Study participants
Data used in this study were from the Chinese Nutrition and Health Surveillance (2010–2012). The CNNHS was a nationally representative cross-sectional study conducted by the National Institute for Nutrition and Health, Chinese Center for Disease Control and Prevention (NINH, China CDC) to assess the health and nutrition status of Chinese residents. The 2010-2012 survey covered all 31 provinces, autonomous regions, and municipalities directly under the central government throughout China (except for Taiwan, Hong Kong, and Macao). The method of multi-stage stratification and population proportional cluster random sampling was adopted. In accordance with economic development, all county-level administrative units were divided into four categories: big cities, small and medium-sized cities, ordinary rural areas and poor rural areas. A total of 150 counties (districts) were selected from four categories of areas as study sites,including 34 big cities, 41small and medium-sized cities, 45 ordinary rural areas and 30 poor rural areas, respectively. The study selected 25 households randomly from each village/community and children aged 6 to 17 years in each family were involved. If the number of children was less than 20 children (10 boys and 10 girls) in each age group in each site, supplementary children would be selected from nearby primary and secondary schools to meet the minimum sample size. The specific sampling method can be referred in the published literature [25]. After excluding missing and abnormal data in height, weight, frequency of eating out and sedentary time, a total of 15261 participants (7685 boys, 7576 girls) aged 6 to 17years were recruited. This study was approved by the ethics review committee of the National Institute for Nutrition and Food Safety, Chinese Center for Disease Control and Prevention (No. 2013-018).All participants’ parents or legal guardian were fully informed the purpose and procedures of the study before enrolling and signed the informed consent.
Data collection
The interview-administered questionnaire was used to collect the information of basic socio-economics, sedentary duration and leisure exercise. The food frequency questionnaire (FFQ) was used to collect eating behaviors over the past week . Participants were asked “How many times have you had breakfast, lunch, or dinner in the past week (seven days)?” and subsequently “How many times in the past week have you had breakfast, lunch, or dinner at a restaurant or school canteen?”. All the questionnaires were collected by trained investigator by face-to-face interview, and the interviews were conducted at the homes of the participants. The children younger than 12 years old finished the questionnaire with the help of their parents or legal guardian. To ensure the reliability of data , evaluation indexes and quality control measures were made at the national, provincial and district levels, quality control was carried out in the early stage, later stage and period of field survey.
Categories of Frequency of Eating Out
In the current study, eating out was defined as eating a meal or more prepared by restaurants. Frequency of eating out was categorized as three levels: 0 time, 1-2 times, or 3 times and over per week.
Anthropometric Measurements
The height was measured with an accuracy of 0.1 cm and the fasting weight was measured with an accuracy of 0.1 kg. All measurements were conducted by well-trained investigators under standard operation procedure. Body mass index (BMI) was calculated as a division of weight in kilograms by the square of height in meters.
Definition of Weight Status
Overweight and obesity was classified based on age- and gender-specific BMI cutoff points among Chinese children[26]. In the meantime, national health standard of Screening standard for malnutrition of school-age children and adolescents was devoted to screen for underweight children[27].
Assessment of Sociodemographic Determinants
Age groups were divided into two categories (6-12, 13-17 years). Participants were divided into four residency groups according to location (urban, suburban, rural and poor rural). Family income was classified into four levels (<20,000 RMB/person, 20,000–40,000 RMB/person, >40,000 RMB/person, and unkown). Leisure exercise was defined as two levels (No/Yes). Sedentary duration status was grouped into two levels (≤2h/d and >2h/d).
Statistical Analysis
All statistical analyses were conducted using SPSS 22.0. The univariate descriptive statistics were conducted for each variable (frequency and percentage), and the frequency of eating out was expressed as a percentage. t tests were used to compare means between different gender, age,leisure exercise and leisure SB groups. Data were run to compare means of region, family-income and BMI status using Chi-square tests. Multivariate logistic regression analysis was used to analyze the relationship between eating out and overweight and obesity. After adjusting for age, region, logistic regression was employed to examine the association of eating out with overweight and obesity, where dependent variable was BMI status (obesity, overweight, normal weight and underweight), and adjusted odds ratios (OR) with 95% confidence intervals (CI) were obtained. Statistical significance was considered at p< 0.05.