Association Between Acculturation and Body Weight

BMJ Open: first published as 10.1136/bmjopen-2017-018768 on 22 June 2018. Downloaded from PEER REVIEW HISTORY

BMJ Open publishes all reviews undertaken for accepted manuscripts. Reviewers are asked to complete a checklist review form (http://bmjopen.bmj.com/site/about/resources/checklist.pdf) and are provided with free text boxes to elaborate on their assessment. These free text comments are reproduced below.

ARTICLE DETAILS

TITLE (PROVISIONAL) Association between Acculturation and Body Weight Status among Migrant Children in Guangzhou, China: A Cross-Sectional Study AUTHORS Huang, Xiaoling; Chen, Wen; Lin, Yanwei; Zhang, Qi; Ling, Li

VERSION 1 – REVIEW

REVIEWER Zhihong Sa Beijing Normal University REVIEW RETURNED 11-Sep-2017

GENERAL COMMENTS General comments: A study of the association between socio-cultural integration and migrant children’s body weight status has important implications for health intervention of this disadvantaged group in China. In my point of view, the manuscript needs drastic revision before it can be accepted for publication. Its potential contribution is weakened by the vague problematization of the issue under study and some methodological flaws.

Specific comments:

Background – http://bmjopen.bmj.com/ 1) The focus of the paper is the association between acculturation and body weight status among migrant children in China. However, the authors did not provide enough background information to help readers better understand the rationale behind the study. For example, what’s the prevalence of childhood overweight/obesity in urban and rural China, respectively? What’s the prevalence of overweight/obesity among local children in Guangzhou (if data is

available)? What aspect of Cantonese or urban culture and related on October 1, 2021 by guest. Protected copyright. behavioral pathways that might lead to change in weight status among migrant children in Guangzhou?

2) A related issue is whether the concept and definition of acculturation is appropriate for the present study. While it’s possible that healthy Cantonese food is a protective factor for overweight/obesity, to what extent other aspects of urban culture (i.e. language, social interaction, custom) may be related to weight status among migrant children? What’s the difference between urban-to-urban migrant children and local children in Guangzhou in terms of their culture and exposure to risk factors for obesity? The clarification of these issues can provide a better justification for their conceptualization of acculturation.

3) It’d be helpful for us to understand the empirical analysis if more relevant literature on childhood obesity in China is included in the background secion. For example, a brief review of empirical evidence on major risk factors for overweight/obesity among migrant

BMJ Open: first published as 10.1136/bmjopen-2017-018768 on 22 June 2018. Downloaded from children in China would be useful.

Methodology – 1) The authors provided literature support for the definition of childhood overweight/obesity in China (p. 7, 2nd paragraph). It’d be helpful if further explanation is provided on method used to construct childhood overweight/obsity in this study, for example, whether age- and gender-specific BMI measures were used to define overweight and obesity.

2) More careful validation of acculturation is needed if the paper focuses on it.

3) Is children’s survey self-administered or based on face-to-face interviews?

4) Level of physical exercise and place of origin are important confounding factors, but they are not included in the analysis.

Results 1) In bivariate analyses, neither acculturation index nor individual factors of acculturation is significantly associated with overweight/obesity (Tables 3 and 4). Acculturation is found to be significantly related to overweight/obesity in model 2 of Table 4. I wonder whether the significant association exists if important correlates of overweight/obesity such as physical exercise is included.

2) Is there interaction effect between acculturation and primary caregiver’s type of migration in their relationship to overweight/obesity?

3) It seems that gender plays an important and interesting role, but

the effect of gender is not thoroughly explored. http://bmjopen.bmj.com/

REVIEWER Myoungock Jang USA REVIEW RETURNED 04-Oct-2017

GENERAL COMMENTS This study discussed very interesting topic regarding domestic migration as related to overweight and obesity in China. Since this journal target international audience, please include more specific on October 1, 2021 by guest. Protected copyright. cultural and environmental characteristics for promoting the understanding of the study outcomes. Here are several specific comments.

Background

More specific economic situation in china would be helpful for audience who is not familiar with China. The authors touched based on the migration phenomena, but readers may want to know what is caused, how actively it is occurred, and what resulted in.

Need more information about obesity in China. If there were more

BMJ Open: first published as 10.1136/bmjopen-2017-018768 on 22 June 2018. Downloaded from statistical information regarding rate of obesity by different geographical areas (i.e., rural vs. urban) etc., it would be much helpful and specific characteristics about Guangzhou may need in the background section. Without any information, the readers may be surprised with the study results.

Authors developed an instrument. Need more information. Is this paper for testing the instrument? Or has it been already developed and tested?

Methods

Comparison of demographic difference between public school and private school children would be helpful.

The reason for using social anxiety and child diet intake as a confounding factors is needed.

For food intake variable, food intake and food preference are different. You may be consistent with terminology by saying food intake.

What is the point of collecting animal food vs. milk, beans and http://bmjopen.bmj.com/ products? One is supposed to be high fat and another is high protein? Overall, you may consider including food intake for your data analysis to see relationship with acculturation and weight status.

on October 1, 2021 by guest. Protected copyright. Results

The sample size difference is too small to compare group difference by weight status. Questionable to see the difference with this much small variation within sample. There is no information about range of BMI. Rather you may consider linear regression modeling to see the association between the score of acculturation and BMI. You may calculate BMI Z-score.

Be consistent with presentation of numbers and percentage

Need to describe why the age group by younger than 14 vs. older

BMJ Open: first published as 10.1136/bmjopen-2017-018768 on 22 June 2018. Downloaded from than 14

I would recommend not controlling food intake for the model. Readers may want to see difference of food intake between groups.

Discussion

Need a definition of international migrant. Is it meaning by country to country?

Need to discuss cultural perspectives of acculturation in china.

When you discuss the diet as a factor for the negative relationship of acculturation and overweight/obesity, readers may want to know their diet behaviors. How did you know they are engaged in healthy diet behaviors? You need to include diet intake variables in your model to explain the relationship.

If the environment in Guangzhou is healthier, it should be mentioned in the background to allow readers have a sense of the environmental effect on weight status.

The first paragraph on page 13 has two main points. If you want to http://bmjopen.bmj.com/ focus on parental influence, please focus on that. The age difference is a different point so that you may have a separate paragraph to discuss this.

VERSION 1 – AUTHOR RESPONSE on October 1, 2021 by guest. Protected copyright.

Reviewer #1s' Comments

Background 1) The focus of the paper is the association between acculturation and body weight status among migrant children in China. However, the authors did not provide enough background information to help readers better understand the rationale behind the study. For example, what’s the prevalence of childhood overweight/obesity in urban and rural China, respectively? What’s the prevalence of overweight/obesity among local children in Guangzhou (if data is available)? What aspect of Cantonese or urban culture and related behavioral pathways that might lead to change in weight status among migrant children in Guangzhou?

Author’s reply: We thank the reviewer for pointing out the need for more background information. We have provided the information about the prevalence of childhood overweight/obesity based on available data in China, including Guangzhou, and other countries. In general, the body weight status

BMJ Open: first published as 10.1136/bmjopen-2017-018768 on 22 June 2018. Downloaded from among children varies across regions in China. Urban areas have higher prevalence of childhood overweight/obesity than rural areas, and southern China has lower prevalence than northern China. (p. 4, 3rd paragraph)

In terms of the association between Cantonese culture and weight status among migrant children in Guangzhou, it can be summarized as follows. First, Guangzhou is the capital city and a metropolitan in Guangdong Province. Influenced by its time-honor history and migration features, Guangzhou has very unique and diverse culture, including diet, language, and custom. For instance, comparing with cities in other provinces, Guangzhou is famous for its Cantonese diet, however, comparing with cities within Guangdong, the diet in Guangzhou is characterized by manifold styles from throughout China and other countries. Second, among different aspects of Cantonese culture, diet and language may play important roles to lead changes in weight status among migrant children in Guangzhou via affecting life habits and social interaction. We have revised the background correspondingly (p. 5, 2nd paragraph).

Furthermore, the body weight status bears regional differences in China, e.g., residents in Guangzhou have lower body weight than their counterparts in other regions, which may be the result of local geographical environment and life habits that have formed throughout its long history.[1-3] The culture in Guangzhou, including diet, language, and custom, is influenced by migration and becomes more and more diverse. For instance, the traditional Cantonese diet is relative healthy compare with the diet in most regions of China, [4] but the diet in Canton is also influenced by diet from other regions and becomes divers. Besides, local residents speak Cantonese, while other dialects was also used for communication within migrants. Language played a role to affect the communication between migrants and local residents, which may have a further impact on social interaction and enable to engage in activities with local resident, including local diet, life habits, custom, and attention to fashion and health. [5] The interaction between local culture and original culture might lead to change in weight status among migrant children in Guangzhou. We have revised correspondingly (p. 5, 2nd paragraph).

2) A related issue is whether the concept and definition of acculturation is appropriate for the http://bmjopen.bmj.com/ present study. While it’s possible that healthy Cantonese food is a protective factor for overweight/obesity, to what extent other aspects of urban culture (i.e. language, social interaction, custom) may be related to weight status among migrant children? What’s the difference between urban-to-urban migrant children and local children in Guangzhou in terms of their culture and exposure to risk factors for obesity? The clarification of these issues can provide a better justification for their conceptualization of acculturation.

on October 1, 2021 by guest. Protected copyright. Author’s reply: We thank the reviewer to point out this critical comment. First, as we explained in the “background” section, acculturation is a multidimensional concept, and reflects the overall adaptation to a new society, including language, social interaction, preferences in life (e.g. diet, dressing, custom, et al). [6, 7] Diet is only a factor of preferences in life. The aim of this study is to explore the association among different dimensions of acculturation and body weight status among migrant children in China. We have concretized the concept of acculturation (p. 6, 2nd paragraph, line 94-98) and added the discussion of acculturation (p. 14, 2nd paragraph).

Second, as we replied in the first comment, although the traditional Cantonese diet is relative healthy compare with the diet in most regions of China, the diet in Guangzhou is diverse and includes both traditional Cantonese diet and other styles. Therefore, whether the diet in Guangzhou is a protective factor for overweight/obesity or not mainly depends on migrant children’s eating habits. Furthermore, other perspectives of culture have both direct and indirect influences on weight status among migrant children. For instance, language played a role to affect the communication between migrants and

BMJ Open: first published as 10.1136/bmjopen-2017-018768 on 22 June 2018. Downloaded from local residents, which may have a further impact on social interaction and enable to engage in local life habits, including diet and custom. 5 (p. 5, 2nd paragraph).

Additionally, China is a country of great cultural diversity, and the differences not only exist between urban and rural areas, bu t also between regions, e.g., southern, northern, western, and eastern China have very different cultures. Therefore, migrant children with urban-urban migration background also face with the acculturation process in Guangzhou. (p. 16, 2nd paragraph, line 315-317).

3) It’d be helpful for us to understand the empirical analysis if more relevant literature on childhood obesity in China is included in the background section. For example, a brief review of empirical evidence on major risk factors for overweight/obesity among migrant children in China would be useful.

Author’s reply: As our replies in the first comment, the research on migrant children’s body weight status in China was scanty. Therefore, we added a brief review on major risk factors for overweight/obesity among migrant children across the world. These factors are dietary habits, lifestyle behaviors, socioeconomic factor and it was mediated by caloric intake (especially calorie‐laden food), energy expenditure (especially physical activity) and lifestyle changes. [8-11] We have revised the background correspondingly (p. 5, 3rd paragraph, line 89-93) .

Methodology 4) The authors provided literature support for the definition of childhood overweight/obesity in China (p. 7, 2nd paragraph). It’d be helpful if further explanation is provided on method used to construct childhood overweight/obesity in this study, for example, whether age- and gender-specific BMI measures were used to define overweight and obesity.

Author’s reply: Migrant children’s overweight and obesity were defined based on the norm established by the working group on obesity in China (WGOC). [12] This WGOC norm uses the 85th percentile and the 95th percentile of the BMI within each age- and gender-specific groups of Chinese children as http://bmjopen.bmj.com/ cutoff points to define overweight and obesity, respectively (Table 1). We have added the definition of childhood overweight/obesity in the manuscript (p. 9, 1st paragraph, line 157-160), and have provided the following table as an appendix.

Table 1 Body mass index reference norm for screening overweight and obesity in Chinese children and adolescents

age male female on October 1, 2021 by guest. Protected copyright. (years) overweight obesity overweight obesity 7~ 17.4 19.2 17.2 18.9 8~ 18.1 20.3 18.1 19.9 9~ 18.9 21.4 19.0 21.0 10~ 19.6 22.5 20.0 22.1 11~ 20.3 23.6 21.1 23.3 12~ 21.0 24.7 21.9 24.5 13~ 21.9 25.7 22.6 25.6 14~ 22.6 26.4 23.0 26.3 15~ 23.1 26.9 23.4 26.9 16~ 23.5 27.4 23.7 27.4 17~ 23.8 27.8 23.8 27.7 18 24.0 28.0 24.0 28.0

BMJ Open: first published as 10.1136/bmjopen-2017-018768 on 22 June 2018. Downloaded from 5) More careful validation of acculturation is needed if the paper focuses on it.

Author’s reply: We agreed with the reviewer and added more details about the validation of acculturation. As mentioned in the manuscript, we have conducted the reliability test of the acculturation scale, and the Cronbach’s alpha of the total acculturation in this study was 0.70 (p. 9, 3rd paragraph, line 174-176).

Besides, we performed factor analyses to measure the acculturation. The factor analyses generated five factors from 12 original items, namely language, social interaction, custom, dressing; and diet. These factors were consistent with the dimensional contents of this acculturation scale, with good discriminant validity (Table 2). We have revised the description in the manuscript (p. 9, 3rd paragraph, line 176-178) and have also attached the following table in the appendix.

Table 2 Factor analysis of the acculturation scale among migrant children in Guangzhou, China in 2016* Factor Numbers of items % of Variance Language 4 21.41 Social-interaction 2 11.70 Custom 2 10.93 Dressing 2 10.99 Diet 2 11.42 Total 12 66.45 * Kaiser-Meyer-Olkin Measure of Sampling Adequacy: 0.72; Bartlett’s Test: P<0.001. Extraction method: Principal Axis Factoring; Rotation method: Varimax.

6) Is children’s survey self-administered or based on face-to-face interviews?

Author’s reply: Our survey on children was based on self-administered questionnaires, and we have revised the manuscript (p. 8, 2nd paragraph, line 139). http://bmjopen.bmj.com/

7) Level of physical exercise and place of origin are important confounding factors, but they are not included in the analysis.

Author’s reply: Thanks the reviewer for indicating the importance of confounding factors.

a) Level of physical exercise: We agreed that the level of physical exercise is an important on October 1, 2021 by guest. Protected copyright. confounding factor in this study. However, we only investigated children’s physical exercise level at school, and the students attended three physical exercise session per week. Therefore, this exercise factor was not analyzed in this study due to a very limited level of variation in the level of physical exercise were found. Moreover, we also noted and discussed the limitation of the study as lacking of comprehensive measurements on physical exercise. After school exercise needs to be further explored in the future study (p. 16, 3rd paragraph, line 324-326).

b) Place of origin: We investigated migrant children’s place of origin in this study. However, due to limited sample size, we categorized their place of origin into two variables, namely caregivers’ type of migration (urban-urban or rural-urban) and province of origin (southern China province vs. northern China province). But the province of origin was not statistically significant (Table 3), so we simplified the model and only included caregivers’ type of migration in the analysis.

BMJ Open: first published as 10.1136/bmjopen-2017-018768 on 22 June 2018. Downloaded from Table 3 Logistic model three on the association between acculturation and overweight/ obesity among migrant children in Guangzhou, China 2016 * Model 3 Variable aOR (95%CI) P Caregivers’ type of migration Rural-to-urban migration (reference) 1 Urban-to-urban migration 2.060(1.155, 3.675) 0.014 Provinces of origin Southern China province 1 Northern China province 0.981(0.447, 2.155) 0.962 Acculturation 0.981(0.964, 0.998) 0.028 * Model 3 was based on the original model 2 and included the interaction effect between acculturation and primary caregiver’s type of migration.

Results 8) In bivariate analyses, neither acculturation index nor individual factors of acculturation is significantly associated with overweight/obesity (Tables 3 and 4). Acculturation is found to be significantly related to overweight/obesity in model 2 of Table 4. I wonder whether the significant association exists if important correlates of overweight/obesity such as physical exercise is included.

Author’s reply: As answered in the Comment #7, the level of physical exercise is one limitation of this study and needed to be considered in the future. In addition, we have added discussion about how acculturation affect physical exercise, which may lessen likelihood of overweight/obesity among migrant children (p. 15, 1st paragraph, line 293-296).

9) Is there interaction effect between acculturation and primary caregiver’s type of migration in their relationship to overweight/obesity?

Author’s reply: We agree with the reviewer that the interaction effect between acculturation and http://bmjopen.bmj.com/ primary caregiver’s type of migration should be considered. Actually, we have analyzed the interaction effect, however, it was not significant (Table 4). Therefore, we simplified the regression model and did not report this figure. In addition, we revised the “Statistical analysis” section, and added the corresponding analytical strategy (p. 11, 2nd paragraph, line 220-222).

Table 4 Logistic model four on the association between acculturation and overweight/ obesity among

migrant children in Guangzhou, China 2016 * on October 1, 2021 by guest. Protected copyright. Model 4 Variable aOR (95%CI) P acculturation * primary caregiver’s type of migration 1.007(0.971,1.045) 0.696 * Model 4 was based on the original model 2 and included the interaction effect between acculturation and primary caregiver’s type of migration.

10) It seems that gender plays an important and interesting role, but the effect of gender is not thoroughly explored.

Author’s reply: We also have analyzed the interaction effect between acculturation and gender, but it was not statistically significant (Table 5). Although we did not report the interaction effect, we have added a separate paragraph in the “discussion” section to further discuss the difference of gender (p. 15, 2nd paragraph, line 304-309).

BMJ Open: first published as 10.1136/bmjopen-2017-018768 on 22 June 2018. Downloaded from Table 5 Logistic model five on the association between acculturation and overweight/obesity among migrant children in Guangzhou, China 2016 * Model 5 Variable aOR (95%CI) P acculturation * gender 0.994(0.941,1.039) 0.746 * Model 5 was based on the original model 2 and included the interaction effect between acculturation and gender.

Reviewer #2s' Comments:

Background 1) More specific economic situation in china would be helpful for audience who is not familiar with China. The authors touched based on the migration phenomena, but readers may want to know what is caused, how actively it is occurred, and what resulted in.

Author’s reply: Thank for advice on enriching background of migration in China. Internal migration is the results of the urbanization and economic reform in China since the late 1950s. Under the permanent household registration (hukou) system in China, resident status is categorized as rural and urban residency. Due to the unbalanced economic development across regions in China, residents move across the country for a better life and form the 247 million internal migrant population (2015), and the majority of them are rural-to-urban migrants. Under the hukou system, the migration process creates social, economic, and cultural challenges and barriers to this internal migrant population. For instance, migrants have limited access to the education, housing, public health service, and social welfare, which are provided for local residents by local governments. Internal migrant children are generated by tailing with their caregivers who migrated from developing areas to relative developed areas within China. They also experienced significant socio-economic and cultural changes and challenges. We have added the related description into the background section (p. 4, 1st paragraph & p. 4, 2nd paragraph).

http://bmjopen.bmj.com/ 2) Need more information about obesity in China. If there were more statistical information regarding rate of obesity by different geographical areas (i.e., rural vs. urban) etc., it would be much helpful and specific characteristics about Guangzhou may need in the background section. Without any information, the readers may be surprised with the study results.

Author’s reply: Our responses to the first comment of the reviewer #1 also apply to this comment. We

has been revised correspondingly (p. 4, 3rd paragraph & p. 5, 2nd paragraph). on October 1, 2021 by guest. Protected copyright.

3) Authors developed an instrument. Need more information. Is this paper for testing the instrument? Or has it been already developed and tested?

Author’s reply: The measurement of acculturation in our study was adapted from the existing acculturation scales and of Unger JB, Nieri T, Zhang W, Barona A and Barry J et al. [13-16, 17] We have conducted reliability test and factor analysis to validate the acculturation scale (see our responses to the 5th comment of the reviewer #1).

Methods 4) Comparison of demographic difference between public school and private school children would be helpful.

Author’s reply: The reviewer’s suggestion would be helpful to understand the difference between public school and private school children. However, we only investigated migrant children from private

BMJ Open: first published as 10.1136/bmjopen-2017-018768 on 22 June 2018. Downloaded from schools. Under the hukou system in China, migrant children without hukou are widely excluded from accessing public schools in receiving cities. Therefore, we did not conduct the survey at public schools.

5) The reason for using social anxiety and child diet intake as a confounding factors is needed.

Author’s reply: We thank the reviewer for this suggestion. In our study, all confounding factors were selected based on the existing references which assessed the relationship between acculturation and body weight among children. We have revised the manuscript correspondingly (p. 10, 3rd paragraph, line 191-192 for mental health and p. 11, 1st paragraph, line 200-201 for food intake).

6) For food intake variable, food intake and food preference are different. You may be consistent with terminology by saying food intake.

Author’s reply: We thank the reviewer for pointing out the terminological standardization. In this study, the food intake variable was measured by counting their consumption frequency on 30 foods, therefore, we consistently used “food intake” in the manuscript, and revised the inconsistent expression (p. 11, 1st paragraph, line 203).

7) What is the point of collecting animal food vs. milk, beans and products? One is supposed to be high fat and another is high protein? Overall, you may consider including food intake for your data analysis to see relationship with acculturation and weight status.

Author’s reply: The food intake variable was measured by counting the study children’s consumption frequency on 30 foods.[18] The category of the food intake was based on the dietary nutritional perspective to evaluate the diversity of food intake, including carbohydrate; animal food; milk, beans and products; vegetables and fruits. Each type of food provides different nutrient for the body. Animal food, and milk, beans and products both mainly provide protein, fat, minerals, and vitamins. However, the former offers more fat, and the latter one is needed for high protein. While carbohydrate based food mainly provides carbohydrate, vegetables and fruits mainly offer dietary fiber and vitamin C. We http://bmjopen.bmj.com/ included four food categories into data analysis (Table 4 in the manuscript), rather than using the total score of food intake, and we have revised the description in the manuscript (p. 11, 1st paragraph, line 204 & p. 11, 2nd paragraph, line 220).

Results 8) The sample size difference is too small to compare group difference by weight status.

Questionable to see the difference with this much small variation within sample. There is no on October 1, 2021 by guest. Protected copyright. information about range of BMI. Rather you may consider linear regression modeling to see the association between the score of acculturation and BMI. You may calculate BMI Z-score.

Author’s reply: Thanks for reviewing carefully and giving us useful advices.

First, we have revised Table 2 as to compare the prevalence of overweight/obesity between different subgroups of each categorical variable (p. 12, 3rd paragraph, line 237-239 & Table 2 in the manuscript). Moreover, we added the information of BMI (p. 12, 3rd paragraph, line 230 & Table 2 in the manuscript).

Second, we have calculated BMI-for-age Z-score (WHO), and built the linear regression modeling to see the association between the score of acculturation and BMI Z-score (Table 6). The result was consist with the result of logistic regression analysis we reported in the manuscript. However, our data did not meet the linear regression preconditions and had a small R square, while the original logistic

BMJ Open: first published as 10.1136/bmjopen-2017-018768 on 22 June 2018. Downloaded from model 2 fitted well (Table 7) with the accuracy of model prediction of 91%. Therefore, we still chose the logistic regression to analyze the association between acculturation and overweight/obesity.

Table 6 Linear model on the association between acculturation and BMI-for-age Z-score among migrant children in Guangzhou, China 2016 * Model 6 Variable B (95%CI) P Acculturation -0.004 (-0.008,1.000) 0.045 * BMI-for-age Z-score was included in the model as the outcome. Model 5 was a multivariate linear model adjusted for gender, age, years of residence in Guangzhou, caregivers’ education, caregivers’ migration type, household monthly income per person, number of children in family, fear of negative evaluation (FNE), social avoidance and distress (SAD), and four categories of food intake. Model ANOVA test: F=1.753, P=0.033; adjusted R2=0.014; Shapiro-Wilk normality test: W=0.933, P<0.001; Durbin-Watson(U)=1.919.

Table 7 Hosmer and Lemeshow test of logistic model 2 on the association between acculturation and overweight/obesity among migrant children* χ2 df P Model 2 4.467 8 0.813 * The model 2 was fitted well (P>0.05).

9) Be consistent with presentation of numbers and percentage

Author’s reply: We have revised the Table 2 and corresponding descriptions in the results section (p. 12, 2nd paragraph, line 236-238 & Table 2 in the manuscript).

10) Need to describe why the age group by younger than 14 vs. older than 14

Author’s reply: The classification of age was mainly based on the mean of age (Mean=13.8, SD=0.9).

Besides, 14 years old is also used as the cut-off point for children and adolescents in China.[19] We http://bmjopen.bmj.com/ have added the description of age classification (p. 10, 1st paragraph, line 185-187).

11) I would recommend not controlling food intake for the model. Readers may want to see difference of food intake between groups.

Author’s reply: Thanks for the reviewer’s recommendation. We have showed the difference of four

categories of food intake between overweight/obesity and normal weight migrant children, which was on October 1, 2021 by guest. Protected copyright. presented in Table 1 in the original manuscript (Carbohydrate based food: P=0.338; Animal food: P=0.798; Milk, beans and products: P=0.749; Vegetables and fruits: P=0.371). As mentioned above (Reviewer #2, Comment #5), food intake had both relationship with acculturation and body weight based on the existing reference. [20] Therefore, food intake was included as a confounding factor. Furthermore, this factor was not significant in the model 2, and we did not further discuss the difference.

Discussion 12) Need a definition of international migrant. Is it meaning by country to country?

Author’s reply: The international migrant in our study was defined as those leave the mother country for another that may or may not have an intent to return. [21] We have added the definition in the first paragraph of the discussion (p. 13, 3rd paragraph, line 264-265).

13) Need to discuss cultural perspectives of acculturation in china.

BMJ Open: first published as 10.1136/bmjopen-2017-018768 on 22 June 2018. Downloaded from

Author’s reply: We have agreed to the reviewer and further to discuss about cultural perspectives of acculturation, especially in the aspects of language and custom in China. A separate paragraph was added in the “discussion” section (p. 14, 2nd paragraph).

14) When you discuss the diet as a factor for the negative relationship of acculturation and overweight/obesity, readers may want to know their diet behaviors. How did you know they are engaged in healthy diet behaviors? You need to include diet intake variables in your model to explain the relationship.

Author’s reply: In our study, acculturation in diet was reflected from two items (What kind of food do you usually eat? 1=all dishes like hometown; 2=most of dishes like hometown; 3=half like hometown and half like Guangzhou; 4=most of dishes like Guangzhou; 5= all dishes like Guangzhou. & What kind of dishes do you like? 1=hometown dishes; 2=local dishes; 3=foreigner dishes.), and high level of acculturation in diet reflected the high consistency with local diet. As discussed above (Reviewer #1, Comment #1), children in Guangzhou had the lower body weight than children in other regions, which may be explained by local diet. Therefore, we assume that migrant children who had high consistency with local diet were more likely to have healthy eating habits. Besides, we also took diet diversity into consideration and included food intake (including consuming frequency on four food categories) in the regression model (Table 4 in the manuscript).

15) If the environment in Guangzhou is healthier, it should be mentioned in the background to allow readers have a sense of the environmental effect on weight status.

Author’s reply: We thank the reviewer’s advice. Our responses to the first comment of the reviewer #1 also apply to this comment. In addition, we have provided the importance of environmental effect on weight status (p. 5, 3rd paragraph, line 92-96) and added the brief description the characteristic of Guangzhou environment and its effect on weight status in the “background” section (p. 5, 2nd paragraph, line 84-89).

http://bmjopen.bmj.com/ 16) The first paragraph on page 13 has two main points. If you want to focus on parental influence, please focus on that. The age difference is a different point so that you may have a separate paragraph to discuss this.

Author’s reply: We agreed with the reviewer and added a separate paragraph to discuss the effect of children’s characteristics (p. 15, 2nd paragraph). Another separate paragraph was represented to

focus on major caregivers’ influence (p. 15, 3rd paragraph). on October 1, 2021 by guest. Protected copyright.

Thanks for your reconsideration.

Sincerely Yours,

Xiaoling Huang

Contact details: Address: 74, Zhongshan Road 2, Faculty of Medical Statistics and Epidemiology, School of Public Health, Sun Yat-sen University, Guangzhou, China E-mail: Xiaoling Huang huangxlxpxc1163.com Corresponding author: Li Ling [email protected] Tel/Fax: 13530115220; 86-020-87333319.

Reference:

BMJ Open: first published as 10.1136/bmjopen-2017-018768 on 22 June 2018. Downloaded from 1. Song F, Cho MS. Geography of Food Consumption Patterns between South and North China. Food 2017;6(5):34. 2. Mai J. A Study on the Relations between obesity and dietary behaviors, physical activities among adolescents: Taking some students in Guangzhou China as an example Sichuan Normal University, 2014. 3. Zhao G, He W. Investigation on Overweight and Obesity of Urban Children and Teenage in Guangzhou. China Sport Science and Technoligy 2008;44(5):103-06. 4. Xue W. Cantonese cuisine: the perfect unity of nutritious food and cultural custom. Science and Technology of West China 2005:17-22. 5. Renzaho AMN, Oldroyd J, Burns C, et al. Over and undernutrition in the children of Australian immigrants: Assessing the influence of birthplace of primary carer and English language use at home on the nutritional status of 4-5-year-olds. International Journal of Pediatric Obesity 2009;4(2):73-80. 6. Barry DT, Garner DM. Eating concerns in East Asian immigrants: Relationships between acculturation, self-construal, ethnic identity, gender, psychological functioning and eating concerns. EatWeight Disord 2000;6:90-98. 7. Unger JB, Ritt-Olson A, KarlaWagner, et al. A Comparison of Acculturation Measures Among Hispanic/Latino Adolescents. J Youth Adolescence 2007;36:555–65. 8. Liu JH, Chu YH, Frongillo EA, et al. Generation and acculturation status are associated with dietary intake and body weight in Mexican American adolescents. The Journal of nutrition 2012;142:298-305. 9. Liu WJ, Liu W, Lin R, et al. Socioeconomic determinants of childhood obesity among primary school children in Guangzhou, China. BMC public health 2016;16:482. 10. Ye JC, Jiao CT, (WGOC) WGoOiC. Empirical Changes in the Prevalence of Overweight and Obesity among Chinese Students from 1985 to 2010 and Corresponding Preventive Strategies. Biomed Environ Sci 2013;26(1):1-12. 11. Nishtar S, Gluckman P, Armstrong T. Ending childhood obesity: a time for action. The Lancet 2016;387(10021):825-27. 12. Working Group on Obesity in China, Ji C. Body mass index reference norm for screening overweight and obesity in Chinese children and adolescents. Chinese Journal of Epidemiology 2004;25:10-15.

13. Unger JB, Gallaher P, Shakib S, et al. The AHIMSA Acculturation Scale: A new measure of http://bmjopen.bmj.com/ acculturation for adolescents in a multicultural society. Journal of Early Adolescence 2001;22:225-51. 14. Nieri T, Lee C, Kulis S, et al. Acculturation among Mexican-heritage preadolescents: A latent class analysis. Social Science Research 2011;40:1236-48. 15. Zhang W, Lei K. The Urban New Immigrants' Social Inclusion: Internal structure, present situation and influential factors. Sociological Study 2008:117-41, 244-45. 16. Barona A, Miller J. Short acculturation scale for Hispanic youth (SASH-Y). Hispanic Journal of

Behavioral Sciences 1994;16:155–62. on October 1, 2021 by guest. Protected copyright. 17. Berry J. Acculturation as varieties of adaptation. In Acculturation: Theory, models and some new findings. Westview, 1980. 18. Shu XO, Zheng W, Potischman N, et al. A population-based case-control study of dietary factors and endometrial cancer in Shanghai, People's Republic of China. Am J Epidemiol 1993;137:155-65. 19. Mo X. Reflections on the Age Boundary of "Young People". Journal of Guangxi Youth Leaders College 2009;19(2):38-40. 20. Wiley JF, Cloutier MM, Wakefield DB, et al. Acculturation determines BMI percentile and noncore food intake in Hispanic children. The Journal of nutrition 2014;144:305-10. 21. Gaston N, Nelson D. International Migration. Oecd Journal Economic Studies 2011;1(3):1-1

VERSION 2 – REVIEW

REVIEWER Myoungock Jang USA REVIEW RETURNED 09-Dec-2017

BMJ Open: first published as 10.1136/bmjopen-2017-018768 on 22 June 2018. Downloaded from

GENERAL COMMENTS Thank you for addressing the comments and revising the manuscript. The manuscript is overall much improved to enhance clarification. Background has a clear description of geographical difference of social and environmental factors related to body weight status. I would recommend to point that the relationship of acculturation and health behaviors depends on where they are acculturated. Guangzhou is considered as healthy environment, and thus children acculturated to the healthy environment adopt healthy lifestyle. If people migrate to healthy environment, then we need to encourage acculturation and help them adopt the healthy lifestyle. Need to check grammars.

Here are minor comments. Abstract Measures: Readers may want to have information of measures for social anxiety, food intake as well as brief description of data analysis.

Background On page4, it is ambiguous to use the terms here, developing countries vs. developed countries. Could you be more specifying the country? For example, United States vs. Asian countries or specific County

Discussion The second paragraph on page 14, it is confusing to divide the level of acculturation by middle and high since there is nowhere to describe the level of acculturation. You described the instrument is used as a continuum. If there is a cutpoint to divide the level of acculturation, you need to describe that in the measurement section.

REVIEWER Zhihong Sa http://bmjopen.bmj.com/ Beijing Normal University REVIEW RETURNED 10-Dec-2017

GENERAL COMMENTS The manuscript has improved in various ways. While the focus on the association between acculturation and body weight status among migrant children in China is new, the present study has brought about some conceptual/policy and methodological concerns

for the authors to consider. on October 1, 2021 by guest. Protected copyright.

1． While a negative association between acculturation and overweight/obesity was found among migrant children in the present study, without taking into account the potentially important confounding effects of physical activities, the observed association between acculturation and overweight/obesity may possibly occur by chance. Acculturation is not significantly associated with overweight/obesity in the bivariate analysis. Despite the significant association between acculturation and overweight/obesity in multivariate analysis, the magnitude of the association between these two variables is modest.

2. Overall, the prevalence of overweight/obesity is lower among rural children than that of urban chidren in China. Although child obesity rate in Guangzhou is lower than that in other big cities, it has been increasing rather rapidly in recent years as suggested in the literature. Children in Guangzhou perhaps like children in other

BMJ Open: first published as 10.1136/bmjopen-2017-018768 on 22 June 2018. Downloaded from Chinese cities are exposed to more obesity-related risk factors than migrant children, which include intake of more high-fat and fast food, insufficient physical activity, longer hours of TV watching and game playing, heavier school work and grantparenting. While the representative data comparing the prevalence of overweight/obesity between local and migrant children in Guangzhou is not available, it is possible that the prevalence of overweight/obesity among local children is similar to (as it is indicated in the present study) or even higher than that of migrant children. In this case, the authors’ policy suggestion for promoting acculturation to prevent childhood overweight and obesity may need to be reconsidered.

3. Results: It’s more useful to show the full model in Table 4.

4. Discussion: The study is based on self-reported weight and height which may has some reporting error. The data limitation should also be mentioned.

The authors mentioned (p. 15, end of 1st paragraph) that “this study suggest that further acculturation services for migrant children should focus not only on the prevention of childhood overweight and obesity but also on the promotion of a healthy body image”. Maintaiing a healthy body weight is an important policy goal, but policy advocacy should be careful not to stigmatize heavy children.

VERSION 2 – AUTHOR RESPONSE

Answers to Reviewers:

Reviewer #1s' Comments http://bmjopen.bmj.com/

The manuscript has improved in various ways. While the focus on the association between acculturation and body weight status among migrant children in China is new, the present study has brought about some conceptual/policy and methodological concerns for the authors to consider.

Author’s reply: Thank you very much for giving us helpful advices to improve our manuscript. We on October 1, 2021 by guest. Protected copyright. have given corresponding replies and revised point by point below.

1) While a negative association between acculturation and overweight/obesity was found among migrant children in the present study, without taking into account the potentially important confounding effects of physical activities, the observed association between acculturation and overweight/obesity may possibly occur by chance. Acculturation is not significantly associated with overweight/obesity in the bivariate analysis. Despite the significant association between acculturation and overweight/obesity in multivariate analysis, the magnitude of the association between these two variables is modest.

Author’s reply: First, in this study, in addition to acculturation, we considered other factors associated with overweight/obesity among migrant children including demographic characteristic of migrant children and their caregivers, children’s social anxiety, and food intake. Second, in this study, we only

BMJ Open: first published as 10.1136/bmjopen-2017-018768 on 22 June 2018. Downloaded from measured migrant children’s physical activities at school but no differences were found, and after school activities were not measured. Consequently, we did not include physical activities as a confounding factor, and we agreed with the reviewer that it is a limitation of the current study. Additionally, we conducted a literature review on after school activities among secondary school students in Guangdong Province, only one manuscript demonstrated that the variation of extracurricular physical activities of middle school students in Guangzhou was small. Virtually all of students (97.25%) participated in extracurricular physical activities from Monday to Friday. [1]. Therefore, comprehensive measurements on physical exercise among students should be considered in the future. We have revised the limitation accordingly (p. 18, 1st paragraph)

2) Overall, the prevalence of overweight/obesity is lower among rural children than that of urban children in China. Although child obesity rate in Guangzhou is lower than that in other big cities, it has been increasing rather rapidly in recent years as suggested in the literature. Children in Guangzhou perhaps like children in other Chinese cities are exposed to more obesity-related risk factors than migrant children, which include intake of more high-fat and fast food, insufficient physical activity, longer hours of TV watching and game playing, heavier school work and grandparenting. While the representative data comparing the prevalence of overweight/obesity between local and migrant children in Guangzhou is not available, it is possible that the prevalence of overweight/obesity among local children is similar to (as it is indicated in the present study) or even higher than that of migrant children. In this case, the authors’ policy suggestion for promoting acculturation to prevent childhood overweight and obesity may need to be reconsidered.

Author’s reply: We thank the reviewer for pointing out this comment. To our best knowledge, there were no data on body weight status among local children in Guangzhou. However, the prevalence of obesity among migrant children found in this study (3.1%) was close to that among students aged 12- 18 (3.4%) in Guangzhou in 2013 [2]. Therefore, we agreed with the reviewer that it is possible that the prevalence of overweight/obesity among local children is similar to that of migrant children, but further http://bmjopen.bmj.com/ empirical evidence on the difference between local and migrant children is needed. We have added this statement in the limitation of study (p. 18, 1st paragraph).

In addition, in general, children in Guangzhou had lower BMI than those in other cities, and had lower overweight/obesity prevalence than that among children in urban China in [2-4]. Therefore, we hypnotized that the healthier body weight status among children in Guangzhou could be explained by on October 1, 2021 by guest. Protected copyright. healthier environment and life habits, such as Cantonese diets and health propaganda in the newly emerging media [5-7].

Based on the above mentioned information and the reviewer’s comments, we have revised our policy suggestion by adding advocating policy and social campaign on healthy body weight among children, not only among migrant children (p. 16, 1st paragraph and conclusion).

Results:

3) It’s more useful to show the full model in Table 4.

BMJ Open: first published as 10.1136/bmjopen-2017-018768 on 22 June 2018. Downloaded from Author’s reply: We have revised and showed the full model in Table 4.

Discussion:

4) The study is based on self-reported weight and height which may has some reporting error. The data limitation should also be mentioned.

Author’s reply: We have added the description about the limitation of the self-reported data (p. 18, 1st paragraph).

5) The authors mentioned (p. 15, end of 1st paragraph) that “this study suggest that further acculturation services for migrant children should focus not only on the prevention of childhood overweight and obesity but also on the promotion of a healthy body image”. Maintaining a healthy body weight is an important policy goal, but policy advocacy should be careful not to stigmatize heavy children.

Author’s reply: Thanks for the reviewer’s advices concerning policy advocacy. Social campaigns on promoting healthy body image have been launched in many countries, and have succeeded in improving immoderate eating and dysfunctional exercise for obesity prevention [8]. Grogan provided evidence to support the effectiveness of programs like psycho-social interventions, which was aimed to promote a healthy body image among both boys and girls [9]. Furthermore, the importance of developing a healthy body image during adolescence should be highlighted on the association between healthy body weight and health consequences, rather than focusing on heavy children’s

weight-related pressures and stigmatization [8, 10]. We have revised the discussion accordingly (p. http://bmjopen.bmj.com/ 16, 1st paragraph).

Reviewer #2s' Comments

on October 1, 2021 by guest. Protected copyright. Thank you for addressing the comments and revising the manuscript. The manuscript is overall much improved to enhance clarification. Background has a clear description of geographical difference of social and environmental factors related to body weight status. I would recommend to point that the relationship of acculturation and health behaviors depends on where they are acculturated. Guangzhou is considered as healthy environment, and thus children acculturated to the healthy environment adopt healthy lifestyle. If people migrate to healthy environment, then we need to encourage acculturation and help them adopt the healthy lifestyle. Need to check grammars. Here are minor comments.

Author’s reply: Thanks the reviewer for giving us useful comments. We have revised one by one, and check grammars in our whole manuscript.

Abstract

BMJ Open: first published as 10.1136/bmjopen-2017-018768 on 22 June 2018. Downloaded from 1) Measures: Readers may want to have information of measures for social anxiety, food intake as well as brief description of data analysis.

Author’s reply: We have agreed with the reviewer and added the description of measures for social anxiety, food intake and data analysis in the measure section (p. 2, 3rd paragraph).

Background

2) On page4, it is ambiguous to use the terms here, developing countries vs. developed countries. Could you be more specifying the country? For example, United States vs. Asian countries or specific County

Author’s reply: The data (developing countries vs. developed countries) was from a systematic analysis on the data from 188 countries and 21 regions globally (Ng et al, Lancet. 2014). The specific countries are hard to be listed. We have added information of this study before showing the data (p. 5, 3rd paragraph).

Discussion

3) The second paragraph on page 14, it is confusing to divide the level of acculturation by middle and high since there is nowhere to describe the level of acculturation. You described the instrument is used as a continuum. If there is a cut-point to divide the level of acculturation, you need to describe that in the measurement section.

http://bmjopen.bmj.com/

Author’s reply: The acculturation in our study was a continuous variable ranged from 0-100 points without a cut-point. In our study, the average total score of acculturation was 41.8 (SD=14.6) out of 100. We have changed the description of the acculturation status among internal migrant children “in the middle level” as the “not high” (p. 15, 2nd paragraph).

on October 1, 2021 by guest. Protected copyright.

Thanks for your reconsideration.

Sincerely Yours,

Xiaoling Huang

Contact details:

Address: 74, Zhongshan Road 2, Faculty of Medical Statistics and Epidemiology, School of Public Health, Sun Yat-sen University, Guangzhou, China

E-mail: Xiaoling Huang huangxlxpxc1163.com

BMJ Open: first published as 10.1136/bmjopen-2017-018768 on 22 June 2018. Downloaded from Corresponding author: Li Ling [email protected]

Tel/Fax: 13530115220; 86-020-87333319.

References:

1. Wang Z, Liu M: Influential Factors and Current Situation of Extra-curriculum P.E. of Students of Guangzhou Middle School. Journal of Harbin Institute of Physical Education 2008, 26(2):64-66.

2. Mai J: A Study on the Relations between obesity and dietary behaviors, physical activities among adolescents: Taking some students in Guangzhou China as an example Chengdu, China: Sichuan Normal University; 2014.

3. State General Administration of Sports of the People's Republic of China: Report on the physical fitness and health survey among Chinese students in 2014: Higher Education Press; 2016.

4. Zhao G, He W: Investigation on overweight and obesity of urban children and teenager in Guangzhou. China Sport Science and Technoligy 2008, 44:103-106.

5. Liu B, Jing J, Mai J, Chen Y, Xu G, Bao P, Yu M, Zhang W, Y P: Secular trends of overweight and obesity prevalence between 2007 and 2011 in children and adolescents in Guangzhou. Chinese Journal of Prevention Medicine 2014, 48(4):312-317.

6. Xue W: Cantonese cuisine: the perfect unity of nutritious food and cultural custom. Science and Technology of West China 2005:17-22.

7. Boyce T: The media and obesity. obesity reviews 2007, 8(Suppl. 1):201-205.

8. Voelker DK, Reel JJ, Greenleaf C: Weight status and body image perceptions in adolescents: current perspectives. Adolescent health, medicine and therapeutics 2015, 6:149-158. http://bmjopen.bmj.com/ 9. Grogan S: Promoting Positive Body Image in Males and Females: Contemporary Issues and Future Directions. Sex Roles 2010, 63:757–765.

10. Antin TMJ: Contradictions in Food Choice and Body Image: Implications for Obesity Prevention. Dissertations & Theses - Gradworks 2012.

REVIEWER Myoungock Jang USA REVIEW RETURNED 12-Jan-2018

GENERAL COMMENTS Thank you for addressing the comments and revising the manuscript. Overall, improved, but still there are some areas need to clarify and reorganize to enhance the scientific writing.

Strengths and limitations of the study P 4 line 15-16: Self-report height and weight are usually related to social desirability. Recall bias is related to an individual’s memory to report the past event or experience. Background Authors tried to improve clarity of the research gaps and purpose of the study. They provided some statistics to help readers understand

BMJ Open: first published as 10.1136/bmjopen-2017-018768 on 22 June 2018. Downloaded from the different environmental factors that may work here for the relationship between acculturation and weight status in a typical area in China. But still needs some work on organizing ideas about what they want to focus on. Page6-7 the authors need to clearly state the concept of acculturation and its influence on health and health behaviors. More specifically, the last paragraph on page 6, the reason of high prevalence of obesity in immigrant children in developed country is that they immigrated to obesogenic society where the environment trigger obesity, not just simply movement increases the risk of being overweight or obese. Need to state that we need to think differently like the migration into healthy environment may decrease the risk of being overweight or obese. Page 7, the last paragraph, the inconsistency is related to the complication of understanding and capture the process of acculturation, not just age difference of children. Need to work on the systematic Page 8, the authors provided confounding factors with no justification why they are important to consider. Please add a brief justification of those factors. And, please clarify the “gap” between sending and receiving communities. What’s the gap here? Is it SES or cultural practices?

Page 8, authors do not need to mention the instrument development if it is not the purpose of the study.

Methods:

Page 10, there is too specific information of the instrument development. If you already developed and reported somewhere else, you do not need to specify the process of development. Page 11, was children’s social anxiety mentioned on the background? The first sentence of the section needs to be moved to

background. http://bmjopen.bmj.com/

Results Page 13, line 45-50, the last sentence of the paragraph may be moved to the third in the paragraph.

Discussion

Overall, did not discussed null findings. I expected to see some discussion about diet and children’s anxiety since they targeted to on October 1, 2021 by guest. Protected copyright. consider those confounding factors to see the relationship. On page 15, line 14-17, the diet and cultural differences within China need to be emphasized. “Healthy lifestyle adaption in the main society may improve weight status in migrant children” is a key message from the study. On page 16, line 10-15, is the diet style (fast food consumption) observed in migrant children as well?

REVIEWER Zhihong Sa School of Sociology, Beijing Normal University, China REVIEW RETURNED 19-Jan-2018

GENERAL COMMENTS The authors addressed the questions in earlier versions of the review appropriately and revised the manuscript carefully. The quality of the manuscipt has improved. I have no further questions and concerns.

BMJ Open: first published as 10.1136/bmjopen-2017-018768 on 22 June 2018. Downloaded from VERSION 3 – AUTHOR RESPONSE

Answers to Reviewers:

Reviewer #1s' Comments

The authors addressed the questions in earlier versions of the review appropriately and revised the manuscript carefully. The quality of the manuscript has improved. I have no further questions and concerns.

Author’s reply: Thank you very much for your helpful advice to improve our manuscript.

Reviewer #2s' Comments

Thank you for addressing the comments and revising the manuscript. Overall, improved, but still there are some areas need to clarify and reorganize to enhance the scientific writing.

Author’s reply: Thank you very much for giving us valuable advices to revise and improve our manuscript. We have given corresponding replies point by point and revised carefully.

Strengths and limitations of the study 1) P 4 line 15-16: Self-report height and weight are usually related to social desirability. Recall bias is related to an individual’s memory to report the past event or experience.

Author’s reply: Thank the reviewer for clearly stating the concept of reporting bias and recall bias. The limitation of self-report height and weight included both reporting bias and recall bias. We have corrected the limitation of self-report height and weight in the manuscript (page 4, line 7-8 & limitation section on page 18, line 13-14).

Background http://bmjopen.bmj.com/ Authors tried to improve clarity of the research gaps and purpose of the study. They provided some statistics to help readers understand the different environmental factors that may work here for the relationship between acculturation and weight status in a typical area in China. But still needs some work on organizing ideas about what they want to focus on.

2) Page 6-7 the authors need to clearly state the concept of acculturation and its influence on health

and health behaviors. More specifically, the last paragraph on page 6, the reason of high prevalence on October 1, 2021 by guest. Protected copyright. of obesity in immigrant children in the developed country is that they immigrated to an obesogenic society where the environment trigger obesity, not just simply movement increases the risk of being overweight or obese. Need to state that we need to think differently like the migration into healthy environment may decrease the risk of being overweight or obese.

Author’s reply: We thank the reviewer for pointing out this critical comment. First, as we explained in the “background” section, acculturation is a multidimensional concept and reflects the overall adaptation to a new society, including language, social interaction, preferences in life (e.g., diet, dressing, custom, et al.) [1, 2]. Particularly, during the migration, the change of environment and adaptation process may influence their diet, physical activity and other life behaviors, which may further be bidirectionally associated with health status. For example, on the one hand, the unhealthy body weight status may be explained by exposure to obesity-related risk factors during and after resettlement in new obesogenic environments, and adaptation to unhealthy behaviors, such as intake of high-calorie food, insufficient exercise, and sedentary [3, 4]. On the other hand, migration into the healthy environment may reduce the risk of being overweight or obese. As a study show that the

BMJ Open: first published as 10.1136/bmjopen-2017-018768 on 22 June 2018. Downloaded from acculturated parents of migrant children were aware of the health issues associated with obesity as migrated and received local attitudes and practices of a healthy lifestyle in main society [5]. We have revised correspondingly (page 6, 3rd paragraph & page 7, 2nd paragraph).

3) Page 7, the last paragraph, the inconsistency is related to the complication of understanding and capture the process of acculturation, not just age difference of children. Need to work on the systematic

Author’s reply: We agreed the reviewer’s comments. Here, we meant to explain children with various age could have notable differences in the self-evaluation of body image across culture, and the views of acculturation might vary person to person who had different cultural background [6]. We have revised in the manuscript (Page 8, 1st paragraph).

4) Page 8, the authors provided confounding factors with no justification why they are important to consider. Please add a brief justification of those factors. And, please clarify the “gap” between sending and receiving communities. What’s the gap here? Is it SES or cultural practices?

Author’s reply: The “gap” between sending and receiving communities is the complex diversity, involving the environmental, socioeconomic, cultural and behavioral difference [7]. We have added a brief statement of the confounding factors and revised the description of the “gap” in the manuscript (page 8, 2nd paragraph).

5) Page 8, authors do not need to mention the instrument development if it is not the purpose of the study.

Author’s reply: The development of instrument in our study was not reported in other articles. In order to help the audience better understand our study process and the instrument validation, we have provided the instrument development and enclosed it as an appendix. We agreed with the reviewer and simplified the introduction of instrument development. The validation was enclosed in Table A-2 in the appendix. We have revised and simplify the related description in the methods section (page 8, http://bmjopen.bmj.com/ 3rd paragraph & page 10, 3rd paragraph).

Methods: 6) Page 10, there is too specific information of the instrument development. If you already developed and reported somewhere else, you do not need to specify the process of development.

Author’s reply: Our responses to the fifth comment also apply to this comment. We have been revised on October 1, 2021 by guest. Protected copyright. correspondingly (page 10, 3rd paragraph).

7) Page 11, was children’s social anxiety mentioned on the background? The first sentence of the section needs to be moved to background.

Author’s reply: The mental health was mentioned in the background. We had moved the first sentence of this section to the background and revised corresponding contents on the background (page 8, 2nd paragraph & page 11, 3rd paragraph).

Results 8) Page 13, line 45-50, the last sentence of the paragraph may be moved to the third in the paragraph.

Author’s reply: We have moved the last sentence of the paragraph to the third in the paragraph (page 13, 3rd paragraph).

BMJ Open: first published as 10.1136/bmjopen-2017-018768 on 22 June 2018. Downloaded from

Discussion 9) Overall, did not discussed null findings. I expected to see some discussion about diet and children’s anxiety since they targeted to consider those confounding factors to see the relationship.

Author’s reply: Previous studies found that social anxiety and food intake had both relationships with acculturation and obesity. We, therefore, took these factors into consideration in the regression models. However, these factors were not significant in the model 2, and we did not further discuss the relationship.

Nevertheless, in our study, acculturation included the acculturation in diet, and high level of acculturation in diet reflected the high consistency with local diet. As discussed in the manuscript (page 16, 1st paragraph), diet in Guangzhou is famous for its healthy Cantonese diet culture, which might be one of the possible explanations of our results on the association between the low levels of acculturation and overweight/obesity. In order to better understand the relationship between social anxiety and acculturation in diet, we also have added related analysis of the relationship. However, their association was not statistically significant in our study (see the following Table). Therefore, we did not further report these figures.

Table 1-1 Linear model on the association between social anxiety and acculturation in diet among migrant children in Guangzhou, China 2016 Variable Model 1 Model 2 B (95%CI) P B (95%CI) P Social anxiety -0.24(-0.53,0.05) 0.110 -0.27(-0.57,0.04) 0.085 * Model 1 was bivariate linear regression model. Model 2 was a multiple linear model adjusted for gender, age, years of residence in Guangzhou, caregivers’ education, caregivers’ migration type, household monthly income per person, number of children in family.

10) On page 15, line 14-17, the diet and cultural differences within China need to be emphasized.

“Healthy lifestyle adaption in the main society may improve weight status in migrant children” is a key http://bmjopen.bmj.com/ message from the study.

Author’s reply: We agreed with the reviewer’s comment. In our study, the prevalence of obesity found (3.1%) was close to that among students aged 12-18 (3.4%) in Guangzhou (2013), in which children had lower BMI than those in other cities. However, the prevalence of obesity among migrants in this age group in other regions of China was scanty. Meanwhile, it is notable that China is a vast country

with diversity, and children in various regions had different prevalence of obesity (Girls: 0.44%- on October 1, 2021 by guest. Protected copyright. 10.00% vs. Boys: 0.56%-21.67%) [8, 9]. The spatial, diet and cultural disparity within China may be the significant reasons of differences in obesity rates and needed to be further explored [10, 11]. We had added and highlighted the diet and cultural differences within China (page 15, 1st paragraph).

11) On page 16, line 10-15, is the diet style (fast food consumption) observed in migrant children as well?

Author’s reply: The fast-food consumption was not investigated in our study. However, we conducted a literature review on fast food consumption among internal migrant children. We found that this diet style (fast food consumption) exist among children in Guangzhou and other multicultural metropolis. Therefore, further empirical evidence among migrant children is needed.

Thanks for your reconsideration.

Sincerely Yours,

BMJ Open: first published as 10.1136/bmjopen-2017-018768 on 22 June 2018. Downloaded from

Xiaoling Huang

Contact details: Address: No. 74, Zhongshan Road 2, Faculty of Medical Statistics and Epidemiology, School of Public Health, Sun Yat-sen University, Guangzhou, China E-mail: Xiaoling Huang huangxlxpxc1163.com Corresponding author: Li Ling [email protected] Tel/Fax: 86-13530115220; 86-20-87333319.

Reference: 1. Barry DT, Garner DM: Eating concerns in East Asian immigrants: Relationships between acculturation, self-construal, ethnic identity, gender, psychological functioning and eating concerns. EatWeight Disord 2000, 6:90-98. 2. Unger JB, Ritt-Olson A, KarlaWagner, Soto D, Baezconde-Garbanati L: A Comparison of Acculturation Measures Among Hispanic/Latino Adolescents. J Youth Adolescence 2007, 36:555– 565. 3. Liu JH, Chu YH, Frongillo EA, Probst JC: Generation and acculturation status are associated with dietary intake and body weight in Mexican American adolescents. The Journal of nutrition 2012, 142:298-305. 4. Liu WJ, Liu W, Lin R, Li B, Pallan M, Cheng KK, Adab P: Socioeconomic determinants of childhood obesity among primary school children in Guangzhou, China. BMC public health 2016, 16:482. 5. Chen JL, Weiss S, Heyman MB, Lustig R: Risk factors for obesity and high blood pressure in Chinese American children: maternal acculturation and children's food choices. Journal of immigrant and minority health / Center for Minority Public Health 2011, 13(2):268-275. 6. Unger JB, Gallaher P, Shakib S, Ritt-Olson A, Palmer PH, Johnson CA: The AHIMSA Acculturation Scale: A new measure of acculturation for adolescents in a multicultural society. Journal of Early Adolescence 2001, 22:225-251. 7. Ebrahim S, Kinra S, Bowen L, Andersen E, Ben-Shlomo Y, Lyngdoh T, Ramakrishnan L, Ahuja RC,

Joshi P, Das SM et al: The effect of rural-to-urban migration on obesity and diabetes in India: a cross- http://bmjopen.bmj.com/ sectional study. PLoS Med 2010, 7:e1000268. 8. Ma J, Cai C, Wang H, Dong B, Song Y, Hu P, Zhang B: The trend analysis of overweight and obesity in Chinese students during 1985-2010. Chin j Prev Med 2012, 46:776-780. 9. Lu S, Pan D, Liu J, Cai L, Zhu Y, yang W, Li X, Ma J, Jing J, Chen Y: Comparisons of obesity prevalence in children aged 7 to 12 years old in Liaoning ang Guangdong Province. Chin J Sch Health 2017, 38:173-176.

10. Mai J: A study on the relationship between obesity and dietary behaviors, physical activities on October 1, 2021 by guest. Protected copyright. among adolescents: Taking some students in Guangzhou China as an example Chengdu, China: Sichuan Normal University; 2014. 11. Song F, Cho MS: Geography of Food Consumption Patterns between South and North China. Food 2017, 6(5):34.

VERSION 4 – REVIEW

REVIEWER Myoungock Jang USA REVIEW RETURNED 29-Mar-2018

GENERAL COMMENTS Overall, the manuscript has been improved so much. The authors well addressed the reviewers' comments. There is a still statistical inconsistency to assess the relationships of the variables. For example, they stated food intake is a mediating

BMJ Open: first published as 10.1136/bmjopen-2017-018768 on 22 June 2018. Downloaded from factor, but they controlled for food intake when they assess the relationship between acculturation and overweight/obesity in children.

There is also a minor change you may need to make.Please be consistent when you use "migrant" within a country, i.e., in China vs. "immigrant" from different country.

VERSION 4 – AUTHOR RESPONSE Answers to Reviewers:

Author’s reply: We really appreciate the reviewer for useful suggestions to improve our manuscript.

Reviewer #2s' Comments

1) There is a still statistical inconsistency to assess the relationships of the variables. For example, they stated food intake is a mediating factor, but they controlled for food intake when they assess the relationship between acculturation and overweight/obesity in children.

Author’s reply: As mentioned in the background (page 8, 2nd paragraph), food intake, as an important factor of energy intake, was controlled to exam the relationship between acculturation and body weight status among immigrant children based on previous research. [1] The aim of the study was to assess the association between acculturation and overweight/obesity among internal migrant children by controlling for food intake and other confounding factors. We have corrected the statement of statistical inconsistency about food intake in our manuscript (page 12, 1st paragraph).

2) There is also a minor change you may need to make. Please be consistent when you use "migrant" within a country, i.e., in China vs. "immigrant" from different country. http://bmjopen.bmj.com/ Author’s reply: Thank the reviewer for clearly stating the difference of migrant and immigrant. We have corrected the statement and kept the concept consistent in the manuscript (background and discussion).

Thanks for your reconsideration.

Xiaoling Huang

[1] Wojcicki JM, Schwartz N, Jimenez-Cruz A, et al. Acculturation, dietary practices and risk for childhood obesity in an ethnically heterogeneous population of Latino school children in the San Francisco bay area. Journal of immigrant and minority health 2012;14:533-9.