Skip to main content
. 2020 Jul 13;22(7):e16021. doi: 10.2196/16021

Table 1.

Characteristics of the included studies (n=12).

Characteristics Number of studies
Study design

Quasiexperiment 6

Randomized controlled trial 6
Type of publication

Journal article 8

Conference proceedings 3

Thesis 1
Country

United States 4

Japan 1

Sweden 1

Turkey 1

Australia 1

United Kingdom 1

China 1

Romania, Spain, and Scotland 1

Global population 1
Year of publication

2018 7

2017 2

2016 1

2015 2
Sample size

<100 6

100-200 3

>200 1
Age (years), mean (range)a 31.3 (22-45)
Sex (male), %b 35
Sample type

Clinical sample 6

Nonclinical sample 6
Settingc

Community 6

Educational 4

Clinical 3
Intervention purpose

Therapy 10

Self-management 2
Intervention platform

Web-based 6

Standalone software 6
Intervention response generation

Rule-based 8

Artificial intelligence 4
Intervention dialogue initiative

Chatbot 9

Both 3
Intervention input modality

Written 9

Spoken 2

Written and spoken 1
Intervention output modality

Written 6

Written, spoken, and visual 3

Spoken and visual 2

Written and visual 1
Embodiment

Yes 6

No 6
Targeted disordersd

Depression 7

Anxiety 4

Any mental disorder 3

Acrophobia 1

Stress 1
Comparator

Pretest vs posttest

No intervention 4

Education 3

High users vs low users 1
Measured outcomese

Severity of depression 6

Psychological wellbeing 3

Severity of anxiety 3

Positive and negative affect 2

Distress 2

Stress 2

Safety 2

Severity of acrophobia 1
Measuresf

PHQ-9g 4

GAD-7h 2

PANASi 2

K10j 2

PSS-10k 2

AQl 1

HAD-Sm 1

OASISn 1

WHO-5-Jo 1

HIQp 1

BDI-2q 1

Adverse events 2
Follow-up periodr

2 weeks 6

4 weeks 6

6 weeks 1

12 weeks 1

aMean age was reported in 10 studies.

bSex was reported in 9 studies.

cNumbers do not add up as one study recruited the sample from more than one setting.

dNumbers do not add up as 4 chatbots focused on both depression and anxiety.

eNumbers do not add up as most studies assessed more than one outcome.

fNumbers do not add up as some studies used more than one tool to assess a single outcome and several studies have more than one outcome.

gPHQ-9: Patient Health Questionnaire.

hGAD-7: Generalized Anxiety Disorder scale.

iPANAS: Positive and Negative Affect Schedule.

jK10: Kessler Psychological Distress Scale.

kPSS-10: Perceived Stress Scale.

lAQ: Acrophobia Questionnaire.

mHAD-S: Hospital Anxiety and Depression Scale.

nOASIS: Overall Anxiety Severity and Impairment Scale.

oWHO-5-J: World Health Organization-5 Well-Being Index.

pHIQ: Heights Interpretation Questionnaire.

qBDI-2: Beck Depression Inventory II.

rNumbers do not add up as two studies assessed outcomes at 2 different points of time.