Table 4 VR-enhanced therapy vs. active condition contrast, post-treatment.

From: The effectiveness of virtual reality based interventions for symptoms of anxiety and depression: A meta-analysis

Anxiety symptoms

 

N

ga

95% CI

I 2

I2 95% CI

NNT

pb

All studies

 

29

−0.02

−0.14 to 0.10

20

0 to 50

83.33

 

Outliers excludedc

 

27

−0.02

−0.13 to 0.08

0

0 to 43

83.33

 

Only studies with >25 randomized per arm

 

12

−0.05

−0.19 to 0.07

1

0 to 59

35.71

 

Only studies involving anxiety disorders

 

23

−0.10

−0.24 to 0.04

26

0 to 55

17.86

 

Subgroup analyses d

Country

N. America

9

−0.12

−0.31 to 0.06

0

0 to 65

14.71

0.198

 

EU

18

0.04

−0.13 to 0.23

39

0 to 65

45.45

 

VR program author

N

8

0.09

−0.21 to 0.40

54

0 to 79

20.00

0.372

among author poole

Y

20

−0.05

−0.18 to 0.07

0

0 to 48

35.71

 

Recruitmentf

Army

2

−0.32

−0.64 to −0.005

0

N/Ai

5.56

 
 

Clinic

7

0.03

−0.17 to 0.23

0

0 to 71

62.50

0.159

 

Community

17

0.001

−0.19 to 0.19

43

0 to 68

1772.4

 

Control group

CBT

18

0.03

−0.09 to 0.16

0

0 to 50

62.50

0.120

 

IE

6

−0.16

−0.41 to 0.08

0

0 to 75

11.11

 
 

IVE

4

−0.35

−0.78 to 0.07

49

0 to 83

5.10

 

Experimental

VRCBT

17

0.09

−0.04 to 0.24

3

0 to 53

20.00

0.016

intervention

VRE

12

−0.18

−0.35 to −0.006

11

0 to 50

9.80

 

Type of anxiety disorder

Flight anxiety

7

0.21

−0.12 to 0.54

41

0 to 75

8.47

0.206

 

Panic disorder

6

−0.05

−0.32 to 0.21

0

0 to 75

35.71

 
 

PTSD

2

−0.32

−0.64 to −0.005

0

N/A

5.56

 
 

Social anxiety

5

−0.18

−0.52 to 0.15

41

0 to 78

9.80

 
 

Specific phobia

3

−0.19

−0.57 to 0.17

14

0 to 91

9.43

 

Incomplete outcome data RoBj

High/Unclear

20

0.02

−0.11 to 0.15

5

0 to 50

83.33

0.326

 

Low

9

−0.12

−0.36 to 0.12

43

0 to 74

14.71

 

Depressive symptoms

All studies

 

13

0.004

−0.20 to 0.21

26

0 to 62

443.11

 

Outliers excludedg

 

12

0.07

−0.10 to 0.25

0

0 to 58

25.00

 

Only studies with >25 randomized participants per arm

 

4

−0.03

−0.27 to 0.20

0

0 to 85

62.5

 

Subgroup analyses

Country

N. America

3

0.14

−0.19 to 0.48

0

0 to 90

12.82

0.410

 

EU

9

−0.04

−0.32 to 0.24

40

0 to 72

45.45

 

VR program author

N

2

0.03

−0.37 to 0.43

0

N/A

62.50

0.901

among author pool

Y

11

−0.001

−0.24 to 0.24

37

0 to 69

1772.4

 

Recruitmenth

Clinic

6

−0.01

−0.25 to 0.23

0

0 to 75

166.67

0.769

 

Community

4

−0.12

−0.79 to 0.55

76

32 to 91

14.71

 

Control group

CBT

10

0.08

−0.10 to 0.28

0

0 to 62

21.74

0.777

 

IE

2

0.02

−0.39 to 0.43

0

N/A

83.33

 

Experimental

VRCBT

8

0.17

−0.07 to 0.43

0

0 to 68

10.42

0.126

intervention

VRE

5

−0.18

−0.57 to 0.20

62

0 to 86

9.80

 

Incomplete outcome data RoB

High/Unclear

9

0.08

−0.11 to 0.29

0

0 to 65

21.74

0.308

 

Low

4

−0.25

−0.88 to 0.37

70

15 to 90

7.14

 
  1. aAll results are reported with Hedges’ g, using a random effects model. Negative effect indicates superiority of the active interventions over the VR-enhanced therapies.
  2. bThe p levels in this column indicate whether the difference between the ESs in the subgroups is significant. (significant results are marked with italic).
  3. cOutliers: Kampmann, 2016 St.2; Muhlberger, 2003.
  4. dSubgroup analysis were conducted using a mixed effects model. Only subgroups with at least 2 studies were included.
  5. eOne study (Wiederhold, 2001) did not contain information about this moderator.
  6. fTwo studies (Meyerbroeker, 2013, Robillard, 2010 St2) did not contain information about this moderator.
  7. gOutliers: Kampmann, 2016 St.2.
  8. hOne study (Robillard, 2010 St.2) did not contain information about this moderator.
  9. iConfidence intervals around I2 cannot be calculated if there are less than 3 groups.
  10. jRoB: Risk of Bias.