Heterogeneity of health-risk behaviours in left-behind children in rural western China: a cross-sectional survey
Yuanyi Ji,Huanyu Xu,Yu Zhang,Qiaolan Liu
DOI: https://doi.org/10.1016/S0140-6736(17)33142-2
IF: 202.731
2017-01-01
The Lancet
Abstract:Abstract Background Chinese children left behind in rural areas while their parents seek work elsewhere experience a lack of parental monitoring and companionship, resulting in higher prevalence of health-risk behaviours. This study aimed to investigate the heterogeneity of health-risk behaviours among rural left-behind children in western China, and the associations between demographic characteristics and health-risk subgroups. Methods In this cross-sectional survey study, rural left-behind children (those who had not seen their parents for u003e6 months) from middle schools (aged 10–17 years) in Sichuan province, selected using a stratified cluster sampling method, were surveyed between November and December, 2015. Participants were given a self-administered questionnaire, which measured eight health-risk behaviours (unhealthy diet, tobacco use, binge drinking, accidental injury, physical inactivity, self-injurious behaviour, unhealthy internet use, and suicidal thoughts), demographics, household characteristics, their parentsu0027 relationship status, and their economic status. Latent class analysis (LCA) was used to identify subgroups of rural left-behind children with distinct patterns of health-risk behaviours. Differences between health-risk subgroups in relation to demographic characteristics were examined using multinomial logistic regression analysis. Informed consent was obtained from children and parents, and ethics approval from the Medical Ethical Committee of Sichuan University, China (number 20140307). Findings 1596 rural left-behind children completed the questionnaire (mean age 14·7 years [SD 1·44]; 712 [45%] boys). Four classes were obtained from LCA: high risk (69 [4%] of 1596), high risk of self-injury and suicidal thoughts (321 [20%]), moderate risk (479 [30%]), and low risk (727 [46%]). The multinomial logistic regression model revealed that, compared with the low-risk group, boys (odds ratio [OR] 2·22, 95% CI 1·32–3·74), and children whose parentsu0027 relationship was poor (OR 4·49, 2·48–9·73) were more likely to be classified as high risk. Children whose parentsu0027 relationship was poor (OR 2·98, 1·93–4·62) and those with high living expenses (OR 2·03, 1·18–3·51) were more likely to have a high risk of self-injury and suicidal thoughts than be at a low risk. Interpretation Our study is, to our knowledge, the first in-depth analysis of health-risk behaviours in left-behind children in rural western China. The clusters of heterogeneous health-risk behaviours in rural left-behind children are likely to require different tailored interventions. Particular attention should be paid to left-behind boys who have higher living expenses and those whose parents have a poor relationship when designing preventive group interventions. Funding National Natural Science Foundation of China (81472994, 30972546).