survey, physical examination and laboratory test were carried out among the participants. Univariate and multivariate Cox regression were applied to explore the associations between diet preference and T2DM.  Results  By the end of study, a total of 2 773 participants, including 874 deaths, were lost to follow-up, with a rate of loss to follow-up of 9.82%. Totally 369 new T2DM incidents were identified till November 2015 and the incident density was 3.13 per 1 000 person years (PYs). Among the participants reporting main dietary pattern of balanced diet, vegetarian diet, and meat or fish diet, the number of T2DM incidents newly identified were 227, 46, and 96 and the T2DM incident density were 2.59/1 000 PYs, 2.88/1 000 PYs, and 6.67/1 000 PYs, respectively. The results of univariate Cox regression revealed that the T2DM risk was 2.58 times higher (hazard ratio [HR] = 2.58, 95% confidence interval [95% CI]: 2.03 – 3.27) among the participants with a main dietary pattern of meat or fish diet than among those with a main dietary pattern of balanced diet. The results of multivariate Cox regression demonstrated that compared to the participants with a main dietary pattern of balanced diet, those with a main dietary pattern of meat or fish diet had a significantly increased risk of T2DM incidence (HR = 2.46, 95% CI: 1.94 – 3.13) after adjusting 11 confounding factors including gender, age, education, occupation, smoking, alcohol drinking, tea drinking, body mass index, hypertension, impaired fasting glucose, and familial history of diabetes.  Conclusion  Meat or fish dietary pattern could increase the incidence risk of T2DM among rural adult residents and the results should be concerned in future prevention and control of T2DM among the population.