Yoga, Meditation, Breathing Exercises, and Inflammatory Biomarkers with Possible Implications in COVID-19: A Systematic Review and Meta-Analysis of Randomized Controlled Trials
Table 5
Sensitivity analysis.
Parameters
Mean difference [95% CI]
Heterogeneity
Overall Effect
Q
I2 (%)
Z
- Value
IL-6
Removing (3) A1.2 studies
−0.42 [−1.34, 0.49]
115.80
96
0.91
0.36
Removing (1) B1.3 studies
−0.53 −1.28, 0.21]
165.60
96
1.40
0.16
Removing (3) C1.2 studies
−0.67 [−1.31, −0.04]
15.36
67
2.09
0.04
Removing (2) C1.3 studies
−0.71 [−1.60, 0.17]
171.86
97
1.58
0.11
TNF-α
Removing (2) A1.2 studies
−1.41 [−3.81, 1.00]
48.33
96
1.15
0.25
Removing (1) B1.3 studies
−2.53 [−4.20, −0.86]
58.10
95
2.96
0.003
Removing (2) C1.2 studies
−4.19 [−5.29, −3.09]
0.75
0
7.44
<0.00001
Cortisol
Removing (1) B1.3 studies
−25.80 [−78.33, 26.73]
63.15
98
0.96
0.34
Sensitivity analysis, showing progressive effects on pooled mean differences of removing data by trials’ risk of bias rating. Interstudy heterogeneity was tested by using the Cochran Q statistic (Chi2) at a significance level of and quantified by the I2 statistic.