Influencing factors of psychological and behavioral development in 1- to 2-year-old children based on random forest algorithm andmultivariate logistic regression analysis model

Abstract

Objectives To investigate the current status of psychological and behavioral development in children and its influencing factors, in order to provide a scientific basis for the prevention and control of developmental issues in children. Methods A total of 694 children aged 1 - 2 years who underwent routine health check-ups at the Child Health Care Department of Guiyang Maternal and Child Health Hospital from July 2022 to December 2023 were selected as the study subjects. Their parents were surveyed on general demographic information, screen exposure, and parent-child interaction levels. The Gesell Developmental Scale was used to assess the children′s psychological and behavioral development. The random forest algorithm was employed to rank the importance of influencing factors, and the screened variables were included in a multivariate logistic regression analysis model. Results The ranking of the importance of factors affecting children′s psychological and behavioral development was as follows: birth weight, gender, parity, gestational age, monthly household income, outdoor activities, feeding modes, screen exposure, parent-child interaction, and mode of conception. Multivariate logistic regression analysis revealed that a birth weight ≥2 500g (OR＝0.241, 95%CI: 0.108 - 0.536) and high-quality parent-child interaction (OR＝0.267, 95% CI: 0.122 - 0.585) were protective factors against the risk of delayed psychological and behavioral development in children. Conclusions The psychological and behavioral development of children aged 1 - 2 years still requires significant attention. Therefore, high-quality prenatal management to control neonatal birth weight and the provision of high-quality parent-child interactions by caregivers are essential to promote early childhood development.

Keywords

psychological behavioral development / random forest model / multivariate logistic regression model