Table 2.

Standardised regression coefficients of characteristics of studies on psychological treatment of depression: multivariate meta-regression analyses (k = 369)

		Full model			Parsimonious model
		Coeff	SE	p	Coeff	SE	p
Recruitment	Community	Ref.
	Clinical	−0.15	0.09	0.11
	Other	0.04	0.09	0.65
Target group	Adults	Ref.
	Older adults	0.04	0.11	0.73
	Student population	0.32	0.16	0.04
	Women with PPD	−0.14	0.12	0.27
	General medical	−0.04	0.11	0.73
	Other	0.05	0.11	0.65
Diagnosis v. cut-off		−0.14	0.07	0.04
Type	CBT	Ref.
	Third wave	0.15	0.15	0.32
	BAT	0.09	0.15	0.55
	Psychodynamic	−0.14	0.18	0.42
	IPT	−0.10	0.13	0.42
	PST	0.09	0.12	0.45
	Supportive	0.00	0.15	0.99
	Other	0.01	0.09	0.91
Format	Individual	Ref.
	Group	−0.03	0.08	0.70
	Guided self-help	0.02	0.11	0.87
	Other/mixed	−0.19	0.14	0.19
Number of sessions (continuous)		0.00	0.01	0.94
Control group	Waiting list	Ref.
	Care-as-usual	−0.12	0.09	0.17	−0.22	0.07	0.002
	Other	−0.27	0.09	0.003	−0.31	0.08	<0.001
Year (continuous)		−0.01	0.00	0.16
Western v. non-Western		0.46	0.11	<0.001	0.42	0.09	<0.001
Risk of bias (continuous)		−0.09	0.03	0.001	−0.11	0.02	<0.001
Intercept		12.76	8.17	0.12	1.08	0.07	<0.001
R2 analogue		0.24			0.26

BAT, behavioural activation therapy; CBT, cognitive behavioural therapy; Coeff, regression coefficient; HAM-D, Hamilton Rating Scale for Depression; IPT, interpersonal psychotherapy; p, this p-value indicates whether the regression coefficient of the subgroups differ significantly from the reference group; PPD, postpartum depression; PST, problem-solving therapy; Ref, reference group; SE, standard error.