Table 4.

Subgroup analyses: depression treatments.

Subgroup analysisa		Effects			Heterogeneity				Subgroup analysis
		Nb	g	95% CI		I2	P value	I2 95% CI		Q value	P value (Q)
Recruitment pathwayc
	Clinical and community+clinical	8	1.05	0.95-1.14		78	<.001	56-89		7.253	.007
	Community	5	1.38	1.16-1.59		96	<.001	94-98		7.253	.007
Specific treatment
	Mixed treatment	7	1.10	0.98-1.22		93	<.001	87-96		0.736	.39
	Disorder-specific treatment	6	1.27	0.91-1.62		97	<.001	95-98		0.736	.39
Diagnosis conductedd
	Interview	7	1.127	0.89-1.35		97	<.001	95-98		0.946	.33
	Questionnaire	5	1.25	1.16-1.34		81	<.001	57-92		0.946	.33
Clinical cutoff/minimal symptom severity
	Yes	4	1.27	1.08-1.45		84	<.001	60-94		0.897	.34
	No	7	1.17	0.98-1.35		96	<.001	94-97		0.897	.34
Treatment duration
	<9 weeks	5	1.17	1.01-1.32		95	<.001	92-97		7.485	.02
	9-13 weeks	4	1.00	0.95-1.05		0	.85	0-85		7.485	.02
	>13 weeks	4	1.37	1.00-1.74		97	<.001	94-98		7.485	.02
Guide cognitive behavioral therapy training (profession)e
	Nonprofessional	4	0.92	0.79-1.05		22	.28	0-88		14.151	<.001
	Other	9	1.27	1.14-1.40		96	<.001	94-97		14.151	<.001
Guide supervision provided
	No	4	0.91	0.75-1.08		39	.18	0-79		10.339	<.001
	Yes	9	1.27	1.13-1.41		96	<.001	94-97		10.339	<.001
Guide training provided
	No	6	0.98	0.94-1.02		7	.37	0-76		21.368	<.001
	Yes	7	1.35	1.20-1.51		95	<.001	91-97		21.368	<.001
Intervention manual provided
	No	9	1.039	0.95-1.137		75	<.001	51-87		10.715	<.001
	Yes	4	1.467	1.23-1.71		97	<.001	95-98		10.715	<.001
Risk of bias—researcher allegiance
	High	7	1.252	1.08-1.42		97	<.001	95-98		1.347	.25
	Low	5	1.119	0.99-1.29		70	.01	23-88		1.347	.25
Risk of bias—confounding (treatment inclusion)
	High	7	1.215	1.06-1.43		96	<.001	94-98		0.985	.32
	Low	5	1.105	0.98-1.23		82	<.001	58-92		0.985	.32

^aTest against “Guidance format: face-to-face vs written guidance,” “Guidance modality: Message, Email, Telephone, F2F,” and “Guide profession” excluded, as there were too few studies included in analysis.

^bNumber of studies.

^cOnly two studies included via the clinical pathway only. We combined the categories “Both, community and clinical” and “clinical” for this analysis.

^dExcluding one study [54], as this is the only study using clinical judgment without specifying the use of an interview or questionnaire.

^eWe grouped all studies involving guides not specifically trained in delivering cognitive behavioral therapy in the category “non-professional” and studies involving psychiatrists, psychologists, or psychotherapists in their guidance in the category “other.”