Table 2c. Gymnast Cohort Regression Results: Non-bone Lean Mass based Models for the Proximal Femur.
| Proximal Femur Outcome |
Model Adj R2 |
Constant | Age | DXA nbFFM | PBF | MEN | GYMHRS | |||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| St β | PCC | St β | PCC | St β | PCC | St β | PCC | St β | PCC | |||
| FNBMC | 0.85 | −0.31 | 0.19* | 0.24 | 0.68*** | 0.72 | 0.11* | 0.27 | 0.13* | 0.24 | −0.01 | −0.01 |
| FNaBMD | 0.63 | 0.43 | -- | -- | 0.53*** | 0.56 | -- | -- | 0.23** | 0.28 | 0.20** | 0.29 |
| NNbCSA | 0.79 | 0.37 | 0.19* | 0.21 | 0.64*** | 0.64 | -- | -- | 0.15* | 0.24 | −0.03 | −0.05 |
| NNCSMI | 0.75 | −0.61 | 0.29** | 0.30 | 0.69*** | 0.64 | -- | -- | -- | -- | −0.17** | −0.27 |
| NN width | 0.49 | 1.90 | -- | -- | 0.78*** | 0.70 | -- | -- | -- | -- | −0.25*** | −0.30 |
| NN ED | 0.28 | 1.75 | -- | -- | 0.60*** | 0.54 | -- | -- | -- | -- | -0.33*** | -0.33 |
| NN CT | 0.63 | 0.09 | -- | -- | 0.53*** | 0.55 | -- | -- | 0.24** | 0.29 | 0.18* | 0.26 |
| NN Z | 0.79 | -0.24 | 0.29** | 0.33 | 0.69*** | 0.67 | -- | -- | -- | -- | −0.11 | −0.20 |
| NN BR | 0.25 | 9.50 | -- | -- | −0.27** | −0.28 | -- | -- | -- | -- | −0.33*** | −0.33 |
| SN Angle | 0.10 | 126.6 | -- | -- | 0.37*** | 0.34 | -- | -- | -- | -- | −0.26* | −0.24 |
| HAL | 0.69 | 64.28 | -- | -- | 0.85*** | 0.81 | -- | -- | -- | -- | −0.04 | −0.06 |
DXA nbFFM: Non-bone Lean Mass; PBF: % Body Fat; MEN: Pre-menarche= 0, Post-menarche=1;
GYMHRS: mean h/wk gymnastic training for the year prior to the focal DXA scan; St β: standardized beta; PCC: partial correlation coefficient, represents square root of percent of variance explained after accounting for the effects of other variables; FN: femoral neck; NN: narrow neck;
BMC: bone mineral content; aBMD: areal bone mineral density; bCSA: bone tissue cross-sectional area; CSMI: cross-sectional moment of inertia; ED: endosteal diameter; CT: cortical thickness; Z: section modulus; BR: buckling ratio; SN Angle: shaft neck angle;
HAL: hip axis length.
Presented FNBMC and FNBMD results use data that are corrected for fan beam magnification error.
Bold type indicates that GYMHRS was not a significant independent predictor within the model.
Models were built as follows: age, nbFFM, PBF, MEN, GYMHRS, Interaction (GYMHRS x MEN),
< 0.05
≤ 0.01
≤ 0.001 for t significance of β.