BACKGROUND: Children's growth is increasingly considered a key mediator of later life outcomes. When examining weight growth, the correlation between repeated observations on the same subject must be regarded as well-modelled. This study aimed to analyze children's weight growth variations and associated factors in Ethiopia, India, Peru, and Vietnam using a fractional polynomial mixed-effects model.

METHODS: This study used longitudinal data from the Young Lives Cohort Study conducted from 2002 to 2016 in Ethiopia, India, Peru, and Vietnam. The study included 7,140 children of 1 to 15 years old. A fractional polynomial mixed-effects model was used to analyze the data.

RESULTS: Ethiopian, Peruvian, and Vietnamese children had significantly higher average body weights than children in India (1.426, P<0.001; 1.992, P<0.001; 1.334, P<0.001, respectively). Girl children's average body weight was significantly 0.15 times less than that of boys (-0.148; P=0.027). The average weight of rural children was significantly 0.671 times less than that of urban children (0.671, P<0.001). Children from Peru and Vietnam had higher rates of weight change than those from India. However, the rate of weight change was lower in Ethiopian children than in Indian children. Children from urban areas had a significantly higher rate of weight gain than those from rural areas.

CONCLUSION: Country, sex, residence, parental education, household size, wealth, good drinking water, and reliable power affected children's longitudinal weight growth. Therefore, WHO and the nation's health ministry should monitor children's weight growth status and these associated factors to plan future action.

KEYWORDS: Children, Weight Growth, Growth Rate, Linear Mixed-Effects Mode.