Behavioural and emotional problems in children affected by parental migration in rural China: a cross-sectional survey

Abstract

BackgroundConcerns for the welfare and human capital development of children of migrant workers in China are growing. An estimated 60 million children were left behind in rural areas, and a further 15 million accompanied their parents to urban areas. The aim of this study was to investigate the effects of migration on psychosocial and developmental outcomes of left-behind children (LBC) and urban migrant children (UMC). MethodsSelf-completion and well-validated questionnaires were administered to primary-school and middle-school children, who were identified by random stratified sampling of schools and within schools in two rural counties and two cities, which are the major feeder and receiving areas for migrants in Zhejiang province, respectively. The questionnaire included migration status and history, and the Strengths and Difficulties Questionnaire (SDQ), which measures emotional disorders, conduct disorders, hyperactivity, peer problems, and prosocial behaviours. Logistic regression was used to define the effect of migration on SDQ outcomes. Informed consent was obtained from children and their parents, and ethical approval from Zhejiang University Research Ethics Committee. FindingsFull data were available for 1876 UMC and 671 LBC aged 9–17 years (1287 boys and 1124 girls [136 were excluded in the analysis because of missing data]; mean age 13·5 years [SD 1·5] for boys and 13·3 years [1·6] for girls). 192 (8%; 120 [6%] UMC and 72 [11%] LBC) had emotional disorders, 298 (12%; 221 [12%] UMC and 77 [12%] LBC) had conduct disorders, 300 (12%; 214 [11%] UMC and 86 [13%] LBC) showed hyperactivity, 357 (14%; 96 [5%] UMC and 261 [39%] LBC) reported peer problems, and 2327 (91%; 1726 [92%] UMC and 601 [90%] LBC) had prosocial tendency. Compared with UMC, LBC manifested more emotional disorders (adjusted odds ratio 1·77 [95% CI 1·26–2·48]; p=0·001) and peer problems (11·21 [8·46–14·85]; p=0·000), but were less likely to be prosocial (0·71 [0·51–0·99]; p=0·04). Large differences were reported between the two groups in emotional disorders, peer problems, total difficulty score, and prosocial tendency, some of the key items of the SDQ. InterpretationMigration with parents, rather than separation from parents, led to more favourable emotional and behavioural outcomes. Follow-up studies are needed to explore the long-term effects of parental rural–urban migration on children. FundingUBS Optimus Foundation.