Objective: To explore the interaction of smoking and diabetes on stroke. Methods: In this case-control study, a face to face questionnaire survey was conducted. Logistic regression models were used to analyze the relationship between smoking or diabetes and stroke. The indicators of interaction were calculated according to the Bootstrap method in this study. Results: A total of 918 cases and 918 healthy controls, who participated in the chronic disease risk factor survey in Xuzhou in 2013, were included in this study. Logistic regression analysis found that cigarette smoking was associated with stroke (OR=1.63, 95%CI: 1.33-2.00), and diabetes was also associated with stroke (OR=2.75, 95%CI: 2.03-3.73) after adjusting confounders. Compared with those without diabetes and smoking habit, the odds ratio of stroke in those with diabetes and smoking habits was 8.94 (95%CI:3.77-21.19). Diabetes and smoking combined interaction index was 3.65 (95%CI: 1.68-7.94), the relative excess risk was 5.77 (95%CI: 0.49-11.04), the attributable proportion was 0.65 (95%CI: 0.42-0.87). Conclusion: The results suggest that there are additive interactions between smoking and diabetes on stroke.
目的： 探讨吸烟和糖尿病与脑卒中的关系，分析两者对脑卒中的交互作用。 方法： 采用病例对照研究方法，通过调查问卷收集脑卒中患者及对照者的研究信息。以2013年徐州市慢性病危险因素调查发现的918例脑卒中患者作为病例组，同期参加慢性病危险因素调查相同例数的健康者作为对照组。采用非条件logistic回归模型分析吸烟和糖尿病与脑卒中的关系，通过Bootstrap法计算相对超额危险度比（RERI）、归因比（AP）和交互作用指数（S）以评价吸烟和糖尿病对脑卒中的相加交互作用。 结果： 吸烟与脑卒中有关联（OR=1.63，95%CI：1.33~2.00）；糖尿病与脑卒中也有关联（OR=2.75，95%CI：2.03~3.73）；吸烟和糖尿病同时存在时，与脑卒中关联增大（OR=8.94，95%CI：3.77~21.19）。吸烟和糖尿病对脑卒中的相加交互作用：S=3.65（95%CI：1.68~7.94），RERI=5.77（95%CI：0.49~11.04），AP=0.65（95%CI：0.42~0.87）。 结论： 吸烟和糖尿病对脑卒中具有交互作用，两因素同时存在时，对脑卒中的危害大于单因素的作用。.
Keywords: Diabetes; Interaction; Smoking; Stroke.