Table 3.
The slope, bias and limit of agreement (LOA) obtained from the Bland-Altman plots of the differences between the LBM from DXA and each of the predictive equations. Pearson correlations are also reported between LBM from DXA and each of the predictive equations. Results are shown for pooled males and females (M + W), males only (M) and females only (W). *highlights the optimum fit. +indicates other fits that are statistically consistent with the optimum
M + W | M + W | M + W | M + W | M | M | M | M | W | W | W | W | |
---|---|---|---|---|---|---|---|---|---|---|---|---|
Slope | Bias | LOA | Pearson | Slope | Bias | LOA | Pearson | Slope | Bias | LOA | Pearson | |
Eq1: Hume 19 | 0.32 | −1.1* | 16.9 | 0.945 | 0.32 | 1.8* | 14.2+ | 0.890+ | 0.08 | −3.8 | 11.3 | 0.861+ |
Eq2: Hume & Weyers 20 | 0.13 | −3.5 | 13.0 | 0.957+ | 0.20 | −2.5+ | 13.0+ | 0.899+ | 0.15 | −4.3 | 12.1 | 0.840 |
Eq3: James & Waterlow 21 | 0.15 | −4.1 | 14.2 | 0.948 | 0.21 | −3.3 | 13.6+ | 0.889+ | 0.33 | −4.9 | 14.0 | 0.776 |
Eq4: Hallynck 22 | 0.15 | −4.1 | 14.2 | 0.948 | 0.21 | −3.2 | 13.6+ | 0.882 | 0.33 | −4.9 | 14.0 | 0.776 |
Eq5: Boer 15 | 0.14 | −3.6 | 13.1 | 0.957+ | 0.20 | −2.6+ | 13.0+ | 0.899+ | 0.15 | −4.5 | 12.1 | 0.840 |
Eq6: Deurenberg 23 | 0.14 | 1.5+ | 13.9 | 0.950+ | 0.31 | 2.0+ | 14.2 | 0.887+ | 0.42 | 1.0+ | 13.3 | 0.810 |
Eq7: Zasadny & Wahl 7 | 0.02+ | −12.8 | 25.0 | 0.830 | 0.12 | −12.9 | 27.5 | 0.525 | 0.10 | −12.7 | 22.4 | 0.451 |
Eq8: Morgan & Bray 16 | 0.05+ | −5.0 | 13.6 | 0.949 | 0.13 | −5.1 | 13.1+ | 0.896+ | 0.33 | −4.9 | 14.0 | 0.776 |
Eq9: Gallagher 24 | 0.04+ | −2.4 | 11.8* | 0.962* | 0.13 | −2.4+ | 12.8* | 0.900* | 0.17 | −2.3 | 10.7* | 0.877* |
Eq10: Janmahasatian 25 | 0.00* | −1.6+ | 12.3 | 0.959+ | 0.21 | −2.3+ | 13.3+ | 0.895+ | 0.14 | −0.9* | 10.7* | 0.876+ |
DXA, Dual-energy X-ray ansorptiometry; Eq, Equation; LBM, Lean body mass; LOA, Limit of agreement.