Table 5.
Results of Subgroup Analyses in TLMA 2.
| K | #ES | g | 95% CI | p | p of Moderator | |
|---|---|---|---|---|---|---|
| Type of Sample | 0.65 | |||||
| Psychiatric | 11 | 30 | −0.09 | [−0.29, 0.11] | 0.40 | |
| Stressed and at-risk | 16 | 29 | −0.09 | [−0.36, 0.18] | 0.51 | |
| Healthy | 8 | 20 | 0.04 | [−0.27, 0.34] | 0.81 | |
| Type of MBI | 0.08 | |||||
| MBSR, MBCT, MBRP | 8 | 20 | −0.01 | [−0.22, 0.20] | 0.92 | |
| Adapted MBSR | 26 | 55 | −0.17 | [−0.42, 0.08] | 0.18 | |
| ACT | 1 | 4 | 0.32 | [−0.20, 0.84] | 0.22 | |
| Duration of MBI | 0.38 | |||||
| 8 weeks | 20 | 43 | −0.05 | [−0.21, 0.10] | 0.49 | |
| <8 weeks | 12 | 28 | −0.10 | [−0.35, 0.15] | 0.42 | |
| >8 weeks | 3 | 8 | −0.27 | [−0.68, 0.14] | 0.20 | |
| Biomarker | 0.20 | |||||
| Stress-related | 22 | 40 | −0.19 | [−0.34, −0.04] | 0.017 | |
| Inflammation-related | 15 | 33 | 0.18 | [−0.03, 0.39] | 0.09 | |
| Indirect | 6 | 6 | −0.001 | [−0.36, 0.35] | 0.99 | |
| Risk of bias | 0.50 | |||||
| Low risk | 6 | 15 | 0.004 | [−0.27, 0.28] | 0.97 | |
| Some concerns | 26 | 59 | −0.13 | [−0.44, 0.18] | 0.41 | |
| High risk | 3 | 5 | −0.29 | [−0.80, 0.22] | 0.26 | |
| Control group | 0.44 | |||||
| Active control | 18 | 42 | −0.07 | [−0.23, 0.09] | 0.37 | |
| Passive control | 17 | 38 | −0.09 | [−0.33, 0.14] | 0.44 |
Note. k = number of studies; #ES = number of effect sizes; g = Hedges g; CI = confidence interval; stress related = AA, ACTH, cortisol, DHEA-S; inflammation related = CRP, IL-6, IL-8, IL-1β, IL-1ra, IP-10, NF-κB, TNF-α; indirect = BDNF, EGF, LTL, NP-Y, SLC6A4 DNAm. The first subgroup in each moderator served as a baseline; entries for the respective other subgroups relate to deviations to this baseline.