1. The data presented the baseline characteristics of the 5364 subjects (aged ≥ 25 years) in three BMI classes for men and women [see Additional file 1
Compared with the normal weight groups, the means of all variables were statistically higher in the overweight and obesity groups, except that the mean of HDL-C was lower for men and women; while those means were lower in the underweight groups, except for the means of HDL-C and age for both genders, and the means of LDL-C and FINS for women.
2. The data demonstrated the standardized prevalences of overweight/obesity [see Additional file 2
According to the Chinese standard and the WHO standard, (1) the standardized prevalences of overweight plus obesity were 41.5% and 31.2%, respectively; (2) among women, the age-specific prevalences increased until age 74 for overweight but until age 64 for obesity, and then declined thereafter; among men, the age-specific prevalences peaked at 65–74 years old for overweight; (3) the age-specific prevalences of overweight were higher in men than in women of 25–54 years old, but the opposite was true for those above 55 years old; for obesity, the age-specific prevalences were higher in women than in men in all groups except for 25–34 years old.
The risk of overweight and obesity (BMI ≥ 25 kg/m2), reflected by adjusted ORs, increased from the age groups (25–34 years in women, and 35–44 years in men) to the 65–74 years age groups (in both genders), and then decreased.
3. The data summarized the risk factors of overweight and obesity for men and women in logistic regression models [see Additional file 3
Among different categories of educational levels (for both genders), smoking behavior(for men), alcohol intake (for men) and household income (for women) the proportions of subjects were significantly different between the two BMI classes (p < 0.05). The proportions of the individuals with the family history of obesity were higher in the overweight and obesity group than in the normal weight group for both genders (p < 0.001).
The associations between the risk of overweight plus obesity and the following factors in multivariate models were as follows: (1) the risk increased by approximately 1.2-fold for both genders with family history of obesity. (2) Current male smokers had a lower risk of overweight and obesity (OR = 0.76, p < 0.05) than nonsmokers. In contrast, current male drinkers had a higher risk (OR = 1.42, p < 0.05) than nondrinkers. (3) Compared with low-educated women, medium- and high- educated women were at lower risk of overweight and obesity, and the corresponding ORs (95%CIs) were 0.64(0.52–0.79) and 0.50(0.36–0.68), respectively.
In addition, we also described some baseline characteristics of smoking and alcohol intake habits as follows.
Among male subjects, the proportions of overweight plus obesity were 30.2%, 34.5% and 36.1% in current smokers, ex-smokers and non-smokers, respectively. As to amount and duration of smoking in 1131 current smokers (1033 men and 98 women), the medians (interquartile range) were 15 cigarettes/day (10 cigarettes/day -20 cigarettes/day) and 20 years (15 years -30 years), respectively.
The frequency distribution of 631 current drinkers (580 men and 51 women) among different alcohol beverages was listed below: 1) Chinese distilled spirit (n = 123), 2) rice wine (n = 246), 3) beer (n = 152), 4) mixed Chinese distilled spirit and rice wine (n = 17), 5) mixed Chinese distilled spirit and beer (n = 21), 6) mixed rice wine and beer (n = 48), or 7) mixed Chinese distilled spirit, rice wine and beer (n = 24). For Chinese distilled spirit, rice wine and beer, the Medians (interquartile range) of consumption were 100 grams (50 g/day-150 g/day), 200 grams (100 g/day-250 g/day) and 1 bottle (1 bottle/day -1 bottle/day), respectively.
4. The data showed associations between metabolic diseases/disorders and BMI [see Additional file 4
The proportions of DM, hypertension, high serum triglyceride, high serum LDL and low serum HDL among the BMI classes elevated with increasing BMI, this trend also could be observed through the corresponding age-adjusted ORs. Although the proportions of high serum total cholesterol in the overweight/obesity classes were higher than those in the normal/overweight classes, there were not significant differences of the proportions between the overweight group and the obesity group, as well as between the normal weight group and the underweight group.
5. Table presented the changes of BMI after an average 3.6-year follow-up
(1) The means of baseline BMI were less than 25 kg/m2 in all age groups except for the 55-year and the 65-year age groups in women. After an average of 3.6 years, the means of BMI were still less than 25 kg/m2 in all groups except for the 55-year age group in women. (2) After 3.6 years, the increases of BMI were statistically significant in 35–44 years old, and the decreases (of BMI) were significant above 65 years for both genders. (3) In general, the annual changes of BMI means were insignificant, ranging from an increase of 0.1 kg/m2 to a decrease of 0.1 kg/m2 for men, and ranging from an increase of 0.1 kg/m2 to a decrease of 0.2 kg/m2 for women. (4) There were no significant changes in the percentages of overweight and obesity in the same-gender and same-age groups after 3.6 years.
Changes of BMI for 1264 men and 1768 women after an average 3.6-year follow-up
6. Figure depicted the percentages switching between two BMI groups after 3.6 years (by gender and age group)
About 85%–93% of the subjects among different age groups remained in their original BMI groups after 3.6 years. At the same time, for young and middle-aged individuals (men aged 25–54 years and women aged 35–54 years), there were higher percentages switching from the underweight and normal weight group to the overweight and obesity group, than the opposite. But the case reversed for the old individuals (men aged 55–64 years and women above 55 years). The percentages transferring between the two BMI groups reached a balance for men above 65 years old.
Figure 1 The percentages switching into another BMI groups (by age group) for men and women after an average 3.6-year follow-up. □ proportion switching from the underweight and normal weight group at baseline to the overweight and obesity group after an (more ...)