The massive flow of migrants from rural to urban areas in China over the past decades has sparked concerns about the development of left-behind children. Drawing on a six-round, longitudinal cohort survey in rural China from 2013 to 2023 that follows children from 6 months to 11 years of age, we analyse the effects of two maternal migration patterns – persistent migration (migration without return) and return migration (migration followed by return) – on the cognitive development and nutrition of left-behind children from infancy to early adolescence. The results show that persistent maternal migration has adverse effects on the cognitive development and increased the BMI of left-behind children. In contrast, maternal migration had no significant effect on either cognitive development or any indicator of nutrition when the mother later returned. Persistent maternal migration had a strong, long-term negative effect on the cognitive development of left-behind children especially when mothers migrate within one or one and a half years after childbirth; maternal migration also had a short-term, negative effect on cognitive development when mothers migrate when the child is between 2 and 3 years old. These effects are likely driven by the lower levels of stimulating parenting practices and dietary diversity provided by the stand-in primary caregivers of left-behind children.

Purpose: Previous research has found an association between with early childhood development (ECD) and paternal beliefs and evidence suggests that paternal beliefs about their own role in child rearing might affect parental involvement. However, there has been no such empirical study in rural China. This study examines the interrelationships among paternal beliefs, parental involvement and ECD among rural Chinese children.

Design/methodology/approach: Data used for this study were collected from 6-to-42-month-old children and their primary caregivers in southwestern China in 2020. ECD was assessed by the Bayley Scales of Infant and Toddler Development-III. Paternal beliefs were assessed with the Role of the Father Questionnaire. The non-parametric regression methods were used to construct the factor scores of ECD. The ordinary least squares models, the mediation models and the bootstrapping approach were employed to investigate the interrelationships of paternal beliefs, parental involvement and ECD.

Findings: A large share of the sample children displayed delays in cognitive, language and social-emotional development. Paternal involvement significantly mediated the link between paternal beliefs and child cognitive and social-emotional development, while maternal involvement mediated the associations between paternal beliefs and child language and social-emotional development.

Originality/value: This study provides a unique contribution by utilizing unique data fathers' beliefs and their involvement in parenting to investigate the underlying mechanisms of how the role of fathers in parenting can lead to the improvement of early child developmental outcomes. This study also provides the first empirical evidence on the role of paternal beliefs in fostering human capital formation during the early stages of life in rural China. This study suggests that shifting paternal beliefs and improving parental involvement are effective pathways to benefit rural children in their early development.

China’s urbanization has created growing peri-urban communities, where children face challenges to early childhood development (ECD) despite proximity to developed urban areas. Little is known about how the environmental factors of ECD differ between rural migrants and new urban residents. To address this, we sampled 77 peri-urban households with 18–24-month-old Han Chinese children to examine the associations between early cognitive and language development, parental self-efficacy, stimulating parenting practices, and the home language environment measured with Language Environment Analysis (LENA). We find that rural migrants and new urban residents exhibited no significant difference in any child or household characteristic except parental residency. There was no significant difference in the family environment factors or early cognitive or language development, either. However, parental self-efficacy and stimulating parenting practices both predicted better cognitive and language development in rural migrant households, whereas only conversation turn counts predicted better language development in new urban resident households.

Background
The quality of parenting can affect the developmental outcomes of young children. This study aims to investigate the associations between parenting quality and the early childhood development of children under age 3 across four major rural subpopulations in China.

Methods
Using a stratified cluster sampling method, 760 children aged 6–36 months and their primary caregivers in four rural subpopulations from four provinces and a metropolis in China were surveyed. Child development was assessed by the Third Edition of the Bayley Scales of Infant and Toddler Development. Parenting quality was measured using the Family Care Indicators. Data were analyzed using descriptive statistics, t-tests, multivariable regression analysis, and linear regression analysis.

Results
Across the four subpopulations, prevalences of delays of the sample children in four domains — cognition, language, social-emotional, and motor development are 52%, 45%, 52%, and 19%, respectively. The proportion of children with any type of delay is 82%, while over half (53%) have delays in at least two areas, and 27% have delays in three or more areas. Child’s mother as the primary caregiver, maternal education levels, and family asset values are all positively associated with the quality of parenting. Notably, low levels of parenting quality in rural China are linked to high rates of developmental delays.

Conclusions
This study demonstrates that the level of parenting quality is significantly associated with early childhood developmental outcomes. Results highlight the need for raising investments in family care to improve early childhood development in different rural subpopulations in China.

The number of acutely food insecure people worldwide has doubled since 2017, increasing demand for early warning systems (EWS) that can predict food emergencies. Advances in computational methods, and the growing availability of near-real time remote sensing data, suggest that big data approaches might help meet this need. But such models have thus far exhibited low predictive skill with respect to subpopulation-level acute malnutrition indicators. We explore whether updating training data with high frequency monitoring of the predictand can help improve machine learning models’ predictive performance with respect to child acute malnutrition by directly learning the dynamic determinants of rapidly evolving acute malnutrition crises. We combine supervised machine learning methods and remotely sensed feature sets with time series child anthropometric data from EWS’ sentinel sites to generate accurate forecasts of acute malnutrition at operationally meaningful time horizons. These advances can enhance intertemporal and geographic targeting of humanitarian response to impending food emergencies that otherwise have unacceptably high case fatality rates.

Achieving minimum dietary diversity (MDD), a crucial indicator of infant and young child diet quality, remains a challenge in rural China, especially for infants aged 6–11 months. This study examined the rate of MDD attainment in rural China, identified its determinants using the Capability, Opportunity, Motivation, and Behavior (COM-B) model and Bayesian network analysis, and estimated the potential impact of improving each modifiable determinant. A multi-stage sampling design selected 1328 caregivers of infants aged 6–11 months across 77 rural townships in China. Data were collected through a cross-sectional survey via in-person household interviews. Bayesian network analysis identified key factors influencing MDD attainment and their interrelationships, while Bayesian inference estimated MDD attainment probabilities. Results showed that only 22.2 % of the sample infants attained MDD. Bayesian network analysis revealed that caregiver knowledge (a proxy of capability), self-efficacy and habits (proxies of motivation), and infant age directly influenced MDD attainment. Social support (a proxy of opportunity) indirectly promoted MDD attainment by boosting self-efficacy and habit. Notably, simultaneous improvements in knowledge, self-efficacy, and habit could increase MDD attainment by 17.6 %, underscoring the potential effectiveness of interventions focused on enhancing caregiver capability and motivation. The critically low MDD attainment rate among rural Chinese infants highlights the urgent need for targeted interventions. Strategies should prioritize enhancing caregiver feeding knowledge, self-efficacy, and habit formation to improve infant dietary diversity. Addressing these key factors could substantially boost MDD attainment in rural China.

Background: The content of children’s screen exposure and interactive coviewing with caregivers are important determinants of early childhood development (ECD) that have been overlooked in past research in resource-limited rural regions. Given the prevalence of digital devices and diverse digital content today, determining screen use practices that minimize the negative impacts on children’s development is crucial for promoting healthy screen use among children.

Objective: This study aims to examine screen exposure among children aged <3 years in rural China and investigate its relationship with ECD outcomes, focusing on duration, content, coviewing, and interaction.

Methods: The sample includes all children aged between 6 and 26 months and their primary caregivers residing in the study area. A survey of screen exposure and household characteristics was conducted for 1052 eligible households. Caregivers reported the duration of screen exposure, defined as the average daily screen time over the past month; the content of exposure, defined by the time spent on educational and child-friendly content; and the caregiver’s presence and interaction with the child during exposure. ECD outcomes were evaluated using the third edition of the Bayley Scales of Infant and Toddler Development assessment scale and the Brief Infant Toddler Social Emotional Assessment. Ordinary least squares regression, logistic regression, and chi-square tests were conducted.

Results: In total, 28.23% (297/1052) of the children in our sample were first exposed to screens before the age of 12 months. Children exposed to screens had an average daily screen time of 27.57 (SD 38.90) minutes. Children who were exposed to screens before the age of 12 months and those who had longer screen time between the ages of 12 and 18 months were more likely to be at risk of motor developmental delays. Children exposed to educational content for >15 minutes on a daily basis had fewer social-emotional or behavioral problems than those with no screen exposure and a lower risk of delay in motor skills development than those exposed to educational content for <15 minutes on a daily basis. Caregiver interaction during screen exposure was associated with a lower risk of cognitive and language delays and better socioemotional skills.

Conclusions: The type of content viewed and how caregivers engaged in children’s screen time were strongly associated with ECD outcomes. Guiding parents to select educational content for their children and engaging in interactive coviewing may better protect children from the negative effects of screen exposure. The findings complement conclusions regarding the impact of screen exposure on ECD in resource-limited rural areas.

Background: Micronutrient powder (MNP or Yingyangbao, a dietary supplement that contains multiple vitamins and minerals) programs can reduce the risk of anemia among children. One such program in China distributed free MNP to children aged 6–24 months in poor rural areas. However, there are indications that the generation of primary caregiver (i.e., parent or grandparent) may influence MNP feeding behavior. The purpose of this study was to investigate the prevalence and influencing factors of effective MNP feeding behavior among parents and grandparents in a rural developing setting.

Methods: We conducted a cross-sectional study of 884 caregivers and children in formerly impoverished rural areas of Sichuan Province using multi-stage random sampling. Data were collected on caregivers’ MNP feeding behavior, their core perceptions and health beliefs about MNP, children’s responses to MNP, MNP delivery patterns, and demographic characteristics. Influencing factors of parents’ and grandparents’ MNP feeding behavior were analyzed using two-level logistic regression.

Results: Total effective MNP feeding behavior rate among caregivers was 40.95%. Grandparents demonstrated higher MNP adoption than parents (χ2 = 4.445, P = 0.035). After controlling for sociodemographic characteristics, grandparents were more likely than parents to achieve effective MNP adoption (OR = 1.360, P = 0.035); child’s preference for MNPs (ORparental subgroup = 1.736, ORgrandparental subgroup = 1.496; P < 0.050) and caregiver’s self-efficacy (ORparental subgroup = 1.157, ORgrandparental subgroup = 1.393; P < 0.050) were influencing factors of feeding behavior for caregivers of both generations. Parents were also influenced by perceived barriers to feeding MNPs (OR = 0.904, P = 0.040), while grandparents were more strongly influenced by child’s discomfort to MNPs (OR = 0.240, P = 0.023) and caregiver’s knowledge about MNPs (OR = 1.557, P = 0.014).

Conclusion: Future efforts to improve the feeding behavior of caregivers in MNP programs both in China and abroad should improve children’s preference for MNP by changing its composition and taste, and increase caregivers’ feeding self-efficacy through health education. Moreover, parents should be taught to cope with time and ability barriers that may limit MNP adoption, while grandparents should be emphasized the importance and potential side effects of MNP.

The effects of psychological factors on complementary feeding behavior have been widely described, yet the mechanisms underlying the complex relationships among personality traits, self-efficacy, and complementary feeding behavior remain unclear. This cross-sectional study was conducted using a multi-stage cluster sampling process to select caregiver-child dyads in Western rural China. Personality traits, self-efficacy, and complementary feeding behavior were evaluated respectively. Both multiple logistic regression and Bayesian network structure (BNs) were used to explore these associations. A total of 787 caregiver-child dyads were enrolled. Results from multiple logistic regression indicated that caregivers with medium (OR = 2.05, p < 0.001) or high (OR = 1.58, p = 0.04) levels of extraversion, as well as those with high self-efficacy recording complementary feeding (OR = 2.08, p < 0.001), significantly increased the likelihood of meeting the criteria for the Infant and Child Feeding Index (ICFI) qualification. Further, BNs were employed to elucidate the pathways of influence, revealing a direct association between the caregiver's level of extraversion, level of self-efficacy and the ICFI. Additionally, the analysis indicated that a caregiver's openness indirectly influenced the ICFI through its influence on self-efficacy regarding complementary feeding. This is one of few studies exploring associations between personality traits, self-efficacy, and complementary feeding behavior. The study highlights the importance of understanding individual differences in caregiving and suggests that interventions should focus on enhancing caregivers' self-efficacy, rather than solely targeting personality traits.