Skip to main content
. Author manuscript; available in PMC: 2022 Jan 1.
Published in final edited form as: J Affect Disord. 2020 Sep 23;278:542–555. doi: 10.1016/j.jad.2020.09.071

Table 2.

Summary of meta-analysis results (overall).

Outcome Random effects model Corrected for publication bias z p-value I2 k
Response RR = 2.0382 [1.5748; 2.6380] RR = 1.4209 [1.0950; 1.8438] 5.41 < 0.0001 62.7% 31
Remission RR = 2.0029 [1.5005; 2.6735] RR = 1.5521 [1.1472; 2.1000] 4.71 < 0.0001 38.5% 24
Score SMD = −1.1430 [−1.4613; −0.8247] SMD = −0.4832 [−0.8453; −0.1212] −7.04 < 0.0001 89.8% 31
Suicidality SMD = −0.3867 [−0.7082; −0.0653] SMD = −0.5034 [−0.8180; −0.1888] −2.36 0.0184 71.3 12
Completion RR = 0.9929 [0.9681; 1.0182] RR = 0.9876 [0.9576; 1.0185] −0.56 0.5773 14.0 31
Dropouts RR = 0.9664 [0.7234; 1.2911] RR = 0.9229 [0.6864; 1.2410] −0.23 0.8173 40.5 24
Adverse events RR = 1.8703 [1.0271; 3.4076] RR = 2.0087 [1.1150; 3.6188] 2.05 0.0406 0.0 14

RR = rate ratio; SMD = standardized mean difference; z = z-score (on normal distribution); I2 = measure of heterogeneity (closer to 100.0 indicates higher heterogeneity); k = number of trials involved in the sub analysis.