Supplementary Table S9. Absolute-load-adjusted sensitivity analyses for principal sEMG findings.
Panel A. Principal contrasts before and after adjustment for absolute load |
| Muscle
/ contrast |
Unadjusted
estimate [95% CI] |
Load-adjusted
estimate [95% CI] |
Change
after adjustment |
Direction
retained |
| PD: RP
- CB (marginal) |
7.96 [5.16,
10.76] |
13.27 [8.57,
17.96] |
+5.31 (66.7%) |
Yes |
| BB:
WVS - SB, lifting |
10.95
[6.93, 14.97] |
12.32
[7.75, 16.90] |
+1.37
(12.6%) |
Yes |
| BB: WVS
- CB, lifting |
10.59 [6.59,
14.59] |
11.42 [7.21,
15.63] |
+0.83 (7.8%) |
Yes |
| BB:
WVS - RP, lifting |
15.67
[11.84, 19.49] |
14.17
[9.68, 18.67] |
-1.49
(9.5%) |
Yes |
| MT: lowering
- lifting (marginal) |
6.15 [2.83,
9.47] |
6.13 [2.81,
9.46] |
-0.02 (0.3%) |
Yes |
| Contrast estimates are in %MVIC. Change is the load-adjusted minus unadjusted estimate; the percentage in parentheses is the absolute percentage change relative to the unadjusted estimate. The load-adjusted model added actual experimental load as a covariate to the attachment × phase mixed-effects model with participant retained as a random intercept. Secondary triceps-brachii contrasts also retained the same direction after adjustment: WVS − CB = 9.61 [4.16, 15.06] and WVS − RP = 8.70 [2.77, 14.63].
|
Panel B. Fixed-effect terms for absolute
load and the load × phase sensitivity model.
| Muscle |
Absolute-load β |
p |
Load × phase β |
p |
| PD |
0.757 |
0.006 |
-0.118 |
0.829 |
| BB |
0.332 |
0.219 |
0.874 |
0.095 |
| MT |
-0.408 |
0.313 |
0.415 |
0.600 |
| TB |
0.097 |
0.818 |
-0.304 |
0.711 |
| LD |
-0.678 |
0.045 |
-0.060 |
0.928 |
|
Absolute-load β is the fixed-effect coefficient for the actual experimental load covariate in the load-adjusted model. The load × phase coefficient is from an additional sensitivity model containing the corresponding interaction term. Absolute load was associated with posterior-deltoid and latissimus-dorsi sEMG (p = 0.006 and p = 0.045, respectively), whereas no load × phase interaction reached statistical significance (all p ≥ 0.095). These models are robustness analyses only; because absolute load is structurally related to attachment-specific 1RM, covariate adjustment does not identify a load-independent causal effect of attachment geometry. PD = posterior deltoid; BB = biceps brachii; MT = middle trapezius; TB = triceps brachii; LD = latissimus dorsi; WVS = wide V-shaped handle; SB = straight bar; CB = cambered bar; RP = rope.
|
|