Reliability of self-reported pubertal development scale for girls in early adolescent: a school population-based study

Objectives: The study aimed to evaluate the correlation between self-reported pubertal developmental scale (PDS) and physically assessed Tanner staging by an experienced pediatrician among girls. Methods: In a school population-based study in Zhongshan, China, we recruited 1,722 girls in grades 1-3 by a multistage stratified cluster random sampling method. Participants completed self-reported PDS questionnaire prior to physical examination. Breast development was evaluated by a female pediatrician combined with ultrasound examination for overweight/obese girls; pubic hair development was evaluated. Otherwise, we tested follicle-stimulating hormone (FSH) and luteinizing hormone (LH) for some participants. Results: We observed a weak association between Tanner-derived composite stage (TDCS) and puberty category scores (PCS) (tau=0.288, p<0.001) among all girls. There was correlation (tau=0.314, p=0.001) between ultrasound-derived composite stage (UDCS) and PCS among overweight/obese girls. Moreover, among overweight/obese girls, PCS was positively correlated with LH (r=0.265, p=0.008), but not FSH (r=0.155, p=0.123), and when the basal LH value was greater than 0.3 mIU/mL, the proportion of PCS stage >= 2 (9/18) was higher than the proportion of TDCS >= 2 (5/18). As for the determination of pubertal onset, when UDCS was used as the gold standard, the specificity of PCS was 0.86 and positive predictive value was 90.00 %. Conclusions: There was a weak correlation between PCS and TDCS among girls early adolescence. Moreover, among overweight/obese girls, combining hormone values, ultrasonographic stage of breast, and the positive predictive value of PCS, we posit that self-reported PDS might be a more reliable method than TDCS to evaluate pubertal development among overweight/obese girls.