OBJECTIVE: To investigate birth-to-childhood tracking of linear growth and weight gain across the distribution of length/height and weight for age z-scores and according to household wealth. METHODS: Data from 614 children from the MINA-Brazil Study with repeated anthropometric measurements at birth and up to age five years were used. Z-scores were calculated for length/height (HAZ) and weight (WAZ) according to international standards. Birth-to-childhood tracking was separately estimated using quantile regression models for HAZ and WAZ, extracting coefficients and 95% confidence intervals (95%CI) at the 25th, 50th, and 75th quantiles. In a subgroup analysis, we estimated tracking between birth and age two years, and between ages two and five years. To investigate disparities in tracking, interaction terms between household wealth indexes (at birth and age five years) and newborn size z-scores were included in the models. RESULTS: Tracking coefficients were significant and had similar magnitude across the distribution of anthropometric indices at age five years (HAZ, 50th quantile: 0.23, 95%CI: 0.11 to 0.35; WAZ, 50th quantile: 0.31, 95%CI: 0.19 to 0.43). Greater tracking was observed between ages two and five years, with coefficients above 0.82. Significantly higher tracking of linear growth was observed among children from wealthier households, both at birth, at the lower bounds of HAZ distribution (25th quantile: 0.30, 95%CI: 0.13 to 0.56), and during childhood, in the entire HAZ distribution at five years. For weight gain, stronger tracking was observed at the upper bounds of WAZ distribution at age five years among children from wealthier households at birth (75th quantile: 0.59, 95%CI: 0.35 to 0.83) and during childhood (75th quantile: 0.54, 95%CI: 0.15 to 0.93). CONCLUSION: There was significant tracking of HAZ and WAZ since birth, with indication of substantial stability of nutritional status between ages two and five years. Differential tracking according to household wealth should be considered for planning early interventions for preventing malnutrition. [ABSTRACT FROM AUTHOR]