Caregiver perceptions of child diet status: What inuenced their judgement

Background: Caregiver perceptions of child diet status can inuence the effect of health intervention, but research focused on the correctness of caregiver's perception is limited. This study aimed to gure out the correctness of caregiver's perception of their child's diet status and nd out the factors which may inuence their judgement. Methods: 815 children from grade 3 rd to 6 th and their caregivers were recruited from two primary schools by cluster sampling. 3-day 24-hour recall was used to record children’s diet, caregiver’s perceptions was obtained through questionnaires. Multivariate logistic regression models were used to explore the inuencing factors of the correctness of caregiver’s perception. Results: Among the 597 children with “balanced” diet, 62.1% were correctly perceived as having a “healthy” diet by their caregivers; but for the 218 children with “imbalanced” diet, just 16.1% were correctly identied. Those who were correctly identied as having an unhealthy diet consumed less fruits and more snacks than those unidentied (p<0.05). Obese children were more likely to be perceived as having unhealthy diet even when they actually had a balanced diet (OR=0.305, p=0.016). Caregivers with high educational level were more likely to correctly perceive children’s healthy diet status (OR=3.628, p=0.038). Conclusions: Caregivers were generally lack of the ability to identify children's unhealthy diet. Obesity, signicantly low consumption of fruits or high consumption of snacks can raise caregivers’ awareness. Extensive health education targeting caregivers is in emergency. in the current study, we aimed to gure out the correctness of caregiver’s perception of their child’s diet status and to nd out the factors which may inuence their judgement.


Background
Among the 20 most populous countries, China ranked rst in terms of the number of obese children in 2015 [1] . 10.5% of the boys and 7.1% of the girls were obesity in China [2] , with the continuous development of the economy, this ratio will keep rising. But at the same time, 2.4% and 4.1% of the Chinese children were stunted or wasted, respectively [3] . The problem of micronutrient de ciency still exist, micronutrients including vitamin D, vitamin A, vitamin B 12 , Fe and Zn were de cient to varying extents [4][5][6] . Child malnutrition is widely exists in countries all over the world, in 2016, 22.9% of children around the world were stunted, 2.5% were wasted, and 6.0% were overweight [7] .
Diet quality, as a potentially modi able factor in maintaining people's health, plays an important role in weight control [8,9] . Diet with less energydense nutrient-poor (EDNP) food and adequate fruits and vegetables can not only help people stay in shape [10][11][12] , but also reduce the risk of negative health outcomes and all-cause mortality [13][14][15][16][17] . Previous studies have revealed that more than 60% of the Chinese school-age children failed to meet the recommendations issued by Chinese Dietary Guidelines on their fruits and vegetables intake [18] , which is undoubtedly detrimental to the development of children. And since child's diet habits might be carried into adulthood and cause long-term health outcomes [19,20] , early detection of children's imbalanced diet and take actions in time is of great importance.
In order to cultivate healthy eating habits at an early stage, researchers and governments have started to organize nutrition education activities targeting school-age children. But in fact, children(especially young children) have little autonomy in their food choices, almost all foods are provided by their caregivers. Thus, caregiver's attitudes are crucial to the effect of interventions [21,22] , if they are unwilling to help their children to change their unhealthy habits, the effect of health education will be small. Previous researches found parents with correct conception on their child's overweight status are more likely to make changes to their children's lifestyles and participate in healthy lifestyle behaviors with their children, while those who failed to recognize their child's weight problem are less motivated to address the problem [23,24] . So we have the reason to judge that caregivers with correct perception on their children's unhealthy diet can also be more willing to participate in the nutrition promotion activities with their children, which could greatly improve the effectiveness of health interventions.
However, to date, limited research focused on caregiver's perception of child's diet. So in the current study, we aimed to gure out the correctness of caregiver's perception of their child's diet status and to nd out the factors which may in uence their judgement.

Participants
Participants were recruited from two primary schools by cluster sampling. Considering that children's diet need to be reported by themselves and the cognitive ability of students in lower grades are limited, so in the current study, we selected children of grade 3rd to 6th and their caregivers as objects. Written consent from both children and caregivers were required in order for both them to participate. The study was approved by Ethics Committee of Wuhan University of Science and Technology (No. 201519).

Demographic Characteristics
After consulting experts, self-designed questionnaire on basic information of children (including age, gender, number of brothers/sisters, and the frequency and location of meals, etc.) and caregivers (including age, gender, relationship with children, annual household income, and caregiver's educational level, etc.) was used to collect the demographic characteristics. A group of research team members, including faculty and postgraduates of the school of public health, helped students to ll the questions about their basic information. Caregivers' questionnaires were brought home by the students, the primary caregiver of the child was asked to ll the questionnaire, then they will be brought back to school by the students the next day. The Cronbach's α coe cient of this questionnaire was 0.776.

Diet Survey And Evaluation
3-day 24-hour recall was used to conduct diet survey. For three consecutive days, the investigators went into the classroom after lunch, using the food size reference models, they asked and recorded all the food consumed by each student in the past 24 hours (including drinks, snacks, interclass meals, etc).
Chinese Children Dietary Index (CCDI) was put forward by Guo Cheng et al [18] , it was developed based on the Chinese dietary intake recommendations (Chinese Dietary Guidelines and Chinese DRIs) and health-promoting behaviors. The effectiveness of CCDI in evaluating the dietary status of Chinese school-age children have been proved, detailed descriptions of the CCDI have been published elsewhere [18] . Since we don't have the water intake and sedentary behavior data, CCDI was slightly modi ed to evaluate the dietary status of children in the current study. The total score is 150 points and if the child got scores over 90 points, he/she will be considered to have a healthy diet, otherwise, his/her diet will be regarded as unhealthy.

Caregiver Perception
Caregiver's perception of their child's diet was assessed using the question 'How would you describe your child's diet status?' and they were given three choices: healthy, unhealthy, and unknown (reason is needed). The accuracy of the perception was assessed by comparing caregiver's perception and child's diet status, caregivers who have mismatch perception with their child's diet were deemed to have incorrectly perceived their child's diet status.

Covariaties
Migrant is classi ed based on the registered place reported by the caregivers, if the child's registered place is in Wuhan, he/she will be regarded as local child, otherwise, he/she will be regarded as migrant child. Primary caregiver is the person who takes care of the child (prepares food for the child) most often. The caregiver's educational level was de ned as the highest degree that the primary caregiver completed at the time of the survey.
Family income refers to the average annual household income, including but not limited to wages, self-employed income and agricultural income.
Children's weight and height were obtained from the data of physical examination made by Wuhan ChangeDong Hospital, height was measured with a precision of 0.1 cm and weight was measured with a precision of 0.1 kg. Children's Body Mass Index (BMI) was calculated, obesity was de ned according to the BMI cutoffs points issued by the National Health Commission of the People's Republic of China [25] .

Statistical analysis
Descriptive statistics included the frequency and percentages for categorical variables and means (SD) for continuous variables. T tests for continuous variables and chi-square tests for categorical variables were used. Univariate and multivariate logistic regression models were used to explore the in uencing factors of the correctness of caregiver's perception. All analyses in this study were performed in Stata (version 13.0; StataCorp, College Station, TX, USA). Differences were considered signi cant if p < 0.05.

Results
Complete data was obtained from 815 child-caregiver pairs, Fig. 1 provides an overview of the recruitment. More than half (57.9%) of the children were migrant children, 50.7% of the families had an average annual income lower than ¥50,000 (Table 1). Most caregivers weren't well-educated, 51.8% of them just got middle school degree or lower. 597 (27.4%) children had imbalanced overall diet, boys and overweight/obese children seemed to be more likely to have an unbalanced diet (p < 0.05)( Table 1).

Caregiver Perception
Of the 815 caregivers, 209(25.6%) couldn't make a clear judgment on their child's diet for they didn't know the de nition of "healthy diet" or didn't pay attention to their children's diet. Compared to the caregivers with clear answer, these caregivers had no difference in educational level, family income and their child's gender, weight status, and diet scores (p > 0.05).

Children's Daily Food Intake
Children's dietary reference values issued by Chinese Dietary Guidelines varied by age and sex [26] , so food intakes were converted to food density(g/1,000 kcal) for the purpose of comparison (except for snacks and beverages), details available from supplementary table. Table 2 shows that children with imbalanced diet had lower intakes of vegetables, fruits, sh, eggs, beans, and milk (p < 0.01), and higher intakes of grain, snacks, and beverages (p < 0.01). In the "imbalanced" group, those who were correctly identi ed as having an unhealthy diet by their caregivers consumed less fruits than those unidenti ed, and they are more likely to consume snacks and beverages than the others (p < 0.05). b T-test for food intake between correct group and incorrect group ( * p < 0.05).
c T-test for food intake between balanced diet group and imbalanced diet group ( △ p < 0.05).

In uence Factors Of Caregiver's Perception On Diet
Migration, child sex, caregiver's association with the child, family income and breakfast habit didn't affect caregiver's recognition on their child's unhealthy diet. Caregivers were more likely to identify unhealthy dietary status among children who consumed less fruits (OR = 0.989, p = 0.031) or those eat more snacks (OR = 1.074, p = 0.004). And they are more likely to correctly identify unhealthy diets (OR = 3.532, p = 0.040) in obese children, but those children were also more likely to be perceived as having unhealthy diet even when they actually had a balanced diet (OR = 0.305, p = 0.016). Moreover, caregivers with high educational level were more likely to correctly perceive children's healthy diet status (OR = 3.628, p = 0.038) ( Table 3) Table 3 Regression models between caregiver's perception status and other socio-demographic predictors, and intake of some kinds of food a

Discussion
The purpose of the current study was to investigate caregiver's conception of their child's diet status and explore the potential factors that may in uence their judgement. To our knowledge, this is one of the rst studies to examine caregiver's perception of child diet status, our results showed that less than a fth of children with "imbalanced" diet were correctly distinguished as having an "unhealthy" diet by their caregivers. Most caregivers cannot correctly identify the unhealthy dietary conditions of their children, the judgment is in uenced by children's consumption of certain types of food or their body type, rather than their overall diet, which is not optimistic.
Children in the "imbalanced" group had lower intakes of several kinds of food except grain and meat, and their snacks and beverages intakes were signi cantly higher, but only extremely low intakes of fruits or high intakes of snacks could call up caregivers' awareness, they didn't pay much attention to the other types of food (including vegetables, sh, beans, eggs, and milk). Chinese children's consumption of meat and sh has been rising in the past decade, meat consumption has basically met the amount recommended by the Chinese Dietary Guidelines, but their sh intakes were still at a low level [27] . The traditional view is that the nutrients in meat, sh and eggs are about the same, among which the red meat was the most nutritious, so most caregivers tend to prepare enough red meat for their children, ignoring their intake of sh and eggs. Most caregivers knew the bene ts of milk and ordered school milk for their children (one box per day distributed during the school days), but some children didn't drink them for different reasons, which may led to caregivers' overestimation of children's milk intake. Caregivers generally know that fruits and vegetables are good for health while eating too many snacks is harmful, and they order school milk for their children initiatively, which showed that most caregivers had the most basic nutritional knowledge, but their knowledge is incomplete, or even incorrect, nutritional education for caregivers needs to be strengthened.
Generally, highly educated caregivers have more access to correct health information and are more likely to make accurate judgements about children's diet. In this study, parents with college degree or above were indeed made more accurate judgement in the "balanced" group, but not in the "imbalanced" group. A study targeted parental perception on child's weight status indicated that what in uenced the accuracy of parent's physique estimation was not parental education, but parental health literacy [28] . Children's diet is associated with caregiver's nutrition knowledge and their attention to the child's diet, so one of the possible reason is that caregivers in the "imbalanced" group are not initiative to the nutrition knowledge, they rarely pay attention to these messages, even if exposed to them. As a result, these caregiver's health literary didn't increase with their educational degree. Another possibly reason is that caregivers of children in the "balanced" group paid more attention to their children's diet behaviors and fully used their knowledge when making the judgement. But for the other group, caregivers may paid less attention and knew little about children's diet status, their knowledge wasn't fully used. Future health education should not neglect highly educated parents.
Whether in the "balanced" group or "imbalanced" group, obese children were more likely to be considered as having an unhealthy diet. Caregivers knew that improper diet can lead to obesity, but they may ignored the in uence of genetics, sports, mental health and other factors, which further proved that they are lack of the correct health knowledge.
Regardless of immigration, educational level and family income, unable to identify unhealthy diets seems to be a common phenomenon. Some people indicated that parents' judge on their child's actual weight is not a cognitive task, but an emotional evaluation [29] , this may also suitable for diet judgement: caregivers tend to believe their children are healthy and normal when they are not sure whether their children's diet is up to the health standard. This tendency may further lead caregivers to refuse to participate in health promotion courses, for people believe that only unhealthy/abnormal children need to participate in such classes, its necessary for us to teach them how to judge diet in a right way.

Strengths and limitations
There are several limitations in this study. Firstly, the participants were recruited from 2 local schools, which is not nationally representative, the results of this study cannot be generalized to other samples. Secondly, In order to increase the compliance of the subjects, the 3-day diet data are collected in the school during the weekday, the data may not represent children's diet on the weekends. Third, since there is no widely accepted indices to measure the diet quality of Chinese children, CCDI is used to assess children's diet status, although not widely used, the effectiveness of CCDI has been proved in other study on children of the same age [18] . Despite these limitations, the study is one of the rst studies to examine parent's perception of child diet status and its in uencing factors, the results can help realize the inadequate health awareness among parents and provide the direction for health intervention.

Conclusions
Caregivers in this study were generally lack of the ability to identify children's unhealthy diet, especially those in "imbalanced" group. Obesity, signi cantly low consumption of fruits or high consumption of snacks can raise caregivers' awareness. But if the child is normal weight or overweight, and they didn't show any special rejection/preference for fruits and snacks, their imbalanced diet are less likely to be noticed. This situation is worrisome, if the caregivers can't notice their child's diet problems, they might be less motivated to change their diet behaviors, increasing the risk of obesity and chronic diseases in the future. Extensive health education targeting caregivers is in emergency.

List Of Abbreviations
BMI-Body Mass Index; CCDI-Chinese Children Dietary Index; DRIs-Dietary Reference Intakes.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. Supplementarytable.docx