Longitudinal modeling of adiposity in periadolescent Greek schoolchildren.

PURPOSE Obesity has an etiology that is multidimensional in nature. Given the dearth of longitudinal data, we examined changes in adipose tissue (Ad) in relation to physical activity levels (PA), aerobic fitness (AF), and energy intake (EI) in Greek schoolchildren, as they progressed from age 12 to 14 yr. METHODS This was a 2-yr and three-time-point (TP) study. Participants (N=210 (TP1); =204 (TP2); =198 (TP3)) were assessed for anthropometry, maturity status, Ad, PA, AF, and EI. Mean values were used for exploratory analyses, whereas two generalized estimating equations (GEE) models examined for longitudinal associations between the studied parameters. The first (GEE1) aimed to extract inherent associations between the dependent (Ad) and independent (PA, AF, EI) variables for the entire study period. For further evidence of association, the second analysis (GEE2) used the independent variables at TP1 and TP2 to predict the dependent variables at TP3. RESULTS Levels of Ad in boys decreased significantly (P<0.05) from TP1 to TP3, whereas the same variable demonstrated a nonsignificant increase (P>0.05) in girls. GEE1 revealed that longitudinal changes in Ad were significantly associated only with PA (beta=-0.16; P<0.001) and AF (beta=-0.09; P<0.05) for all schoolchildren. Similarly, GEE2 revealed that the main factors (at TP1 and TP2) predicting the development of Ad (at TP3) were PA (beta=-0.14; P<0.001) followed by AF (beta=-0.10; P<0.05). CONCLUSION With respect to data presented, we established that longitudinal changes in Ad are mainly accompanied by changes in PA and, to a lesser extent, AF levels.