The adult height percentage (AHP) is an indicator of maturity state, which reflects variation in growth rate. Several methods estimates adult height; however, its similarities or differences have not been documented in a concrete way in literature. The aims of the present work were to compare three common methods of AHP estimation, to identify children and adolescents secular changes and to develop AHP reference values in Portuguese population. Skeletal age, weight and height were measure in 799 children and 736 girls from 7.0 to 16.5 years; in addition, parents height was self-reported by them to estimate the AHP by TW3, RWT and KR methods. ANOVA was used to estimate differences between TW3, KR and RWT methods, as well as Bland-Altman graphs. Also, Kruskal-Wallis test was applied. No differences were found between TW3 and KR methods in all age groups, in both sexes, when subjects were classified by chronological age (P> 0.05). Likewise, no secular changes were observed in AHP (P> 0.05). Not only TW3 and KR protocols can be interchangeable each other because they did not present differences in the AHP estimation at different ages and in both sexes. However, secular changes were not observed in AHP estimation by these methods.

KEY WORDS: Maturation; Growth; Adult height predicted; Non-invasive methods.