Peer-reviewed article

Cohort Profile: The Shanghai Children’s Health, Education and Lifestyle Evaluation, Preschool (SCHEDULE-P) study

Xuelai Wang, Yunting Zhang, Jin Hua Zhao, Wenjie Shan, Zichen Zhang et al. · 18 Dec 2020

Abstract

Early childhood development (ECD) is defined as the period from conception up to school entry. It is a unique opportunity for children’s cognitive, social, emotional and physical development.1 Neuroscientific evidence has proved that both adverse and positive experiences would affect brain development, influencing subsequent health and development.2,3 Thus, beneficial effect of intervention of this period has been shown in adult health, 4 competence, 5 reduction of violence and crime 6 and development of the subsequent generation.7 According to the global estimation, 43% of children in low-income and middle-income countries (LMICs) are at risk of poor development, 8 and they are subject to an average of 26% loss of adult income.9 Given the importance of ECD, it was included in Sustainable Development Goals (SDGs) to ensure that all children have access to quality ECD, care and pre-primary education so that they are ready for primary education.10 Although previous cohort studies have explored how early circumstances can influence later development, the risk factors and current levels of ECD have not been assessed systematically. The Millennium Cohort study, started in 2000, followed the lives of around 19 000 children born in the UK, which has found that circumstances such as child poverty and family issues influence later health and development and thus has informed the development of early years policy and practice and service provision in disadvantaged children.11 However, this cohort mostly focused on physical health and cognitive abilities in early years.12 In 2017, built on empirical evidence of ECD, a consensus was reached that multiple factors could influence the acquisition of competencies, and that children’s early development requires health, nutrition, security and safety, responsive caregiving and early learning, collectively referred to as nurturing care.13 In 2018, the World Health Organization (WHO), United Nations International Children's Emergency Fund (UNICEF) and World Bank put forward ‘nurturing care’ in earliest years as the essential foundation for child development, which is also a framework for helping parents and caregivers to promote children’s development.1 To track the progress of SDG and identify the gaps on ECD, the Countdown to 2030 has collected data on five domains of nurturing care covering 138 LMICs.14 However, data availability is far from satisfactory. There has not been a cohort designed under nurturing care framework, to research its current status and effects on ECD. Population-based measurement of ECD is another challenge, since there are no well-defined tools.15 Early Child Development Index (ECDI), a 20-question scale developed by UNICEF, was designed to measure the developmental status of children aged 24–59 months.16 The former ECDI was used in the UNICEF's Multiple Indicator Cluster Surveys (MICS) to assess basic developmental milestones of children from 94 countries.17 However, not all items on the ECDI are age-appropriate, which may overestimate children’s developmental competence.18 In addition, MICS lack data from China on ECD status. China has the second largest number of children under 14 years of age in the world. Thus, investigation on ECD status in China is crucial to promote ECD measurement and intervention worldwide. This ultimately led to the initiation of Shanghai Children’s Health, Education and Lifestyle Evaluation, Preschool (SCHEDULE-P) study. It aimed to investigate the effects of nurturing care on ECD status of children aged 3–6 years from kindergartens in Shanghai and to explore the scientific evidence for policy making and implementation for nurturing care. We focused on the preschool stage, since it is a critical transient period from parenting care to school education. As one of the largest and most developed cities in China, Shanghai attracts a large number of people from other domestic cities, including migrant workers who do not have urban hukou (household registration) or Shanghai Residence Cards.19 Children of migrant workers are considered vulnerable due to the lack of opportunities for accessing early education and to poor living environment.20 Therefore, SCHEDULE-P study could provide enriched data on both risk and protective factors and population-level ECD status in children from families with differing socioeconomic status, which will serve as a robust basis for policy making and implementation of nurturing care. The SCHEDULE-P study was designed and implemented by the Child Health Advocacy Institute, Shanghai Children’s Medical Center, Shanghai Jiao Tong University School of Medicine. This institute is a subordinate component unit of the Ministry of Education—Shanghai Key Laboratory of Children’s Environment Health, which undertakes extensive cohort surveys on ECD assessment and intervention, child health policy, network and monitoring for child health, environmental factors in child growth and development, and birth cohort. The study was approved by the Institutional Review Board of Shanghai Children’s Medical Center Affiliated Shanghai Jiao Tong University School of Medicine (SCMCIRB-K2016022-01). All 3-year-old children living in Shanghai are required to enter kindergarten and enrol in the registration system of the Shanghai Municipal Education Commission (SMEC). According to this database, a total of 167 597 enrolled children entered 1584 kindergartens in 2016, excluding international kindergartens. We selected a representative sample from newly enrolled children by a stratified cluster random sampling design. First, all 16 districts in Shanghai were set as primary sampling units (PSUs). We then defined secondary sampling units (SSUs) based on the kindergarten’s ownership and level. In Shanghai, the kindergartens are divided into two categories, public or private kindergarten, and they are rated at different levels depending on the teaching quality, child development, health care, facility and equipment, kindergarten management and others, with the demonstration level as the highest and the third level as the lowest. There are a total of nine categories of SSUs (Figure 1). In each SSU, one or two kindergartens were randomly selected, and all the enrolled children in the junior grade were invited to this survey. Finally, 20 899 children aged 3–4 years from 191 kindergartens were randomly selected as subjects. The stratified sample units of kindergartens of different levels in Shanghai The stratified sample units of kindergartens of different levels in Shanghai In addition, we designed a field survey to assess cognitive function of children in the senior grade. We carried out stratified cluster random sampling and first set districts as PSUs. The Pudong district was purposely selected because its population accounts for half of the total population in Shanghai. We then randomly selected three districts from the central urban area, suburbs and outer suburbs, which were the Changning, Baoshan and Songjiang districts, respectively. In each PSU, SSUs were divided into five categories including: (i) public demonstration level and first level; (ii) private demonstration level and first level; (iii) public second level; (iv) private second level; and (v) private third level. Because the first category makes up around half of the population, two kindergartens of this category were randomly selected, and one kindergarten of the other four categories were randomly selected. The kindergarten of private third level in the Baoshan district which we selected in 2016 was closed down, so we replaced it with two kindergartens of the same level in the Jiading district. In each kindergarten, two complete classes were randomly chosen. In private third level kindergartens, all children were selected to ensure enough sample size for children of migrant workers. Finally, 20 kindergartens from 191 kindergartens were randomly recruited, including 1082 senior grade children. Sampling weights were computed using inverse probability weighting, which represented the inverse of the combined selection probability in each sampling stage, including the non-response rate. A combined design was used with a three-wave online questionnaire survey from junior grade to senior grade and one field survey in the senior grade. To ensure that the large-scale online survey would be implemented smoothly, we performed a pilot survey aimed at 19 757 enrolled children in 2015, which was carried out over 3–15 June 2016. Based on this pilot survey, we modified the questionnaire via expert consultation and optimized the implementation protocol under strict quality control. At the beginning of each wave, we and SMEC hosted an orientation meeting and assembled contacts of the district education commission and kindergarten representatives to participate. At this meeting, we introduced the purpose of SCHEDULE-P study, experience from previous waves, technical procedures and timetable of survey. We also provided a standard training for kindergarten representatives to ensure they carried out this survey following the uniform process. The kindergarten representatives were responsible for checking the children’s personal information and guiding parents to finish an online survey within 2 weeks. Children’s personal information was first obtained from the SMEC database. After the information was double-checked by kindergarten teachers, a unique login code was generated for each child and sent to the parents by teachers. Using the login code, all parents accessed an online questionnaire to check the child’s personal information (including name, gender, the birthdate and kindergarten) and gave informed consent to participate in this survey. The online survey has been performed in three waves since 2016. The first wave was carried out 10–24 November 2016, followed by the second wave 12–26 April 2018 and the third wave 22 April–5 May 2019. Furthermore, the field survey was performed between 29 April and 6 June 2019 (Figure 2). The three-wave recruitment processes. The reasons of no participant include transfer to another kindergarten, kindergarten closed down, long-term sick leave, and others The three-wave recruitment processes. The reasons of no participant include transfer to another kindergarten, kindergarten closed down, long-term sick leave, and others The online survey consisted of self-designed questionnaires and international standardized questionnaires, covering risk factors of ECD and nurturing care outcomes. The risk factors included general information, five components of nurturing care framework, and lifestyle. Nurturing care outcomes mainly focused on ECD status, social-emotion and cognition. The field survey assessed cognitive function development, such as intelligence, social-emotion, attention and language by neuropsychological tools. Table 1 presents the information collected online and in field surveys. Information collected from the online survey and the field survey in the SCHEDULE-P study Information collected from the online survey and the field survey in the SCHEDULE-P study A young child’s health is associated with the caregiver’s physical and mental well-being.21 Therefore, we investigated environmental exposures (during pregnancy), birth conditions (i.e. birthweight, gestational age, admission to neonatal intensive care unit), immunization, physical health, and mental health of the caregiver/mother. In this component, we focused on breastfeeding. We also collected the self-reported height and weight of children. This means observing and responding to a child’s movements, sounds and requests, as well as other social interactions. We chose the family structure and primary caregiver as risk factors and used the Index of Child Care Environment (ICCE) questionnaire to assess a child’s home environment. ICCE measures child care environment through 13 questions with a total score and four subscales: Human Stimulation, Social Stimulation, Avoidance of Restriction and Social Support.22 To assess parent-child interaction, we used the Chinese Parenting Child Interaction Scale (CPCIS), a useful assessment tool with good validity and reliability.23 CPCIS has 12 items to examine the frequency of Recreation Activity, Learning Activity, Reading Activity and Interaction with Environment. Younger children are most vulnerable to environmental risks, physical pain and danger. Poverty and low income are indicators of childhood adversity. We researched socioeconomic statuses, such as family income, maternal education, paternal education, parents’ marital status, and maltreatment, including physical abuse, neglect, emotional maltreatment and sexual abuse. During the preschool period, a child’s lifestyle is gradually cultivated, which will not only affect the current situation of ECD but also form the basis for the lifestyle at school age.24 In addition to the nurturing care framework, we paid particular attention to sleep and screen exposure, which have been reported as closely related to child development.25,26 The Children’s Sleep Habits Questionnaire (CSHQ) is a useful sleep screening instrument for measuring sleep patterns and sleep problems in children aged 4–10 years.27 The reliability, content validity and construct validity of the Chinese version of CSHQ are acceptable.28 CSHQ includes eight domains: bedtime resistance, sleep-onset delay, sleep duration, sleep anxiety, night wakings, sleep-disordered breathing, parasomnias and daytime sleepiness. Furthermore, high screen exposure can impair a child’s cognitive development and social communication skills. A self-designed questionnaire has been used to assess a child’s daily time consumption on a smartphone, PAD, television or other devices and gadgets with different contents, including educational programmes, cartoons, video games and other internet media [e.g. ‘Over the last month, how long did your child spend on screen media (TV, computer, phone, tablet etc.) each day?’].29 We also asked if the caregiver has interaction with the child when the child is exposed to screen (e.g. ‘When your child is watching TV, do you or other caregivers accompany with your child?’). School readiness is considered to be an important indicator of ECD due to its effectiveness as a predictor of children’s future achievement.30–32 In this study, school readiness was measured with the early Human Capability Index (eHCI).33 The eHCI was designed to assess the development of children aged 3–5 years as a population-level instrument across diverse cultures. It includes a total of 58 questions covering nine domains of child development (Physical Health, Verbal Communication, Cultural Knowledge, Social and Emotional Skills, Perseverance, Approaches to Learning, Numeracy, Reading and Writing) and allows for a holistic approach to child development.34,35 The Chinese version of the eHCI has been locally adapted to capture the unique cultural aspects of China.36 For each domain, the score is a continuous variable ranging from 0 to 1, with 1 being the best and 0 the worst. An overall development score is a mean value of nine domains and is regarded as an indicator of early child development. The Strengths and Difficulties Questionnaire (SDQ) is a brief behavioural screening questionnaire in social psychology problems (i.e. emotional symptoms, conduct problems, hyperactivity-inattention, peer problems) and personal strengths (i.e. prosocial behaviour) for the past 6 months.37 It has been validated in China and is widely used in many studies.29,38,39 A high SDQ score indicates high psychiatric risk, whereas a prosocial behaviour score is positively correlated with prosocial behaviour. A field survey including neuropsychological assessments and laboratory assessments was conducted to investigate cognitive function outcomes in the senior grade. We carried out a Combined Raven’s Test (CRT),40 Test of Emotion Comprehension (TEC),41 MOXO Continuous Performance Test (MOXO-CPT)42 and Diagnostic Receptive and Expressive Assessment of Mandarin-Comprehensive (Dream-C)43 to evaluate the child’s intelligence, social-emotion, attention and language development. Laboratory measurements included hair cortisol levels and finger-length ratio to evaluate the child’s stress44 and sexually dimorphic features.45 We recruited student volunteers whose field was medicine or psychology from medical schools or normal universities and trained them to operate a cognitive assessment in a standardized way. In the first wave, parents of 20 324 children consented to participate in the survey, with a response rate of 97.25%. The average age of children was 3.72 ± 0.30 years. Parents of 18 316 children consented to participate in the second wave, with a follow-up rate of 90.12%. The average age of children was 5.15 ± 0.29 years. In the last wave, parents of 17 233 children consented to continue the survey, with a follow-up rate of 84.79%. The children averaged 6.17 ± 0.29 years of age. Table 2 shows the baseline demographics and socioeconomic status of target children and the distribution of kindergartens of different levels. Baseline characteristics of the subjects in the SCHEDULE-P study in 2016 RMB, Renminbi, the currency of China. Baseline characteristics of the subjects in the SCHEDULE-P study in 2016 RMB, Renminbi, the currency of China. Zhao et al. examined the media use among preschool children, 29 which indicated that preschool children were exposed to the screen for 2.8 [95% confidence interval (CI) 2.7, 2.9] h per day, with 78.6% (95% CI 77.8, 79.3) of children spending more than 1 h on media per day, exceeding the limit of screen use from the American Academy of Pediatrics guideline of 1 h/day.46 Children who had excessive screen exposure were at increased risk of psychosocial problems. Moreover, parent-child interaction was a main mediator accounting for the effect of screen time on psychosocial well-being.29 Shan et al. explored maltreatment among children aged 3–4 years in Shanghai.39 The prevalence of parent-reported child maltreatment was 2.70% (95% CI 2.38, 3.05). Maltreated children had an increased risk of developing psychosocial problems, particularly those with a low frequency of parent-child interactions. The main strength of the SCHEDULE-P study is that it is a large-scale and representative cohort to depict the development trajectories of kindergarten children in Shanghai, with a baseline and two annual follow-up waves. It covered a complete range of child development domains following the nurturing care We designed a data system to child development and nurturing care which were critical for policy and This study has a high response rate that is mainly to a online and communication between the research and all the kindergartens. It set all items as required to It also set a normal range for to due to We the of each kindergarten via the management system and provided and to kindergartens with low The main of this study is that the information we collected from the online questionnaire was It was to a which was a of a study. For birth conditions were the most to have since it information from years However, we did to most of the with a standardized questionnaire to ensure and is that we only carried out an assessment for the last we did not conduct the for the first wave as a which limit from a time All SCHEDULE-P data are and by the research at Shanghai Children’s Medical There is an for using the After the is approved by the data can be with for research on can be by the study and The Shanghai Children's Health, Education and Lifestyle Evaluation, Preschool (SCHEDULE-P) was a large cohort study to explore how nurturing care early childhood development (ECD) in children aged 3–6 years in Shanghai. SCHEDULE-P three online surveys and one field survey. The online surveys followed up on the development trajectories of children from junior grade to senior grade in The field survey examined cognitive assessed by neuropsychological tools. The SCHEDULE-P covered a complete range of based on a nurturing care framework, such as and socioeconomic child care health status, child behaviour and social and mental health of The response rate of each wave was one of the highest among this of survey. In 2016, 20 899 junior grade children from 191 kindergartens were randomly selected as and 20 324 subjects the first wave, with a response rate of 97.25%. The second survey in the grade included 18 316 with a follow-up rate of 90.12%. Finally, 17 233 subjects in the last wave, with a follow-up rate of 84.79%. This was by Shanghai Education of China Shanghai Children's Health Shanghai Municipal Commission of Health and and Commission Shanghai Shanghai Municipal Education Commission and Shanghai Jiao Tong University School of Medicine on Child Center