Table 2 Meta-analytic and heterogeneity estimates for the overall effect, and sub-group moderation effects, of intervention efficacy

From: A meta-analysis of persuasive design, engagement, and efficacy in 92 RCTs of mental health apps

   

Effect size estimates

Heterogeneity estimates

  

k

Hedge’s g

SE

[95% CI]

p

Cochran’s Q

I2

Overall effect

92

−0.43

0.05

[−0.53, −0.34]

<001

Q (91) = 414.76, p <001

83.4

Sample type

 Non-clinical

28

−0.30

0.08

[−0.45, −0.15]

<001

Q(27) = 128.28, p < 001

82.8

 Sub-clinical

37

−0.49

0.07

[−0.62, −0.36]

<001

Q(36) = 132.07, p < 001

74.5

 Clinical

27

−0.49

0.12

[−0.72, −0.26]

<001

Q(26) = 138.97, p < 001

87.5

Study outcome

 Depression

29

−0.42

0.08

[−0.57, −0.27]

<001

Q(28) = 113.94, p < 001

79.2

 Anxiety

8

−0.71

0.15

[−1.01, −0.41]

<001

Q(7) = 20.87, p = 004

70.5

 Stress

21

−0.38

0.07

[−0.52, −0.24]

<001

Q(20) = 59.37, p < 001

67.5

 Social anxiety

4

−0.55

0.24

[−1.02, −0.08]

0.021

Q(3) = 19.89, p < 001

81.2

 PTSD

5

−0.28

0.10

[−0.48, −0.09]

0.005

Q(4) = 4.43, p = 351

16.8

 Body image/eating disorder

5

−0.73

0.15

[−1.03, 0.43]

<001

Q(4) = 9.89, p = 042

65.5

 Psychosis

2

−0.08

0.43

[−0.93, 76]

0.851

Q(1) = 2.46, p = 117

59.4

 Suicide/self-harm

3

−0.21

0.23

[−0.66, 24]

0.358

Q(2) = 3.68, p = 159

46.3

 Mental health (broadly)

2

0.43

0.55

[−0.64, 1.51]

0.433

Q(1) = 18.72, p < 001

94.7

 Postnatal depression

3

−1.08

0.92

[−2.88, 72]

0.240

Q(2) = 67.85, p < 001

98.7

Human support within intervention

 No

66

−0.44

0.06

[−0.55, −0.33]

<001

Q(65) = 307.33, p < 001

83.8

 Yes

26

−0.41

0.09

[−0.59, −0.23]

<001

Q(25) = 107.40, p < 001

82.7

Intervention type

 CBT (only)

34

−0.54

0.09

[−0.71, −0.37]

<001

Q(33) = 147.62, p < 001

82.7

 CBT (+others)

11

−0.37

0.12

[−0.62, −0.13]

0.003

Q(10) = 34.80, p < 001

72.4

 Mindfulness-based

18

−0.37

0.09

[−0.53, −0.20]

<001

Q(17) = 50.66, p < 001

73.8

 Psychoeducation

6

−0.53

0.14

[−0.81, −0.24]

<001

Q(5) = 29.58, p < 001

83.8

 Others

20

−0.33

0.13

[−0.58, −0.07]

0.012

Q(19) = 112.13, p < 001

88.5

  1. k number of studies included.