Table 3.
Multiple Moderator Model: SBP Response to Dynamic RT (k=69)a
| Moderator Dimension/Level | (95% CI)b | β | P Value | SBP ∆ (mm Hg)c | |
|---|---|---|---|---|---|
| Resting SBP of RT sample, mm Hg | −0.311 | 0.011 | |||
| Normal=115±11 (k=16) | 0.00 (−0.23, 0.23) | 0.0 (−2.5, 2.5) | |||
| Prehypertension=130±13 (k=41) | −0.23 (−0.39, −0.07) | −3.0 (−5.1, −1.0)d | |||
| Hypertension=142±14 (k=14)f | −0.41 (−0.64, −0.19) | −5.7 (−9.0, −2.7)d , e | |||
| Race/ethnicity of RT sample | 0.354 | 0.002 | |||
| White samples (k=40) | −0.00 (−0.20, 0.20) | 0.0 (−2.6, 2.6) | |||
| Nonwhite samples (k=31)f | −0.36 (−0.56, −0.16) | −4.7 (−7.3, −2.1)d | |||
| BP medication use of RT sample | 0.261 | 0.034 | |||
| Taking BP medication (k=14) | −0.03 (−0.29, 0.23) | −0.4 (−3.8, 3.0) | |||
| Not taking BP medication (k=57)f | −0.33 (−0.48, −0.17) | −4.3 (−6.2, −2.2)d | |||
| RT exercises performed/session | −0.221 | 0.043 | |||
| <8 RT exercises=6 (k=37) | −0.11 (−0.29, 0.08) | −1.4 (−4.4, 1.0) | |||
| ≥8 RT exercises=12 (k=32)f | −0.34 (−0.55, −0.12) | −4.4 (−7.2, −1.6)d | |||
| Primary study outcome | −0.238 | 0.032 | |||
| BP focused outcome (k=34)f | −0.30 (−0.49, −0.10) | −3.9 (−6.4, −1.3)d | |||
| Non‐BP focused outcome (k=37) | −0.06 (−0.26, 0.14) | −0.8 (−3.4, 1.8) | |||
| Additive SBP modelf | Non‐white samples | −1.02 (−1.36, −0.67) | −14.3 (−19.0, −9.4) | ||
| White samples | −0.66 (−0.97, −0.35) | −9.2 (−13.6, −4.9) | |||
Resting SBP is presented as mean±SD. ∆ indicates change; BP, blood pressure; k, number of observations; RT, resistance training; SBP, systolic BP; β, standardized coefficient represents unique variance explained by moderator.
Multiple R 2 (variance explained by model, adjusted for number of moderators)=67.1%; I2 residual (variance unexplained by model)=27.3%.
Predicted weighted mean effect size (d+); estimate of the magnitude of SBP reduction among the RT group relative to control, while statistically controlling for the presence of each moderator shown in the above model (held constant at their mean), except for moderator/level of interest. This model also controls for (not shown): 1 versus 2 RT groups (β=−0.235; P=0.050) and SBP×RT exercises interaction (β=0.174; P=0.133).
SBP ∆= (95% CIs) back‐converted to mm Hg. For each moderator/level of interest, (95% CIs) were transformed arithmetically using the SD corresponding to the sample mean (130±13 mm Hg): ∆= (95% CI)×13 mm Hg, with the exception of normal SBP: ∆= (95% CI)×11 mm Hg; hypertension SBP and additive SBP model: ∆= (95% CI)×14 mm Hg.
(P<0.05): normal SBP; white samples; taking medication; 6 exercises; non‐BP outcome.
Normal and prehypertension SBP.
Indicates the moderator dimensions/levels that conferred the largest SBP reductions and were used to generate the additive SBP model.