Early stature prediction method using stature growth parameters.

Area of Science:

• Human Biology
• Medical Science
• Growth Dynamics

Background:

• Accurate human stature prediction is a long-standing challenge in medical science and biology.
• Existing methods often rely on radiographic assessments, limiting their applicability.

Purpose of the Study:

• To develop a non-radiographic method for predicting final adult stature.
• To create a prediction model applicable during the early pubertal growth period.

Main Methods:

• Applied Preece and Baines model 1 (PB1) and Jolicoeur-Pontier-Pernin-Sempe (JPPS) nonlinear growth curves to serial stature data of 400 Korean children.
• Derived five biological parameters, including take-off (TO) related variables, from the growth curves.
• Developed a multiple linear regression equation for final stature prediction, incorporating TO-related parameters estimated via linear interpolation.

Main Results:

• The final stature prediction model demonstrated excellent validity and accuracy in cross-validation.
• Prediction accuracy was positively correlated with the time elapsed since the take-off phase.
• The inclusion of TO-related parameters enhanced the model's predictive capabilities.

Conclusions:

• Multiple regression analysis incorporating biological parameters provides a valid and accurate method for stature prediction.
• The developed method, utilizing TO-related parameters, enables earlier growth evaluation and prediction compared to existing approaches.