Effect of birth size and proportionality on BMI and skinfold thickness in early adolescence: prospective birth cohort...

To assess the effect of birth weight, length and ponderal index at birth on body mass index (BMI) and skinfold thickness in early adolescence.

Population-based, prospective birth cohort study in Pelotas, Brazil. Out of 5249 cohort members, 87.5% were traced at 11 years of age. Birth weight, length and ponderal index (birth weight/birth length(3) x 100) -- treated as continuous variable and divided into fourths -- were the exposure variables. BMI at 11 years, triceps and subscapular skinfolds were the outcomes. Confounders included sex, gestational age, parity, maternal schooling, family income, maternal smoking during pregnancy and maternal BMI.

All three exposure variables were significantly associated with BMI in early adolescence, but the strongest effect was observed for birth weight. Each unit (Z-score) increase in birth weight was associated with 0.46 kg m(-2) increase in BMI at 11 years. The equivalent coefficients for ponderal index and birth length were 0.22 and 0.21 kg m(-2), respectively. High birth weight, length and ponderal index were also associated with increased values for triceps and subscapular skinfolds at 11 years, and with increased prevalence of obesity.

Of the three birth exposures studied, birth weight presented the strongest effect on anthropometry in early adolescence. Ponderal index, a proportionality indicator associated with infant mortality, hospitalizations and anthropometry in infancy, was also associated with anthropometry in early adolescence, but its predictive value for the latter is lower than that of birthweight alone. All three birth size indicators studied presented poor predictability of the later risk of obesity.