Rural Food Environment, Caregiver Inuence and School-Aged Children’s Unhealthy Food Choice: New Evidence From Rural China

A large number of unhealthy foods with low nutrition content and safety issues exist in the rural food consumption environment of China. Guiding rural children to avoid eating unhealthy foods through family intervention has become much more critical. However, when current rural caregivers are once left behind by their parents and lack of family education in their childhood, how would they affect their children’s unhealthy food choices? The purpose of this study is to highlight the new evidence of current rural caregivers’ inuences on children’s unhealthy food choices in China. Our results showed that sex and were more easily to than girls and younger children


Abstract
Background A large number of unhealthy foods with low nutrition content and safety issues exist in the rural food consumption environment of China. Guiding rural children to avoid eating unhealthy foods through family intervention has become much more critical. However, when current rural caregivers are once left behind by their parents and lack of family education in their childhood, how would they affect their children's unhealthy food choices? The purpose of this study is to highlight the new evidence of current rural caregivers' in uences on children's unhealthy food choices in China.

Methods
Based on a survey of 5611 pairs of rural school-aged children and their caregivers in seven provinces, the signi cant effects of caregivers' types, parenting attitudes, and socio-demographic characteristics on children's unhealthy food choices were evaluated through an ordered probit model.

Results
This study found that rural parents did not perform positive in uence than rural grandparents as conventionally believed. When the caregiver type changing from parents to grandparents, the percentage of children who mildly accept unhealthy foods increased 6.94%, while the percentage of children who moderately and severely accept unhealthy foods decreased 6.48% and 0.46%, respectively. In addition, this study underscored the signi cance of caregivers' neglectful parenting attitudes on the impact of children's unhealthy eating behavior. When caregivers' parenting attitudes changing from doting attitude to neglectful attitude, the percentage of children who mildly accept unhealthy foods decreased 9.75%, while the percentage of children who moderately and severely accept unhealthy foods increased 8.77% and 0.99%, respectively.

Conclusions
Our study proposes new evidence that the current rural parents' inherent intention and preference for those unhealthy foods formed in their childhood would be much higher than rural grandparents and might have performed a signi cantly negative in uence on their children. Our ndings also have policy implications for health promotion professionals who look for intervention targets of caregivers to improve their health education effectiveness for children. For the health education toward rural caregivers, instead of focusing on the caregiver types, segmenting those rural caregivers with neglectful parenting attitude would be the key step to identify the intervention targets.

Page 3/20
Background Evidence has indicated that the food environment in many rural areas is affected by high rates of poverty and limited access to su cient and nutritious food compared to that in urban areas [ 1 ]. However, little attention has been paid to the rural food environment lled with foods with adequate supply but made with substandard components or are unhygienically prepared [ 2 -3 ]. Rural China is facing this serious problem now and a large number of unhealthy foods with low nutrition content and safety issues exist in the rural food consumption environment. Under such circumstances, guiding rural children to avoid eating unhealthy foods through family intervention has become much more critical. Parents are usually recognized as crucial health promoters compared to other caregivers [ 4 -5 -6 ]. The parental migration from the rural area to the urban area of China for better employment opportunities has always been criticized in the literatures for causing the weakening role of rural parents [ 7 -8 -9 ]. However, considering parental migration phenomenon has occurred over forty years, those left behind children without parental in uences have grown up and become new generation of parents. How would they in uence their children? Would their eating perceptions and behaviors formed in their childhood further in uence their children? Rural caregiver in uence on children's unhealthy eating behavior need new evidence and further investigation.

School-aged children's unhealthy food choice under current rural food environment
In rural area of China, foods processed by small-scale manufacturers in unhygienic conditions with bad packaging material are widely distributed and sold by street vendors or grocery shops in unsanitary environments. In addition, food fraud problems, such as imitated foods and those with illegal ingredients and additives, commonly existed in the rural market which may result in serious public health consequences when the adulterant is toxic or allergenic [ 10 -11 ]. It has been reported that more than 80% of food safety incidents and 90% of foodborne diseases originate in the rural areas of China [ 12 -13 ]. The food environment in the rural area has posed the serious threat to the health of rural residents, especially the health of rural children. News have reported some rural children have experienced diarrhea, vomiting, and allergic reactions after eating unsafe foods, such as spicy strips [ 14 -15 ]. In 2011, a one-year-old girl in rural area died after eating fried chicken to which an excessive amount of nitrite had been added [ 16 ].
However, when widely exposed in this food environment, rural school-aged children are extremely attracted to these unsafe foods because they are cheap, colorful and tasty [ 17 -18 ]. After school, street vendors and grocery shops near to school are frequented by children who are looking for their favorite snacks [ 19 ]. It is a dangerous sign that rural children are accepting and preferring these unhealthy foods in their daily diets. Since eating habits formed during childhood have a life-long in uence on health, protecting children from the access of the foods with poor quality and safety under the current rural environment have become critical and urgent.
Rural caregiver's in uence on school-aged children's health and food choice The signi cant family in uences on children's eating behaviors have been widely studied [ 20 -21 -22 ].
However, the situation of caregiver in uence in rural China is complicated. Over the past 40 years, millions of rural parents have left their hometown and migrated to urban areas for better employment opportunities, resulting an estimated 28.52% of children are left and cared by their grandparents or other relatives [ 23 ]. The long-term absence of parents has been widely reported as a negative in uencing factor for rural children's both physical and mental health [ 24 -25 -26 ]. Many studies have reported that the left behind children's healthy dietary perception levels are signi cantly lower than other children [ 27 -28 -29 ].
Their common viewpoint is that the left behind children are less restricted by grandparents or other caregivers than by parents and more easily to access unhealthy foods [ 30 -31 ]. However, there are also studies indicating no signi cant difference in the nutrition status, food safety perceptions and unhealthy eating behaviors between left behind children and children of non-migrants [ 32 -33 ]. Zhou et al. even pointed out that both groups of children perform poorly on health and nutrition outcomes regardless of parental migration status [ 34 ].
A fact which should not be ignored is that the current generation of parents might once be left behind children who were lack of parental education in their childhood. When they become parents, how would they in uence their next generation? Would the role of rural parents be still critical as health promoter as expected compared to other caregiver types? This study attempted to apply empirical studies to explore the caregiver in uence on children's unhealthy food choice from the perspective of intergenerational effect.
In addition, caregivers' parenting attitudes, eating habits and other socio-demographic characteristics have been reported to in uence children's eating di culties and promote children's nutritionally-balanced diet [ 35 -36 ]. However, what type of parenting attitude speci cally negatively affect rural children's eating behavior toward unsafe foods has been also rarely discussed. Therefore, we proposed our research questions: could current rural parent in China perform better at in uencing children to make healthier and safer food choices than the other caregiver types? What factors of rural caregivers signi cantly affect children's safe eating behavior?
To summarize, the aim of this study was to (1) assess the new evidence of how caregiver types affecting their children's unhealthy food choice in the current rural food environment; (2) explore the signi cant in uencing factors of caregivers on children's unhealthy food choices.

Participants
This study utilized a cross-sectional design. To ensure better representativeness of the Chinese population, a large-scale survey was launched in the rural areas of 7 provinces located in the north (2 provinces), east (2 provinces), central (1 province), southwest (1 province) and south area (1 province) of China in 2018. A total of 6203 pairs of fourth to sixth grades children and their caregivers from 13 elementary schools were recruited as study participants. After excluding 592 refusals or invalid responses, 5611 valid questionnaires were obtained with an effective response rate of 90.5%. The sociodemographic characteristics of the surveyed children and their caregivers are presented in Table 1. In total, 2872 boys and 1739 girls with an average age of 11.4 years (SD = 1.02 years) participated in the survey. Furthermore, 56.1% of the caregivers were female and 89.2% were aged less than 60 years. Most caregivers were the children's parents (80.7%), 17.8% were their grandparents and 1.5% were children's relatives or others, indicating the existence of left behind children. In addition, 56.3% of caregivers were farmers, 82.2% did not receive a high school education, and 83.7% of their families had a monthly income of less than 6,000 RMB. These demographic characteristics are representative of current conditions in the rural areas of China.

Procedures
Since the participants in the study are both children and their legal guardians, addressing ethical considerations, at the beginning of the questionnaire a statement was clearly explained the purpose of the study and informed that all participants (including caregivers and their children) completing the questionnaire was optional and if caregivers completed and returned the questionnaires, we would assume that they approved us to use the information both they and their children provided. Therefore informed consent to participate in the study has been obtained from all participants, including legal guardians of minors.
Trained researchers explained to the children of fourth to sixth grades how to complete the questionnaires. Children completed their part independently at school and took the rest of the questionnaire home. The major caregivers were asked to complete their sections and then children returned them to teachers within three days. The researchers collected all the returned questionnaires.

Measures
Rural caregivers' parenting attitudes related to children's eating habit and their segmentation The rural caregivers' parenting attitudes related to children's eating habit was rst determined. We gave a full consideration of the general requirements of the parental attitudes in the child feeding questionnaire in the literature [ 37 ] and the actual conditions of rural China and designed nine statements. Speci cally, the descriptive statements with random number explored rural caregivers' attitudes toward developing eating habit for nutritionally-balanced diet (Nos 1, 2, 3, 7 and 8) and food safety assurance (Nos 4), knowledge level toward obesity (Nos 5) and food safety (Nos 6), and attitudes about the necessity to guide children to properly purchase foods (Nos 9). The caregiver participants were asked to score their answers on a ve-point Likert scale (1=strongly disagree, 2=fairly disagree, 3=agree, 4=fairly agree, 5=strongly agree) to indicate their agreement with each statement. An exploratory factor analysis was conducted to summarize rural caregivers' parenting attitudes along major dimensions. Thereafter, Kmeans cluster methods were used to segment the participants into distinctive, homogeneous clusters based on their agreements with the dimensions. Rural children and their caregivers' unhealthy food choices under designed eight scenarios with safety risks.
Eight scenarios, all closely related to the real food consumption environment in the rural areas of China, were designed. Each of the eight scenarios involved one type of unhealthy food with safety risk, including foods that were moldy, dirty, unpackaged, foul smelling, imitated, out-of-date, sold by unsanitary street vendors, and had caused problems pertaining to foodborne diseases before. Both children and caregivers' food choices under each scenario were examined and compared.
Effect of rural caregivers' types, parenting attitudes, unhealthy food choice and other socio-demographic characteristics on their children's unhealthy food choice Caregiver factors in uencing children's choices on the unhealthy foods in the eight scenarios were examined. Since the sum of each child's positive responses toward the eight scenarios takes values of (0-8), which are not interval in nature, we expressed the children's choice for unhealthy foods as an ordinal ranking with three categories of arbitrary width: positive choices in zero to one scenario, positive choices in two to four scenarios, and positive choices in ve to eight scenarios. Thus an ordered probit estimation technique was applied to assess the impacts of caregivers factors on the children's choice. All of the analytic plan, including the construction of variable scales, was pre-speci ed. We chose the caregivers' types, parenting attitudes, unhealthy food choice and other socio-demographic characteristics (sex, age, education level, occupation, and income) as independent variables. Children's age and sex were set as control variables. The marginal effects of the independent variables and control variables on children's intention were also analyzed. When more unsafe foods were selected, a change in the sign of the marginal effect from negative to positive reflects an increase in marginal utility, while a positive to negative change indicates a decrease in marginal utility. We also applied the data in the ordered logit model and OLS model and the results showed that the regression coe cients and signi cant regression results of each model were consistent with those of the ordered probit model and the robustness test was passed.

Statistical analysis
Factor-cluster analyses and children's difference comparison were conducted using SPSS v21 (SPSS, Inc., Chicago). The ordered probit model analysis and marginal effects were evaluated in STATA 15.0.

Rural caregivers' parenting attitudes related to children's eating habit and their segmentation
We investigated rural caregivers' parenting attitudes toward healthy and safe eating behavior. According to participants' agreements with the nine statements, we applied factor analysis to group them into three major dimensions that accounted for 58.38% of the total variance ( Table 2). The Kaiser-Meyer-Olkin test value was 0.857, and the result of Bartlett's test was 9357.166 (p<0.001), indicating that the data matrix was suitable for a factor analysis. Cronbach's alpha coe cients were above or around 0.6, con rming acceptable internal consistency. Dimension 1 contained four statements that re ected caregivers' very casual attitudes regarding shaping children's healthy and safe eating habits. Thus, Dimension 1 was named a "casual attitude." Dimension 2 consisted of three statements expressing that caregivers' parenting attitudes were affected by their lack of knowledge on food safety and nutrition. Therefore, it was called an "ignorant attitude." Dimension 3 included two statements and was entitled a "childrenoriented attitude", because the caregivers indulged their children and allow them to do whatever they wanted.
A K-means cluster analysis was used to segment parents with similar sensitivities along the three attitude dimensions. The respondents were distributed into four clusters (Table 3). A discriminant analysis con rmed that the resultant clusters signi cantly varied in all three dimensions (p < 0.01), proving the reliability of the segmentation. Rural caregivers in Cluster 1 (n=1148) were mainly affected by the children-oriented attitude, and thus referred to as "doting caregivers." The caregivers in Cluster 2 (n=1326) were named "unknowledgeable caregivers," because they mainly accepted the viewpoint proposed in the ignorant attitude statements. While those in Cluster 3 (n=452) agreed with all three attitude dimension statements and demonstrated extremely negative parenting attitudes. Thus, they were called "neglectful caregivers." Finally, almost half the caregivers in Cluster 4 (n=2685) disagreed with all three attitude dimension statements. They were named "quali ed caregivers." The segmentation process provided an in-depth understanding of the parenting attitudes of rural caregivers on the importance of both nutrition and food safety. Regarding rural parents, almost half of them demonstrated negative parenting attitudes.
Rural children and their caregivers' unhealthy food choices under designed eight scenarios with safety risks Table 4 provides the results of both rural children and caregivers' food choices under eight scenarios. Rural children showed a relatively higher percentage for accepting foods that were partly moldy (37.1%), unpackaged (35.8%), and foul smelling (24.7%), which all could potentially threaten children's health. Unsurprisingly, children and caregivers illustrated similar tendency at unhealthy food choice. Both rural children and caregivers showed the higher percentage for accepting unpackaged foods with lower prices but high risk of cross contamination and moldy foods.
Effect of rural caregivers' types, parenting attitudes, unhealthy food choice and other socio-demographic characteristics on their children's unhealthy food choice We further investigated the caregiver in uence on children's unhealthy food choice using ordered probit model. All the variables were de ned in Table 5 with their mean value and standard deviation. The multicollinearity test showed that all the variance in ation factors of the variables were less than four, indicating there was no multicollinearity problem.
Our empirical results were showed in Table 6. Estimated ordered probit model was statistically signi cant with a Likelihood Ratio Test Probability <0.0001, indicates joint signi cance of all coe cient estimates.
The in uence of caregiver types on children's unhealthy food choice was examined. Compared to children cared for by their parents, those cared for by the left-behind parent and relatives did not signi cantly differ in terms of being accepting of foods with safety risks. Surprisingly, children cared for by grandparents showed signi cantly lower acceptability for unhealthy foods than those cared for by their young parents (p<0.05). When the caregiver type changing from parents to grandparents, the percentage of children who mildly accept unhealthy foods increased 6.94%, while the percentage of children who moderately and severely accept unhealthy foods decreased 6.48% and 0.46%, respectively, which indicated that the factor of caregiver type could better in uence children with mild or moderate intention on unhealthy foods. Our results showed that contrary to the popular belief, children cared by grandparents did not show any worse performance at unhealthy food choice than those cared by parents, but even better.
Our results also con rmed the signi cant in uence of caregiver's parenting attitudes. Compared to doting attitudes, caregivers with quali ed parenting attitudes on healthy eating behavior signi cantly help children keep away from those unhealthy foods (p<0.01). The effect of caregivers' doting parenting attitudes and unknowledgeable parent attitudes on children's unhealthy food choice showed no signi cant difference. However, children raised by caregivers with neglectful parenting attitudes showed signi cantly worse performance than children of caregivers with other negative parenting attitudes. The results of marginal effect showed that when caregivers' parenting attitudes changing from doting attitude to neglectful attitude, the percentage of children who mildly accept unhealthy foods decreased 9.75%, while the percentage of children who moderately and severely accept unhealthy foods increased 8.77% and 0.99%, respectively. Our results indicated that it was caregivers' neglectful parenting attitude which signi cantly contributed to children's unhealthy food choice compared to other negative parenting attitudes.
Unsurprisingly, caregivers' acceptability toward unhealthy foods positively affected their children's acceptance. When their acceptability increased from mild to moderate, the percentage of children who mildly accept unhealthy foods decreased 11.39%, while the percentage of children who moderately and severely accept unhealthy foods increased 10.54% and 0.85%, respectively. In addition, our empirical results showed that the caregivers' sex, job type and family income also signi cantly affect children's choice. Children cared for by female caregivers tended to accept unhealthy foods more often than did those cared for by male caregivers (p<0.05). Children from higher-income families demonstrated a better capability to recognize foods with health issues than those from lower-income families. Compared to farmers, caregivers who once were rural migrant workers had a signi cant positive in uence on their children's performance. While caregivers running businesses seemed to be the worst role models, likely because they had limited time to spend with their children. Our results showed that caregivers' age and level of education were not signi cant factors. Children's sex and age also signi cantly affected their choice. Boys were more easily to accept unhealthy foods than girls and younger children preferred to unhealthy foods.

Discussion
This study aimed to provide contemporary insights into the protection of rural children's health from eating those cheap but unhealthy and unsafe foods through the contribution of rural caregivers. Children cared by grandparents are usually reported to have food perception bias and incline to access unhealthy foods compared to children cared by parents, because grandparents have lower level of education, older age, and tendency to indulge their grandchildren [30]. However, our results showed that caregivers' age and level of education were not signi cant factors in uencing children's intention. Surprisingly, rural parents did not necessarily perform better in terms of shaping better eating habits of their children than did other types of caregivers. Rather, rural grandparents even performed signi cantly better than the parents, which seems to be contrary to the common viewpoints that parents were the best food educators We provided a different insight to explain the failure of the current generation of rural parents in being good role models to their children. The average age of rural children in our survey is about 11 years old and the range of their parents' ages were between 35-45 years old. The generation of current parents has experienced a completely different childhood from their parents and even grandparents. From a historical perspective, when current rural parents were children about 40 years ago, the rapid development of the economy and limited regulatory resources in rural China promoted the explosive growth of unhealthy foods in the rural market [ 38 ], which indicates that the current rural parents were much more familiar with those unhealthy foods than their parents because they might start to eat them at their young age. On the other hand, parental migration began to appear at the same period in rural China, resulting in the absence of parental education and control toward healthy eating behaviors. Both of these two historical backgrounds contributed to the same consequence that the current parents become the rst generation who had unrestricted access to plenty of tasty and colorful foods but with substandard components or are unhygienically prepared, which never happened at their parents' childhood [ 39 ]. Thus current parents' inherent intention and preference for those unhealthy foods would be much higher than current grandparents, which could further negatively affect the performance of their education toward their children. Therefore, under the background of Chinese speci c historical period and culture, we propose a new evidence related to rural caregivers' in uence on children's unhealthy food choices, which was that the current rural parents' unhealthy eating perceptions and habits have started to exert intergenerational effects on their children's food choice. The main contribution of our ndings was that the current policies of encouraging rural children to grow up with their parents might help to solve most of the problems of being left-behind, however, they could not effectively protect rural children to keep away from the foods with safety and health risks.
We further investigated the in uence of caregivers' parenting attitudes and found that the children cared for by different caregiver clusters differed signi cantly in terms of their unsafe food intention and preference. The cluster of neglectful caregivers demonstrated the most passive parenting attitudes and their children performed much worse performance for unhealthy foods even than children of doting caregivers and unknowledgeable caregivers. This result implied that in rural areas, caregivers' doting attitude and lack of related knowledge due to a lower level of education, were not the key barriers for effective health education for children. Rather, the key barrier was their neglectful attitude. we propose segmenting the population based on their parenting attitudes toward health education and start from those caregivers with neglectful attitude as the key intervention target group.
The signi cance of sociodemographic factors showed similar results with those in previous studies [ 40 -41 ]. In general, children cared by female caregivers more likely choose unhealthy foods than children cared by male caregivers, which might because female caregivers were more indulgent and obedient to children. Rural children from the family with higher income condition performed better indicating that a family environment with a better economic condition may help children stay away from unsafe foods [ 42 ]. Caregivers used to be rural migrant workers showed better education effect perhaps because they have gained more food safety and health knowledge when living in the big city.

Conclusions
In summary, based on a large-scale survey on rural children and their caregivers in rural China, the ndings of this research highlighted the new evidence of how rural caregivers signi cantly affected their children's unhealthy food choices. Rural parents did not perform positive in uence compared to rural grandparents as conventionally believed. Our study provided new insights to explain this phenomenon under the background of Chinese speci c historical period, which was that the current rural parents were the rst generation who had unrestricted access to plenty of unhealthy foods in their childhood in China and their inherent intention and preference for those unhealthy foods would be much higher than current grandparents and might have performed a signi cantly negative in uence on their children. In addition, this study underscored the signi cance of caregivers' neglectful parenting attitudes on the impact of children's eating behavior. This nding suggests that launching comprehensive food safety and health education toward rural children to intervene their food choices at early time is urgent to minimize the intergenerational effects in the future. Our ndings also have policy implications for health promotion professionals who look for intervention targets of caregivers to improve their health education effectiveness for children. For the health education toward rural caregivers, instead of focusing on the caregiver types, segmenting the rural caregivers with neglectful parenting attitude would be the key step to identify the intervention targets.

Study Limitations And Future Research
There are several constraints in this study. First, the research did not investigate the living conditions of rural parents in their childhood, for example, whether they were left-behind children, which could help con rm our viewpoints. Future research is needed to develop the study and con rm our research conclusions. Second, this study focused only on the family in uence on rural children's intention and preference. However, for school-aged children, school education on food safety was also an important aspect. The next step in this research would be examining the effectiveness of both family and school education in rural children's intention, preference, and behavior change for unhealthy foods. Third, our study has not been examined by the intervention practice. We will launch the real health education intervention practice to evaluate their effectiveness in the future research.

Declarations
Ethics approval and consent to participate: The present study proposal has been approved by the institutional academic board of Renmin University of China. Since only a questionnaire survey was employed to obtain the data, the study was considered to belong to exempt category review for ethical approval due to minimal risk to respondents. Waiver from approval was granted by the institutional academic board of Renmin University of China. Since the participants in the study are both children and their legal guardians, addressing ethical considerations, at the beginning of the questionnaire a statement was clearly explained the purpose of the study and informed that all participants (including caregivers and their children) completing the questionnaire was optional and if caregivers completed and returned the questionnaires, we would assume that they approved us to use the information both they and their children provided. Therefore, informed consent to participate in the study has been obtained from all participants, including legal guardians of minors.

Consent for publication
Not applicable.

Availability of data and materials
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.