High body mass index (BMI) predisposes to several chronic diseases, but a large-scale systematic and detailed study of dose-response relationship between BMI and chronic diseases has not been reported previously. In this study, we aimed to investigate the relationship between BMI and 3 chronic diseases (hypertension, dyslipidemia and MetS) in northeast China. A sample of 16412 participants aged 18~79 years old were included in Jilin province in 2012. The lambda-mu-sigma (LMS) method was applied to examine the trend of BMI by age, and the restricted cubic splines were used to investigate the non-linear associations (dose-response curve) between BMI and chronic diseases. It was pointed out that BMI increased rapidly when young, then kept steady in middle age, and finally declined slowly in old age, and accordingly age was divided into 3 segments, which were different by gender. The odds ratios (ORs) of BMI for the chronic diseases increased relatively slowly when young, then increased dramatically in middle-age and old population, especially for men. Further, the ORs of BMI among non-smokers were lower than those among smokers, and the same trend was shown to be more apparent among drinkers and non-drinkers. The risk of BMI for common chronic diseases increased dramatically in middle-aged, especially for men with drinking and smoking habits.