Longitudinal Association Between Body Mass Index z-Score and Puberty: Structural Equation Modeling Analyses

Changes in the timing of puberty may reflect shifts in population health, including the rising prevalence of overweight and obesity. Therefore, this study aimed to analyze the longitudinal association between body mass index (BMI) in childhood and pubertal development 5 years later among Brazilian students.

This longitudinal study included 494 students aged 7-10 years. Data were collected in 2007 and 2012. BMI z-scores were calculated. Pubertal development was self-assessed using Tanner stages, and girls reported age at menarche. Structural equation modeling was used to assess the effects of the 2007 BMI on sexual maturation (SM) in 2012, adjusting for socioeconomic status (SES), birth weight, breastfeeding, physical activity, and dietary patterns (DP).

No statistically significant association between BMI and SM was observed in either sex. Among boys, higher adherence to DP IV (milk, coffee with milk, cheese, breads/biscuits) (β = -0.21) and higher SES (β = -0.21) were associated with normal/late SM. Among girls, a higher 2007 BMI z-score (β = -0.27) had a direct negative effect on age at menarche, while DP II (ultra-processed foods) showed an indirect negative effect on age at menarche, mediated by the 2007 BMI z-score (β = -0.05).

This study found that in girls, higher childhood BMI was associated with an earlier age at menarche. In boys, DP IV and SES were associated with normal/late SM. These findings highlight the significance of monitoring puberty timing at the population level and the need for sex-sensitive, prospective research to elucidate the determinants of earlier puberty, especially in low- and middle-income countries.