Subjects
The subjects were children under 6 years of age (0 to 71 months) and their caregivers from rural Hunan Province investigated between August and November 2019. The sample size was determined according to relevant equations for cross-sectional studies[19]. Since the stunting prevalence among rural children was estimated to be 16%[20], the size of a test α was 0.05, permissible error d was 0.10, the designed effect of complex samples was 2 and the non-response rate was 20%, and the final sample size was determined to be 5040 (= 2100×2× 1.2).
Subjects were selected by multistage stratified cluster sampling. One province of China consists of several cities, and one city consists of several districts and several counties (districts and counties are urban and rural respectively). One county contains several towns, and one town has several villages[18]. The 14 cities in Hunan were divided by the economic condition into three levels: high, moderate and low. Then 2 cities from each economic level, 2 counties from each selected city, 2 towns from each selected county, and 3 villages from each tested town were randomly selected. From each village, all eligible children and their caregivers were included into our subjects. Totally, 5529 children from 72 villages covering 24 towns in 12 counties were involved.
Data collection
This study consisted of a questionnaire survey and anthropometric measurements. The questionnaire included children’s factors (gender, age, birth weight, preterm birth, left-behind children, only child, passive smoking, regular physical examination), maternal gestational factors (age at delivery, gestational gain weight, moderate/severe anemia, pregnancy comorbidity), caregivers and family factors (type of caregivers, ethnicity, education level, occupation, family size, family income, family food expenditure). All children received anthropometric measurements, including length/height, and weight.
Definition of variables
The birth weight <2500 g, 2500-3999 g, and ≥ 4000 g were considered as low birth weight, normal birth weight, and macrosomia, respectively. Birth at <37 gestational weeks was regarded as premature birth. Left-behind children referred to those children whose parents (both or either) worked in other places and did not live together with them [18]. Only child was the only child born by one couple, and has no siblings. Passive smoking meant a nonsmoker inhaled at least 15 minutes every day the smoke exhaled by smokers for at least 1 day within 1 week. Maternal gestational weight gain was determined by the final weight of the mother measured at late pregnancy before delivery subtracted by the weight at early pregnancy[21] and was divided into four groups of <10, 10-14.99, 15-19.99, and ≥20 kg. Maternal moderate/severe anemia was defined as a hemoglobin level <100 g/L[22]. The maternal hemoglobin concentration in this study was the concentration in the third trimester of pregnancy, which was obtained based on the participants’ recall for their hemoglobin detection during pregnancy. The maternal pregnancy comorbidities included gestational diabetes mellitus, gestational hypertension, pregnancy associated with cardiac diseases, gestational liver diseases, and thyroid dysfunction. Caregivers were those who took care of the diets, living and personal security for children and were divided into two types: parents, and grandparents/others. Ethnicity of caregivers was divided into Han and minorities. The education level of caregivers was classified into primary school or below, junior high school, senior high school, college or above. The occupation of caregivers was divided into housework, government agencies staff, business service staff, farmer, others. Family size was defined as the total number of family members and involved the members with economic relations and joint budget and diets, and was separated into ≤4, 5-6, and ≥7.
Anthropometric measurements
The investigators used unified instruments to measure the length/heights and weights of children according to standardized methods, which were described by the Technical Specification for Children Health Check Service (China Ministry of Health, 2012). The lengths and weights of children aged 0-23 months were measured by using an FSG-25-YE lying-form infants and young children precision medical examination meter (Shanghai Betterren Medical Tech Co., Ltd.). The heights and weights of children aged 24-71 months were measured using an HX-200 stadiometer and an HCS-50-RT electronic scale respectively (Liheng Instrumentation LTD., Shanghai, China). The accuracies of instruments for length/heights and weights were 0.1 cm and 0.05 kg, respectively.
Evaluation criteria for children physical development
The commonly-used indices for children physical development are length/height for age, weight for age, weight for length/height, and body mass index (BMI) for age. BMI was calculated using the ratio between children's weight in kilograms and length/height in meters squared (kg/m2): BMI = weight (kg)/ height2 (m2). The children physical development was evaluated using Z-score recommended by WHO: Z score = (analyzed index - median of reference standard)/standard deviation of reference standard. The WHO Child Growth Standards involve two age groups: 0-5 years (0-60 months) and 5-19 years (61-228 months), which are 2006 Child Growth Standard[23] and 2007 Child Growth Standard[24]. Hence, the physical development of children was evaluated according to the two age groups above.
- Children aged 0-60 months: length/height for age z score (HAZ), weight for age z score (WAZ) and weight for length/height z score (WHZ) were calculated according to WHO 2006 Child Growth Standard. HAZ <-2 was defined as stunting, WAZ <-2 was defined as underweight, and WHZ <-2 was defined as wasting.
- Children aged 61-71 months: HAZ, WAZ, and BMI for age z score (BMIZ) were calculated according to WHO 2007 Child Growth Standard. HAZ <-2 was defined as stunting, WAZ <-2 was defined as underweight, and BMIZ <-2 was defined as wasting.
Quality control
The investigators were the child health care doctors selected from the county-level maternal and child health care hospitals of the corresponding counties. Prior to the survey, all the investigators were trained unifiedly, and only the qualified ones were allowed to take part in on-site survey. The instruments were calibrated before and during investigations. The physique measuring staff measured the length/height and weight of children in strict accordance with the specifications of the instruments. During the survey, all copies of the questionnaire were checked by quality control personnel: each copy should be filled in in a complete and standard way. Any illogical or missed response should be corrected in time. Data were double-inputted on Epidata 3.1 and tested in terms of consistency. For any inconsistent data, the original copy should be checked to ensure the high quality of any inputted data.
Statistical analysis
HAZ, WAZ, WHZ and BMIZ were computed using WHO anthropometric macros in SPSS (igrowup_SPSS and WHO2007_SPSS) [25, 26], and statistical analyses were conducted on SPSS 25.0 (IBM, Chicago, IL, USA). The statistical description of categorical data was used proportion or rate. The stunting, underweight and wasting prevalence of children with different characteristics were compared by chi-square test. The significant variables identified from univariate analyses were involved into multivariate logistic regression analyses of stunting, underweight, and wasting. The independent variables were selected according to stepwise regression (forward: LR). All statistical tests were two-tailed, and the significant level was P<0.05.
Ethics approval and consent to participate
The study protocol was approved by the Ethics Committee of Hunan Provincial Maternal and Child Health Care Hospital (No.2019-S036). The study was conducted in accordance with the Declaration of Helsinki. Written informed consents were obtained from all the caregivers of children involved in this study.