The Difference of Being Overweight or Obese between the North and South of China

Objectives: The geographical environment, dietary culture, food patterns, and obesity rates are substantially different between the North and South of China. Determining the geographical distribution and local dietary patterns involved in being overweight or obese is useful for designing intervention strategies. Methods: Residents between 18 and 65 years old (n=10,863) from 11 Chinese provinces (ve Northern provinces and six Southern provinces) were selected to compare dietary patterns, BMI, and health-related information from the China Health and Nutrition Survey packages in 2011. Linear and logistic regression analyses were performed to assess the strength of the association among geographic variables, the obesity problem, and dietary patterns. Results: The overall prevalence of being overweight or obese was 10.51% higher in the North than in the South. Northern dietary patterns feature a high intake of wheat and soybeans, whereas Southern dietary patterns feature a high intake of rice, vegetables, meat, and poultry. The estimated coecient of regional variables surrounding dietary score is 1.494; surrounding the odds ratio for being overweight is 1.681, whereas surrounding the odds ratio for obesity is 2.035. Multivariate logistic regression including both the variable of South–North areas and Northern dietary patterns showed a signicant correlation with being overweight or obese. Conclusion: Northern areas and their local dietary patterns are more likely to contribute to being overweight or obese. These ndings provide support for tracking the progression of obesity, epidemics, and policies that target the ‘‘obesogenic’’ environment, promoting opportunities for persons to access healthy dietary patterns and nutritional balance.


Introduction
Being overweight or obese is an increasingly serious public health problem affecting every country. This problem is not only associated with developed countries, but has also begun to emerge in the developing world [1], especially in densely populated China. From 1991 to 2011, the percentage of overweight adults rose from 20.5-42.3%, and obesity increased from 3.75-11.3% [2]. It is essential to tackle this issue, as being overweight or obese has long-term in uences on the development of diseases, including diabetes and hypertensive disease [3]. In China, it is not only the obesity phenomenon that is a growing problem; there is also the issue that Northerners have a higher prevalence of being overweight or obese than Southerners [4][5][6]. It is generally believed that the causes of obesity are the socio-cultural environment and lifestyle [7]. However, no data are available regarding a reduction in the prevalence of the problem from the perspective of dietary culture and regional differences, which is required information for drawing up purpose-designed policies for public health intervention.
In china, the Qinling-Huaihe Line is widely recognized to divide China geographically into two regions (the geographical boundaries between the North and the South of China are displayed in Fig. 1, with the edges of certain parts of the Chinese territory not fully shown on the map) in which there are major differences in terms of the geography and environment [8]. Because of the geographical environment, residents' have fostered a diverse and rich dietary culture throughout their long historical activity, and developed some differences of culture and diet between the North and South of China [9]. Dietary patterns represent a broader prospective on the interactions between nutrients and food intake, indicating that certain patterns are high predictors of risks [10,11]. Previous studies have demonstrated the relationship between dietary patterns and obesity [11,12]. However, to our knowledge, very few data are available concerning geographical environment and dietary culture in terms to dietary patterns and nutrient intake research in epidemic. In addition, no research has explored the impact of geographical differences and dietary culture on the associations between dietary patterns and being overweight or obese in China.
Determining the rates of being overweight or obese within regions and dietary patterns is very useful for tracking the progression of the epidemic and establishing the most effective places to target intervention strategies. The goal of the present study is to study the differences in rates of being overweight or obese based on the geographically in uenced dietary cultures in North and South China. It is further intended to explore the causes of these differences, identifying the factors associated with geography and dietary patterns. Finally, we hypothesize that geography and dietary pattern differences are the main reasons for being overweight or obese, and we hope to identify more appropriate dietary patterns and lifestyle options to address the growing prevalence of this problem.

Data
The data used in this study were taken from the Chinese Health and Nutrition Survey (CHNS) for 2011. The CHNS is an ongoing, open-cohort, international, collaborative project between the University of North Carolina and the Chinese Centre for Disease Control and Prevention, and is also the only longitudinal, household-health-based survey in China [13]. The purpose of the CHNS is to investigate the impact of the Chinese government's policies on family planning, health, nutrition, and implemented programs. It also provides information on health and nutritional status during the great Chinese economic transformation. These data mainly measure the impact of health behaviors, nutritional status, and outcomes, as well as changes in demographic, family, individual, economic, and social factors.
The present study used a multistage random cluster to sample 10,863 residents (North = 4,411; South = 6,452) aged 18 to 65 years old when other samples with incomplete information were excluded. The regional variables included eleven provinces and were selected for this paper ( Fig. 1): in the North, this included Heilongjiang, Liaoning, Beijing, Shandong, and Henan; in the South, this included Jiangsu, Shanghai, Hubei, Chongqing, Hunan, Guizhou, and Guangxi. A small section of Henan lies south of the Qinling-Huaihe Line, whereas Jiangsu and Hubei have minor regions located north of the Qinling-Huaihe Line; however, we still consider Henan a Northern province and Jiangsu and Hubei Southern provinces because the Qinling-Huaihe Line is not divided by the provincial boundaries, but by rivers and mountains.

Outcome variables
Measurements were taken of height (m), weight (kg), and BMI (weight/height 2 ) by trained health workers following standardized protocols from the WHO. The clinical classi cation criteria indicated that being overweight is de ned as a BMI ≥ 24 kg/m² and obesity is de ned as a BMI ≥ 28 kg/m² [14].
Dietary patterns ware presented by dietary score. Individual food consumption was recorded through a 24-hour recall method for three consecutive days for each participant. The eating times and weight of each type of food the participants consumed over the course of three days were recorded to determine the total average consumption of each category of food per day. The dietary scores ranking the reasonable intake levels of food patterns were obtained through the Dietary Balance Index (DBI) [15,16], which is a comprehensive statistical technique for assessing under-and over-nutrition in Chinese residents, thus providing a useful tool to monitor dietary healthiness [16]. The DBI method is similar to the Healthy Eating Index [17] and the Diet Quality Index [18].
The closer the dietary score is to zero, the more reasonable the dietary intake is. In contrast, the higher the dietary score is, the more likely the dietary intake equates to over-nutrition. Scoring details of the DBI method can be found in Table S1.

Independent variables
The independent variables in this study is dietary quality score and regional variables, dietary score is not only the dependent variable for former stage but also the independent variable for latter stage. The control variables were household income, education level, age, sedentary time, and some health-related information about parents collected via in-home interviews and general information questionnaires. The sample's education level was divided into ve levels (none, primary school, lower middle school, upper middle school, and university).
Sedentary time includes all the sedentary activities taking place at home, such as watching TV, playing games, sur ng the Internet, and reading. This study did not directly consider the variable of "average number of times per week of engaging in physical or outdoor activity" because individual exercise patterns were inconsistent. As a complement, the sedentary time variable (minutes per week) was used to control for the effect of physical activity on obesity analysis. In addition, individual registration, smoking and alcohol drinking are also considered in our present study. Participants were selected in the 18-to-65 age range to eliminate the effects of the changing dietary patterns in children or the elderly. In addition, we also considered how the family environment factors into children's obesity and nutritional balance, such as parental dietary patterns, obesity, hypertension, and diabetes [19].
In this study, regional variables were de ned in three different forms by the span of geographic space: regional type I, regional type II, and regional type III. Regional type I is divided into 11 areas according to the division of administrative provinces in China (the region sequentially moves from the southernmost Guangxi province to the northernmost Heilongjiang province). Regional type II is divided into three areas: North (Heilongjiang, Liaoning, Beijing, and Shandong), Central (Henan, Jiangsu, and Hubei) and South (Shanghai, Chongqing, Hunan, Guizhou, and Guangxi). Regional type III is divided into two areas: The South and the North. From regional type I to regional type III, the span of geographical space gradually increases.

Extraction of dietary patterns
Dietary patterns were extracted using principal component factor analysis based on 12 main categories of food [20]. Two dietary patterns were obtained from participants' food consumption records with eigenvalues (> 1), and the variance explained by each factor. The sign de ning these linear combinations, called factor loadings, determined the correlations of each food category with the factor. Food categories with loadings on a factor were used to describe the characteristics of dietary patterns; higher loadings mean the categories share more variance with the factor. Labelling of the factors is mainly descriptive and based on our explanation of dietary pattern structures, and participants were labelled according to dietary-pattern-speci c factor scores. Scores for each dietary pattern were counted as the sum of the factor loading coe cients and standardized daily consumption of each food category associated with the corresponding pattern.
In this study, we used two kinds of scores for diet: one was calculated via the DBI, and the other was obtained using factor analysis. The DBI method can comprehensively show the quality of food intake, which mainly scores the level of food intake according to the level of each food (found in Table S1), and nally obtain the dietary score through a summation. In contrast, the factor analysis method mainly indicates the characteristics of speci c dietary patterns, scoring points by the weighted sum of the combination coe cients of a certain pattern. Therefore, we have mainly employed dietary score via the DBI method to compare and study the food intake of residents between the North and South, but using the score calculated via factor analysis to present the characteristics of the two newly extracted Southern and Northern dietary patterns, in the nal regression analysis ( Table 6).

Statistical study
Related data were analyzed via Stata version 14 (Stata Corp, College Station, TX, USA) to determine the causes of the different rates of being overweight or obese between China's North and South. First, a series of descriptive analyses were conducted regarding the participants' BMI and demographic characteristics (e.g., education level, household income, age), rates of being overweight or obese, the degree of different categories of food intake, and parental health-related conditions (e.g., BMI, hypertension, diabetes) and food intake. The DBI method was also applied as a dietary score measure to assess the quality of dietary intake, in which a high score means more food intake and over-nutrition being more likely. Comparisons of the changes in prevalence across regional type III were made using t-tests for continuous variables and χ 2 tests for binary variables (allowing for unequal variance between two groups). Second, we assessed correlations between various categories of food intake and regional type III, and hoped to extract two kinds of dietary patterns using factor analysis. Furthermore, we carried out a series of regression analyses to assess how regions have in uenced dietary scores and being overweight or obese, adding certain control variables related to parental obesity-related information. Finally, on the basis of the existing relationships between regional variables and dietary scores, we further took being overweight or obese as a dependent variable to conduct multivariate logistic regression with the two dietary patterns extracted using factor analysis and the South-North regional dummy to verify that these relationships with different dietary patterns and geographical variations have a signi cant impact on rates of being overweight or obese.

Results
Signi cant increases in mean BMI were observed when moving from the more southern to the more northern areas ( Table 1). The BMI in the North was 23.732%, whereas in the South, it was 22.694% (p < 0.001). Household income, education level, and sedentary time also exhibited statistical signi cance (p < 0.05), which was lower than the signi cance level for BMI. In addition, both Southern parents' BMI levels were higher than those of parents living in the North (see Table 1; all p ≤ 0.001 for BMI), and a portion of the parents' obesity-related diseases also showed signi cance between the two regions. These differences in BMI for both participants and their parents demonstrated that there was a highly signi cant difference in rates of being overweight or obese between the South and North of China. Southern areas, the prevalence shows a roughly decreasing trend (Fig. 2). In addition, the average rate of being overweight including obese in the North was higher than that in the South by 10.51%. Figure 2 portrays how there are signi cant differences in the rates of being overweight including obese between the North and South of China.
As there are signi cant differences in the rates of being overweight including obese between the North and South, we further veri ed whether there is a difference in dietary patterns. Analyzing the statistical description of food intake and dietary scores, we observed that most categories of food feature signi cant differences between the two areas ( Fig. 3). Intake of rice, vegetables, poultry, meat, sh, and alcohol in the South was greater than that in the North, and statistical signi cance was reached for all categories except poultry and alcohol (p < 0.1). In contrast, intake of wheat, soybeans, fruits, tubers, eggs, and dairy in the North was higher than that in the South, and statistical signi cance was reached except for soybeans (p < 0.1). In addition, rice and wheat were the largest difference in food categories between the North and South; the average consumption of rice in the South was 165.9 g higher than that in the North, whereas the average consumption of wheat in the North was 158.5 g higher than that in the South. Furthermore, there were also signi cant differences in dietary scores between the North and South. The dietary scores of Northerners were higher by 1.1 points than those of Southerners on average, which means that the total quality of food intake in the North is more likely to re ect over-nutrition than in the South. In conclusion, there are signi cant differences in dietary patterns among residents between the North and South.
In addition, we found that most categories of food were highly correlated with regional variables from the South to the North (Table 2). Wheat, soybeans, fruits, tubers, eggs, and dairy had signi cantly positive correlations with regional variables, whereas rice, vegetables, poultry, meat, and sh had signi cantly negative correlations with regions, demonstrating that there were signi cant differences with respect to geographical variation in the dietary patterns of Chinese residents. In addition, the estimated coe cient of wheat and rice was the largest but opposite, which reveals that the difference between wheat and rice was the most signi cant between the two areas, and such results were consistent with Song's ndings [9].  (Table 3). Finally, we obtained factor scores of the two types of dietary patterns using the sum of the factor loading coe cients and standardized daily consumption of each food category, and we de ned them as new variables for Southern dietary patterns and Northern dietary patterns, which is generally consistent with the actual information surrounding the consumption of various types of food in the two regions [9]. All of these estimated coe cients for regional variables had a signi cant correlation (p < 0.01) with dietary score (Table 4), and this gradually increased from 0.287 to 1.494 with a rise in geographic span (from model 1 to model 3). In model 1, the geographic span gradually changed from South to North according to adjacent provinces for all 11 provinces. Furthermore, in model 2, the geographic span was rst changing from the Southern to Central areas, and then to the Northern areas, and the size of the geographic span was larger than regional type I. Finally, in model 3, the geographic span changed directly from South to North, adjusted for the three Central provinces (Henan, Jiangsu, and Hubei) into Northern and Southern areas mainly according to their geographical locations, and the geographic span of regional types III were the greatest than that of regional types I and II. In conclusion, the larger the geographic span, the greater the difference in dietary scores. Furthermore, the dietary score rose from the South to the North, indicating that Southerners' dietary patterns were less prone to over-nutrition than those of Northerners. After adjusting for income, age, education level, sedentary time, registration, smoking, drinking and parents' dietary scores, the estimated coe cient of the regions was still signi cant (p < 0.1). Moreover, education level and sedentary time were signi cantly negatively correlated with dietary scores, which was consistent with Woo's results [21]. The samples' dietary scores were also signi cantly correlated with those of their parents, which may be because there is a similar dietary difference in generations between the North and South [22].  Furthermore, we evaluated the causes of being overweight or obese regarding rate differences through regional factors. All of the odds ratios (OR) for regional variables exhibited a signi cant correlation (p < 0.01) with being overweight or obese (Table 5); the OR for being overweight gradually increased from 0.923 to 1.681, and the OR for obesity gradually rose from 0.902 to 2.035 with the increase in geographic span (from model 5 to model 7 and from model 9 to model 11). The variation trend of the OR coe cient was consistent, as per Table 4. After adjusting for income, age, education level, sedentary time, and parents' obesity-related information, the regional OR coe cients were still of signi cance. In conclusion, the possibility of becoming overweight or obese increases gradually from the Southern areas to the Northern areas, and the greater the geographic span, the faster these rates increase. Furthermore, the samples' rates of being overweight or obese were also signi cantly correlated with those of their parents' obesity-related history. Note: Models 5 to 7 and models 9 to 11 only considered the different sizes of the geographic span for being overweight and obesity, respectively. Model 8 and model 12 includes all of model 7 and model 11 respectively, and further adjusts for control variables household income, age, education level, sedentary time and parents' histories of obesity and chronic health-related information. It should be noted that a large number of samples are lost in the regression model of 8 and 12 (n = 1,515 residents (North = 565; South = 986)), because these models include parental factors, and the matching steps between the samples and their elders are di cult.
According to the aforementioned studies, there are signi cant differences in dietary patterns and rates of being overweight or obese for residents of the two regions. Southern rates of being overweight or obese and dietary scores were lower than those of the North. Therefore, we further veri ed the in uence of both dietary differences and geographical differences in terms of the problem of residents being overweight or suffering from obesity. In Table 6, we show the two new dietary patterns extracted using factor analysis, which can approximately represent the characteristics of Northern and Southern dietary patterns rather than arti cially dividing two local dietary patterns by geographical boundaries. overweight or suffering from obesity. These phenomena also suggest that geographical differences not only directly affect the prevalence of being overweight or obese, but also indirectly impact the prevalence of being overweight or obese through in uencing dietary patterns between the two areas.

Discussion
Being overweight or obese is one of the most important social and health problems, and its prevalence is increasing continuously worldwide, including in China [4,23,24]. Currently, it is generally believed that lifestyle and diet are the primary causes of being overweight or suffering from obesity [6,25], and that the future solution for such epidemiological issues should come from political policies based on prevention of those aspects related to health and social customs surrounding being overweight or obese. However, not enough is yet known about these factors, which furthermore may vary from place to place [7,26]. In this study, we systematically analyzed the differences in dietary patterns and the prevalence of being overweight or obese between the South and North of China. The prevalence of being overweight or obese in the South was 10.51% less than that of the North, with this prevalence falling when moving through regions from Northern areas to Southern areas. Moreover, the Southerners' dietary score regarding the quality of food consumed was also lower by 1.1% than that of Northerners. People in the North have a high intake of wheat, tubers, fruits, etc., whereas individuals in the South eat more rice, vegetables, meat, poultry, etc. From the South to the North, all of these estimated coe cients regarding dietary patterns and being overweight or obese were positively and signi cantly correlated with the geographical variable. In addition, with the spatial span increase de ned by three regional types from the South to the North (neighboring provinces, South-Central-North areas, and South-North areas), the estimated coe cients for dietary score increased from 0.287 to 1.494, the OR for being overweight rose from 0.923 to 1.681, and the OR for obesity increased from 0.902 to 2.035. Finally, after placing regional variables and the two dietary patterns extracted using factor analysis into the same model, the OR of the three variables for being overweight were statistically signi cant at 1.464, 1.223 and 1.242, respectively, but only the OR of Northern dietary patterns and regional type III for obesity were statistically signi cant at 1.697 and 1.199, respectively. In short, residents from the South were also less likely to be overweight or obese than those from the North before and after adjustment for individual socio-demographic characteristics and household factors [5].
Firstly, there are signi cant differences in dietary patterns between the North and South of China. One plausible contributor is regional differences in the geographical environment. China has a natural, geographical dividing line-the Qinling-Huaihe Line, which consists of the Qinling Mountains and Huaihe River-which leads to differences between the North and South in many respects [8]. This natural dividing line is also a dividing line in terms of climate. In the North, there is a temperate monsoon climate with little rainfall. In contrast, in the South, there is subtropical monsoon climate with plenty of rain [27]. Owing to the difference in climate between the North and South, the residents' farming methods are also different. Residents in the North mainly cultivate droughttolerant crops, such as wheat, whereas residents in the South primarily grow subtropical cash crops, including rice and tea [28]. As a result, over the course of history, residents have established a variety of dietary cultures with local geographic characteristics, and the types of local agricultural products available may be responsible for some of the differences in dietary patterns [29]. For example, wheat became a main staple food in the North, and rice a major staple in the South. The present study's ndings are consistent with the actual geographical and cultural characteristics of China as well as ndings of previous works [9]. Furthermore, we applied factor analysis to extract two dietary patterns that roughly represent the difference in dietary characteristics between the North and South; these extracted dietary patterns also show that Southern dietary patterns are rich in rice, vegetables, meat, and poultry, whereas Northern dietary patterns are rich in wheat, soybeans, etc. A number of studies have established that the "modern high-wheat" pattern is rich in fat and induces a high risk of obesity and metabolic syndromes, whereas "vegetable-rich" and "healthy patterns" have had no associations with metabolic syndromes [10,30]. Zhang's work showed that traditional Southern dietary patterns feature a high intake of vegetables, sh, and meat, or what constitutes a balanced dietary pattern that may help to reduce the risk of chronic diseases, such as abdominal obesity and hypertension [31]. In addition, climate conditions may contribute to the differences in nutritional intake. The North is colder than the South, and as such, Northerners need to take in more energy to resist the cold, so their total food consumption is more than that in the South [32]. Meanwhile, Southern residents usually eat coarse ber foods, such as corn and buckwheat, which can easily enhance a sense of satiety and control the amount of energy consumed [9,33]. As a result, the differences in dietary patterns may be contributed to the geographical differences in climate and culture.
Secondly, the odds of being overweight or obese change rapidly and with positive signi cance as the geographic variables moving from South to North. When the region variable sequentially moves from one province to another across all 11 provinces from the Southern areas to the Northern areas, the prevalence of being overweight or obese slowly rises, but it increases quite quickly when the region variable directly moves from the Southern to the Northern areas. These phenomena and positive estimated coe cients indicated that the prevalence of being overweight or obese is signi cantly correlation with regional variables. This may also be explained by differences in physical activity [6,34,35]. The difference between energy intake and output partial contributed by physical activity means that rates of being overweight or obese are higher in the North than in the South [5]. Another reasonable contributing factor may be dietary patterns. The present ndings indicate that Chinese residents established local characteristics of lifestyle and diet cultures throughout long-term historical activities because of the existence of a boundary line between the North and South, which further affects residents' dietary patterns.
Overall, differences in residents' rates of being overweight or obese may derive from characteristics of the longestablished culture or lifestyle as impacted by an individual's residence, features of the dietary culture, and the regions themselves; persons in the North have greater eating scores and are more overweight or obese than those in the South. Swinburn and Egger have outlined an ecological framework for understanding rates of being overweight or obese, and they distinguished speci c types of environments, including physical, economic, and socio-cultural environments [36,37]. Hawkins and Gri ths found that the United States's geography affects early childhood rates of being overweight through dependence after adjustment for individual socio-demographic characteristics [38]. Angela also demonstrated that pre-adolescent children in Germany are considered overweight with marked geographical differences in prevalence [39]. Regions with unique environmental, cultural, or policyrelated (e.g., infrastructure) traits may provide or support opportunities for physical activity and/or access to a healthy diet, which have different in uences on the prevalence of being overweight or obese.
Based on the geographical differences and dietary habits formed throughout its long history, there are certain dietary and cultural differences between Northern and Southern China. However, the differences may not only be seen in contemporary residents, but also their elders, and the present study determined that Northern elderly residents have higher rates of being overweight or obese and dietary scores than those of Southern elders. The rates of fathers and mothers being overweight or obese in the North are 19.66% and 18.81% higher than those in the South, respectively, and the dietary scores of those in the North are 1.72 and 0.88 points higher than those in the South, respectively. Furthermore, parents' rates of obesity had a positive and signi cant correlation with participants' rates of being overweight or obese, and parents' dietary scores were also signi cantly correlated with participants' dietary scores. This difference and this correlation indicate that geographical differences may have a signi cant impact on dietary patterns for each generation, and local dietary patterns and geographical differences further affect residents' nutritional balance [40,41]. Furthermore, residents' household income, age, education level, and sedentary time also exerted some in uence on the prevalence of being overweight or obese.
Persons with higher household income, age, and sedentary time were more likely to be obese, but those with higher education levels were less prone to obesity [42][43][44]. Currently, with the rapid increasingly economy and concomitant changes in lifestyle, there has been a markedly high intake of fast foods in China [45]. However, because most fast foods have high fat and energy levels, thus leading a greater risk of hypertension, glucose, and being overweight or obese [46]. It is necessary to be aware of the problems of rapidly changing dietary patterns and obesity-related diseases.
The present research has several strengths compared to other studies. We selected 11 representative provinces to demarcate the differences in dietary patterns and rates of being overweight or obese between the North and South of China, and for the rst time, we studied the causes of these differences in an empirical manner considering geographical differences and dietary patterns, speci cally taking into account traditional Chinese culture [9,47]. In addition, we comprehensively described speci c dietary patterns of residents from the North and South while comparing the differences in the dietary structure of residents by the method of DBI, which could serve as a robust predictor of being overweight or obese and lead to effective policy interventions. Furthermore, the study provides information regarding the relationships among the South-North geographical difference, dietary patterns, and the prevalence of being overweight or obese. Northerners in the study ate differently from Southerners, at least from a qualitative perspective, probably consuming more food and leading to more opportunities for obesity. Finally, based on the natural geographical dividing lines in China, we discussed the historical origin of the differences in residents' dietary patterns and the causes of the differences in rates of being overweight or obese, combining the characteristics of an individual's living environment and traditional culture.
Our results also suggest that a series of relevant policies should be put in place to support the improvement of dietary patterns and living habits, thereby removing barriers to healthy eating and nutritional balance.
Several limitations of this study did exist. We did not speci cally evaluate the factors of economic status, family environment, physical activity, education level, registration, smoking, drinking, or genetic factors, and simply accounted for them as control variables in this paper. Further work is necessary to consider differences in the surrounding environment between the North and South to determine how these differences may promote a healthier nutritional balance. In addition, because of the limited information on nutrition, we did not assess the differences in nutritional composition between Northerners and Southerners. Certainly, there are other factors that may cause differences in the problems of being overweight or obese.     The prevalence of being overweight including obesity, in 11 Chinese provinces. Note: The designations employed and the presentation of the material on this map do not imply the expression of any opinion whatsoever on the part of Research Square concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. This map has been provided by the authors.

Figure 2
The prevalence of being overweight including obesity, in 11 Chinese provinces. Note: The designations employed and the presentation of the material on this map do not imply the expression of any opinion whatsoever on the part of Research Square concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. This map has been provided by the authors.

Figure 2
The prevalence of being overweight including obesity, in 11 Chinese provinces. Note: The designations employed and the presentation of the material on this map do not imply the expression of any opinion whatsoever on the part of Research Square concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. This map has been provided by the authors.