Participants
The baseline characteristics of the study sample can be found in the supplementary file as Table A. There was a significant difference in age between the groups. Specifically, participants in the CG were older than those in the IG (p=0.002, dunb=0.73 [0.28, 1.19]). The adherence rate during the intervention period was established at 97.60%.
Physical and Cognitive Function
At baseline, the CG exhibited significantly higher values compared to the IG in the following measures: chair-stand (p=0.003, dunb=0.68 [0.23, 1.14), chair-stand velocity (p=0.003, dunb=0.68 [0.23, 1.14]), HDS test for the dominant side (p=0.010, dunb=0.59 [0.15, 1.05]), HDS test for the non-dominant side (p=0.026, dunb=0.51 [0.07, 0.96]), and in MMSE (p=0.002, dunb=0.71 [0.26, 1.17]). While the IG presented significantly higher values in the TUG test (p=0.020, dunb=-0.53 [-0.98, -0.09]).
The ANCOVA results, summarized in Table 1, demonstrate highly significant effects of the group factor in favor of the IG after the intervention program across all measures.
Table 1: Analysis of covariance (ANCOVA) results considering the group factor for cognitive and physical function.
Variables
|
Control Group
|
Intervention Group
|
ANCOVA Effects
|
Pre
|
Post
|
Δ (95% CI)
|
Pre
|
Post
|
Δ (95% CI)
|
F
|
p
|
η²p
|
Cognitive Function
|
|
|
|
|
|
|
|
|
|
|
MMSE (score)
|
28.79 ± 1.49
|
28.10 ± 1.64 *
|
-0.69 [-1.15 to -0.23]
|
27.03 ± 3.15
|
29.07 ± 0.98 *,§
|
2.05 [1.17 to -2.29]
|
25.099
|
<0.001
|
0.248£
|
Physical Function
|
|
|
|
|
|
|
|
|
|
|
Chair Stand (rep)
|
15.33 ± 4.16
|
14.23 ± 4.36 *
|
-1.10 [-1.70 to -0.50]
|
12.83 ± 3.09
|
24.32 ± 5.36 *,§
|
11.50 [10.0 to 12.99]
|
212.061
|
<0.001
|
0.736£
|
Chair Stand V. (m/s)
|
0.59 ± 0.17
|
0.58 ± 0.17
|
-0.02 [-0.04 to 0.01]
|
0.49 ± 0.11
|
0.77 ± 0.12 *,§
|
0.27 [0.24 to 0.31]
|
183.315
|
<0.001
|
0.707£
|
TUG Test (sec)
|
6.10 ± 1.39
|
6.60 ± 1.64 *
|
0.49 [0.26 to 0.73]
|
6.75 ± 0.98
|
5.28 ± 0.68 *,§
|
-1.47 [-1.69 to -1.25]
|
132.081
|
<0.001
|
0.635£
|
SMBT (m)
|
2.29 ± 0.70
|
2.19 ± 0.73
|
-0.09 [-0.23 to 0.05]
|
2.05 ± 0.56
|
2.51 ± 0.62 *,§
|
0.46 [0.36 to 0.56]
|
37.575
|
<0.001
|
0.331£
|
6MWT (m)
|
501.93 ± 90.32
|
483.81 ± 94.19 *
|
-18.12 [-32.18 to -4.07]
|
483.71 ± 60.89
|
541.49 ± 65.52 *,§
|
57.77 [41.26 to 74.29]
|
47.779
|
<0.001
|
0.386£
|
HDS test, DS (kg)
|
27.87 ± 9.53
|
26.79 ± 9.75 *
|
-1.08 [-1.96 to -0.20]
|
22.99 ± 6.37
|
27.95 ± 8.16 *,§
|
4.96 [3.73 to 6.19]
|
61.656
|
<0.001
|
0.448£
|
HDS test, Non-DS (kg)
|
25.90 ± 9.67
|
25.50 ± 9.50
|
-0.40 [-1.21 to 0.41]
|
21.63 ± 6.69
|
26.15 ± 7.42 *,§
|
4.53 [3.46 to 5.59]
|
48.215
|
<0.001
|
0.388£
|
|
Abbreviations: MMSE, mini mental score examination; rep, repetitions; V, velocity; m/s, meters per second; sec, seconds; m, meters; kg, kilograms; SMBT, seated medicine ball throw; 6MWT, six-minute walking test; HDS, handgrip strength; DS, dominant side;
Pre- and post-values data are presented as mean and standard deviation, whereas mean difference as mean and 95% confidence interval.
Values in bold represent significant differences at p<0.05.
*, p<0.05 vs. pre-values.
§, p<0.05 vs. Control Group’s Δ.
ηp2 values thresholds:
&, small effect: ηp²= 0.010 to 0.059;
#, medium effect: ηp²= 0.060 to 0.140;
£, large effect large: ηp²> 0.140.
|
To complement the study results, Figure 1 displays Cohen’s dunb (∆ changes: post- minus pre-values) between groups for cognitive and physical functions. Remarkably, IG exhibited significant improvements in all measures: MMSE (t(77)=5.58, p<0.001, dunb=1.24 [0.77, 1.73]), HDS test for the non-dominant side (t(77)=7.42, p<0.001, dunb=1.65 [1.15, 2.18]), HDS test for the dominant side (t(77)=8.04, p<0.001, dunb=1.79 [1.28, 2.33]), 6MWT (t(77)=7.07, p<0.001, dunb=1.58 [1.08, 2.09]), SMBT (t(77)=6.46, p<0.001, dunb=1.44 [0.95, 1.95]), TUG test (t(77)=-12.22, p<0.001, dunb=-2.72 [-3.37, -2.13]), chair stand velocity (t(77)=14.74, p<0.001, dunb=3.28 [2.63, 3.99]), and in chair stand (t(77)=15.72, p< 0.001, dunb=3.502 [2.82, 4.24]).
Lastly, following the intervention, there were significant within-IG increases in MMSE (t(39)=4.74, p<0.001, dunb=0.86 [0.46, 1.29]), chair stand (t(39)=15.59, p<0.001, dunb=2.58 [1.96, 3.29]), chair stand velocity (t(39)=17.87, p<0.001, dunb=2.30 [1.76, 2.92]), SMBT (t(39)=9.43, p<0.001, dunb=0.76 [0.53, 1.00]), 6MWT (t(39)=7.08, p<0.001, dunb=0.89 [0.59, 1.23]), and HDS test for the dominant side (t(39)=8.14, p<0.001, dunb=0.66 [0.45, 0.89]), as well as for the non-dominant side (t(39)=8.61, p<0.001, dunb=0.63 [0.44, 0.84]). Conversely, significant decreases were observed in the TUG test (t(39)=-13.33, p<0.001, dunb=-1.71 [-2.20, -1.28]).
For the CG, there were no significant changes in chair stand velocity, SMBT and HDS test for the non-dominant side, however, there were significant reductions in MMSE (t(38)=-3.05, p=0.004, dunb=-0.43 [-0.74, -0.14]), chair stand (t(38)=-3.73, p=0.001, dunb=-0.25 [-0.41, -0.11]), 6MWT (t(38)=-2.61, p=0.013, dunb=-0.19 [-0.35, -0.04]); and HDS test for the dominant side (t(38)=-2.49, p=0.017, dunb=-0.11 [-0.20, -0.02]). Finally, there was a significant increase in the TUG test (t(38)=4.23, p<0.001, dunb=0.32 [0.16, 0.49]). Figure 2 illustrates the Cohen’s dunb pre-post differences after intervention for all measures in both groups. Additionally, visual representations of variations and differences in mean values for peak torque before and after the intervention are depicted in Figure 3.