BMI与糖尿病关联强度剂量-反应关系分析

发布时间：2018-05-24 23:31

  本文选题：限制性立方样条 + 糖尿病　； 参考：《中国卫生统计》2017年06期

【摘要】：目的使用限制性立方样条模型探讨BMI连续变化与糖尿病患病关联强度的剂量-反应关系。方法采用代表中国45岁及以上中老年人群的2011年中国健康与养老追踪的基线调查数据,共纳入受访者9708名。应用非条件logistic回归和限制性立方样条模型分析BMI与糖尿病患病关联及剂量-反应关系。结果 2011年我国45岁及以上中老年人糖尿病患病率(95%CI)为14.77%(14.06%~15.48%)。调整混杂因素(性别、年龄、文化程度、饮酒、午睡、腰围、总胆固醇和C反应蛋白)之后,超重(OR=1.21,95%CI:1.04~1.41)、肥胖(OR=1.64,95%CI:1.37~1.97)与糖尿病患病相关具有统计学意义。限制性立方样条分析结果显示,不论性别和年龄分层,BMI连续变化与糖尿病患病的关联强度都呈非线性剂量-反应关系(非线性检验,P0.001)。结论高BMI(超重/肥胖)是糖尿病患病的独立、剂量依赖性的危险因素,提示以关注BMI连续变化(而非传统BMI水平)为重点的糖尿病预防将变得更为迫切。
[Abstract]:Objective using a restrictive cubic spline model to explore the dose response relationship between the continuous changes of BMI and the intensity of diabetes mellitus. Methods the baseline survey data of Chinese health and pension tracking in 2011, representing 45 years and older in China, were used to include 9708 people. The non conditional logistic regression and restrictive cube were used. A spline model was used to analyze the relationship between BMI and diabetes and the dose response relationship. Results the prevalence rate of diabetes (95%CI) was 14.77% (14.06%~15.48%) in Chinese aged 45 and older in 2011. After adjusting for confounding factors (sex, age, education, drinking, siesta, waist, total cholesterol and C reactive protein), overweight (OR=1.21,95%CI:1.04~1.41), The correlation between obesity (OR=1.64,95%CI:1.37~1.97) and diabetes was statistically significant. Restrictive cubic spline analysis showed that the continuous changes in BMI and diabetes were both nonlinear dose response relationship (nonlinear test, P0.001), regardless of sex and age stratification. Conclusion high BMI (overweight / obesity) is a diabetes disease. The independent, dose-dependent risk factors suggest that diabetes prevention, which focuses on continuous BMI changes (rather than traditional BMI levels), will become more urgent.
【作者单位】： 中南大学湘雅公共卫生学院流行病与卫生统计学系;
【分类号】：O212;R587.1
，

本文编号：1931083