Cigarette smoking, alcohol use, and the risk of diabetes in Guangdong population

doi:10.3969/j.issn.1001-9448.2001.06.005

Abstract：Objective　To examine the association between smoking, alcohol consumption and the risk of diabetes and the impacts of smoking and alcohol on postprandial plasma glucose(PG), blood pressure, body mass index(BMI) and waist-to-hip ratio(WHR) in patients with diabetes and IGT.Methods　Data from 1998 Guangdong provincial diabetes interview survey, a population-based cross-sectional study including. 11 767 residents aged 20～74 years［male 5　462, female 6 305,mean age (43.8±13.6) years］, was used for analysis. Standard oral glucose tolerance test(75 g) and physical examination were conducted on all subjects. Questionnaire included questions about physical activity, family history, economic state, average daily consumption of alcohol and cigarette, etc. Multivariate logistic regression and GLM-General factorial analysis were used. Results　In Guangdong population, the main independent predictive factors of diabetes were age (OR=2.053, 1.568～2.687), family history (OR=3.210, 2.019～5.104), exercise (OR=1.644, 1.217～2.221), BMI(OR=2.538, 1.905～3.381), WHR(OR=2.255,1.690～3.008) and hypertension(OR=3.088, 2.246～4.245). Affer adjusted for age, physical activity, family history and economic state, the odds ratio of smoking＞25 cigarettes a day was 1.037(0.367～1.535), drinking＞7 times per week was 1.261(0.718～2.214); but there were significant interaction between smoking or drinking and age, family history or exercise condition, after controlling for confounders, the odds ratios of diabetes were 3.910(1.953～7.826) for smoking＞25 cigarettes a day and age＞50 years, 2.251(1.144～4.431) for smoking＞25 cigarettes a day and lower physical activity, 2.421(1.130～5.185) for drinking＞7 times per week and age＜50 years, 8.608(2.140～34.631) for drinking＞7 times per week with family history. Among the patients with diabetes or IGT, current smoking had higher WHR but lower BMI than never and ever smoker, current drinker only had lower BMI than ever drinker. Conclusion　Either heavy smoking or drinking may be a dependent, modifiable risk factor for diabetes in Guangdong population. It has significant interaction with other independent risk factor for diabetes in Guangdong population. It has significant interaction with other independent risk factors-age, family history and physical activity. Smoking may lead to abnormal body fat distribution in diabetes and IGT.