TABLE 2.

Effects of psychological treatments for depressed college students compared with control groups: Hedges’ g^a

		N comp	g	95% CI	I2	95% CI	P b	NNT
All studies		22	0.89	0.66~1.11	57	23-72		2.13
Two outliers excluded		20	0.96	0.75~1.17	42	0~65		1.99
Only BDI		17	1.02	0.78~1.26	46	0~68		1.89
One ES per study (highest)		15	0.79	0.53~1.05	64	29~78		2.36
One ES per study (lowest)		15	0.70	0.48~0.93	53	0~73		2.63
Subgroup analyses
Compensation	Yes	6	0.53	0.18~0.88	71	0~85	0.01	3.42
	No	16	1.08	0.81~1.36	36	0~64		1.81
Recruitment	Announcements	13	0.96	0.65~1.26	52	0~73	0.38	1.99
	Clinical referrals	5	1.07	0.53~1.62	12	0~68		1.82
	Screening	4	0.63	0.17~1.08	79	4~90		2.91
Depression	Mood disorder	4	1.16	0.57~1.76	44	0~80	0.06	1.70
	Self-report	16	0.95	0.69~1.20	49	0~70		2.01
	Subthreshold	2	0.26	−0.31~0.82	75	xxx)		6.85
Psychological treatment	CBT	14	0.88	0.61~1.14	64	25~78	0.19	2.15
	BA	4	1.27	0.69~1.84	0	0~68		1.59
	Other	4	0.54	−0.01~1.08	40	0~79		3.36
Format	Group	14	0.66	0.43~0.90	50	0~71	0.003	2.78
	Individual	8	1.35	0.96~1.73	27	0~67		1.52
Control group	Waiting list	12	1.13	0.81~1.46	30	0~64	0.02	1.74
	Care-as-usual	6	0.90	0.55~1.25	54	0~80		2.10
	Other	4	0.33	−0.10~0.76	66	0~86		5.43

^*P < 0.05,

^**P < 0.01,

^***P < 0.001.

CI, confidence interval; N_comp, number of comparisons; NNT, numbers-needed-to-treat.

^aAccording to the random effects model.

^bThe P-values in this column indicate whether the difference between the effect sizes in the subgroups is significant.