Prepubertal Growth Trajectory and Pubertal Onset

Adiposity has been reported to be a major contributor to earlier pubertal timing, but most pediatric studies have relied on body mass index (BMI) from single or short-term measurements. Studies tracking growth repeatedly from birth to puberty are needed.

To explore the association of growth trajectory and cumulative exposure to different levels of adiposity (CEA) with pubertal onset across the first decade of life and identify sensitive periods for possible weight intervention.

This population-based cohort study included data from 2 birth cohorts: the Longitudinal Study of Australian Children (LSAC), conducted from March 2004 to September 2021, with more than a 10-year follow-up, and the Tianjin Birth Cohort Study (TBCS) in China, conducted from May 2021 to April 2024, with follow-up from December 2010 to April 2024. Data analysis was conducted from April 2023 to November 2025. Participant inclusion required at least 4 anthropometric measures in the LSAC and 9 in the TBCS and completed measures of pubertal onset in both cohorts.

Prepubertal BMI trajectories, CEA, and rates of BMI increase at each age.

Pubertal status and timing were obtained by a parent-reported Pubertal Development Scale. A latent class growth mixed model was used to identify prepubertal BMI trajectories. An interval regression model and the Cox proportional hazards regression model were used to examine the association of the exposures with the age and risk of pubertal onset.

A total of 3354 Australian children (1723 boys [51.37%]) and 1105 Chinese children (563 girls [50.95%]) were included. At the last round, the mean (SD) ages were similar across sexes within each cohort, while children in LSAC (14.83 [0.61] years) were overall older than those in TBCS (10.63 [0.60] years). Girls in BMI trajectory groups characterized as high-level or increasing were younger at pubertal onset (from β = -0.36 [95% CI, -0.66 to -0.07] years to β = -1.51 [95% CI, -2.68 to -0.35] years) and were associated with increased risk of pubertal initiation (from hazard ratio [HR], 1.35 [95% CI, 1.04 to 1.74] to HR, 2.80 [95% CI, 1.69 to 4.63]). A higher CEA and average CEA (both >2) were associated with an earlier age at pubertal onset (from β = -0.04 [95% CI, -0.05 to -0.03] years to β = -0.85 [95% CI, -1.48 to -0.23] years), with a greater effect size after averaging. Consistent results were found in boys of the LSAC but not those of the TBCS. Sensitive ages of 3 to 4 years were identified, at which BMI increase was associated with pubertal timing, with greater effect sizes of pubertal timing (from β = -1.35 [95% CI, -2.00 to -0.71] years to β = -3.41 [95% CI, -4.13 to -2.68] years) and greater risk of pubertal onset (from HR, 1.85 [95% CI, 1.29 to 2.67] to HR, 5.59 [95% CI, 3.73 to 8.37]) than at other ages. Notably, the effect sizes of CEA within this period were greater than that outside it.

In this cohort study, high-level or increasing prepubertal growth trajectories and greater CEA were associated with earlier and higher risk of pubertal onset. These findings highlight the importance of considering CEA in relation to early pubertal onset and suggest that ages 3 to 4 years may be an important intervention period for earlier pubertal onset monitoring.