Table 4.

Results of sensitivity analyses (all with response as outcome)

	Pairwise meta‐analyses						Network meta‐analysis
	N	RR	95% CI	I2	95% CI	Egger	RR	95% CI	χ2 (df), p
Optimized pharmacotherapy
Combined vs. psychotherapy	15	1.27	1.08‐1.49	53	0‐73	0.38	1.20	1.08‐1.35	1.20 (3), 0.75
Combined vs. pharmacotherapy	25	1.16	1.02‐1.33	52	15‐68	0.91	1.19	1.06‐1.33
Psychotherapy vs. pharmacotherapy	45	0.99	0.91‐1.08	44	14‐60	0.75	0.99	0.91‐1.08
Optimized psychotherapy
Combined vs. psychotherapy	9	1.33	1.18‐1.52	14	0‐60	0.42	1.25	1.08‐1.47	0.47 (3), 0.93
Combined vs. pharmacotherapy	19	1.27	1.06‐1.49	72	53‐81	0.33	1.27	1.10‐1.45
Psychotherapy vs. pharmacotherapy	30	1.00	0.90‐1.11	49	15‐66	0.44	1.00	0.89‐1.12
Low risk of bias (self‐report rated as low risk)
Combined vs. psychotherapy	6	1.33	0.98‐1.79	49	0‐78	0.24	1.45	1.16‐1.82	1.06 (3), 0.79
Combined vs. pharmacotherapy	17	1.32	1.02‐1.67	70	45‐80	0.88	1.27	1.05‐1.54
Psychotherapy vs. pharmacotherapy	16	0.87	0.76‐0.99	41	0‐66	0.08	0.87	0.74‐1.03
Low risk of bias (self‐report rated as high risk)
Combined vs. psychotherapy	5	1.52	1.32‐1.75	0	0‐64	0.59	1.43	1.15‐1.79	0.02 (3), 1.00
Combined vs. pharmacotherapy	13	1.27	0.96‐1.67	70	39‐81	0.87	1.23	1.02‐1.49
Psychotherapy vs. pharmacotherapy	14	0.88	0.79‐1.00	30	0‐62	0.04	0.86	0.74‐1.01
Placebo controlled excluded
Combined vs. psychotherapy	16	1.28	1.10‐1.49	50	0‐70	0.56	1.32	1.18‐1.47	0.66 (3), 0.88
Combined vs. pharmacotherapy	39	1.30	1.15‐1.47	58	37‐70	0.02	1.27	1.15‐1.41
Psychotherapy vs. pharmacotherapy	48	0.94	0.87‐1.03	40	10‐57	0.68	0.96	0.88‐1.05
One effect size per study
Combined vs. psychotherapy	19	1.25	1.09‐1.43	46	0‐67	0.40	1.25	1.12‐1.39	0.27 (3), 0.97
Combined vs. pharmacotherapy	43	1.27	1.12‐1.43	61	43‐71	0.02	1.25	1.14‐1.37
Psychotherapy vs. pharmacotherapy	50	0.99	0.90‐1.08	50	26‐63	0.45	1.00	0.92‐1.09

RR – relative risk

Significant results are highlighted in bold prints