Introduction and aims Assessing pubertal growth is complex, due to variation in age when puberty begins. We have developed a new puberty phase specific (PPS) growth reference, constructed using Dutch national cross-sectional data, recalibrated to match the UK 1990 reference. It uses Tanner staging simplified into three phases: Pre puberty (Tanner stage 1), In puberty (2 and 3) and Completing puberty (4 and 5). The aim of this analysis was to assess the validity of the PPS reference when applied to UK children, and the impact its use might have on the assessment of pubertal growth.

Subjects and method We used the Chard data set: longitudinal height and weight data collected on 124 healthy UK children from 1981 to 1988 for ages 8.3-16.6 years. There were 1-14 measurement occasions per child, 1,252 in total, all with exact age and Tanner pubertal staging. All measures were converted into SD scores (SDS) based on the UK90 reference and the new PPS reference.

Results Within each phase, the measurements fitted closely to the PPS reference (mean height SDS by phase: Pre 0.1, In 0.1, Completing 0.3; mean weight SDS: Pre −0.1, In 0.1, Completing 0.1); PPS SDS showed little trend with age in each phase, in contrast to UK90 where the SDS fell significantly (see table 1).

For 72 children with measurements in both the Pre and Completing phases, a change of more than one centile space (0.67 SDS) over time was seen in only 15% (11) for PPS height SDS and only 26% (17) for PPS weight SDS.

Conclusions Children entering puberty relatively late tend to appear shorter and lighter based on the conventional UK90 chart, but do not when compared to a reference that adjusts for phase of puberty. The PPS reference shows a good fit to UK children and should allow growth though puberty to be tracked more accurately.