Condition category
Mental and Behavioural Disorders
Date applied
31/10/2018
Date assigned
19/11/2018
Last edited
19/11/2018
Prospective/Retrospective
Retrospectively registered
Overall trial status
Completed
Recruitment status
No longer recruiting

Plain English Summary

Background and study aims
Students in need of mental health care face many barriers including cost, location, availability, and stigma. Studies show that digital mental health solutions offer a less intensive and more cost-effective alternative for reducing symptoms of depression and anxiety. Artificial intelligence (AI) offers a scalable solution as the demand for affordable, convenient, lasting, and secure support grows. The aim of this study is to assess the feasibility of using an AI chatbot called Tess to deliver emotional support and reduce self-identified symptoms of depression and anxiety in college students.

Who can participate?
University students enrolled in the USA are invited to participate regardless of gender, race, or socioeconomic status. As healthy volunteers, students are expected to speak English, be over the age of 18, and to sign the informed consent in order to participate.

What does the study involve?
Participants are randomly allocated to one of three groups. The control group receive an e-book with no access to the AI, Tess, until after the study has ended. Test group 1 receive unlimited access to Tess with daily check-ins for two weeks. Test group 2 receive unlimited access to Tess with bi-weekly check-ins for four weeks.

What are the possible benefits and risks of participating?
By participating in this study, students gained free access to a unique and on-demand emotional support service that usually costs $5.00 month. Students are also offered an incentive for participating in the form of a $20 Amazon e-gift card upon completion of the study. Students are offered a secure and confidential space to talk about sensitive issues they may otherwise have a hard time sharing with others. While there were no clinical risks observed in the results of this study, one potential risk may have been that interacting with the AI system, Tess, may lead a student to desire more support in a clinical setting which may require financial and other resources to achieve.

Where is the study run from?
The study was conducted by primary investigator, Russell Fulmer, PhD from Northwestern University in Illinois, USA. Dr. Fulmer collaborated with the vendor called X2 AI Inc., which funded and provided the AI service called Tess to deliver support to participants during the study.

When is the study starting and how long is it expected to run for?
July 2017 to October 2017

Who is funding the study?
X2 AI Inc. (USA)

Who is the main contact?
Dr Russell Fulmer
russell.fulmer@northwestern.edu

Trial website

https://www.facebook.com/ChatWithTess/

Contact information

Type

Scientific

Primary contact

Ms Angela Joerin

ORCID ID

http://orcid.org/0000-0003-3667-9636

Contact details

995 Market St. Suite 209
San Francisco
94103
United States of America

Additional identifiers

EudraCT number

ClinicalTrials.gov number

Protocol/serial number

10.2196/mental.9782

Study information

Scientific title

Using psychological artificial intelligence (Tess) to relieve symptoms of depression and anxiety: randomized controlled trial

Acronym

Tess RCT

Study hypothesis

This study aimed to assess the feasibility and efficacy of using an integrative psychological AI, Tess, to reduce self-identified symptoms of depression and anxiety in college students.

Ethics approval

This study was considered exempt from ethics approval and registration in a public trials registry, since it involved a nonclinical population of college students. The support delivered by the artificial intelligence service called Tess does not qualify as a clinical intervention. Participants in the control group received only a brief self-help e-book written by the National Institute of Mental Health (NIMH). Program evaluation outcomes from this study will support content enhancements for future IRB approved studies.

Study design

Randomized controlled trial

Primary study design

Interventional

Secondary study design

Randomised controlled trial

Trial setting

Internet

Trial type

Prevention

Patient information sheet

Condition

Providing on-demand support for students with self-identified, and non-clinical range, symptoms of depression and anxiety

Intervention

Confirmed participants were randomized via a computer algorithm that automatically generated a number between 0 and 2, and then placed into one of three groups.
No Intervention: Control Group participants were sent a link to NIMH's self-help eBook.
Test Group 1 participants were granted unlimited access to the Tess chatbot through Facebook messenger for 2 weeks.
Test Group 2 participants were granted unlimited access to the Tess chatbot through Facebook messenger for 4 weeks.

Intervention type

Behavioural

Phase

Drug names

Primary outcome measure

1. Depression severity assessed using the Patient Health Questionnaire-9 (PHQ9) at baseline, 2 weeks, and 4 weeks for the control group, at baseline and 2 weeks for Test Group 1 and at baseline and 4 weeks for Test Group 2
2. Generalized anxiety disorder severity assessed using Generalized Anxiety Disorder-7 (GAD7) questionnaire at baseline, 2 weeks, and 4 weeks for the control group, at baseline and 2 weeks for Test Group 1 and at baseline and 4 weeks for Test Group 2
3. Positive and negative mood assessed using the Positive and Negative Affect Schedule (PANAS) questionnaire at baseline, 2 weeks, and 4 weeks for the control group, at baseline and 2 weeks for Test Group 1 and at baseline and 4 weeks for Test Group 2

Secondary outcome measures

User satisfaction assessed using a survey at week 4 for the control group, week 2 for Test Group 1 and at week 4 for Test Group 2

Overall trial start date

01/07/2017

Overall trial end date

16/10/2017

Reason abandoned (if study stopped)

Eligibility

Participant inclusion criteria

1. Current enrollment at a university in the United States
2. Aged 18 years and older (screened at the first level via checkbox confirmation)
3. Able to read English (implied)
4. Informed consent signed
5. To guard against compromise, for example, from malicious bots, all potential participants were sent an email requesting that they respond using their university email denoting their confirmation

Participant type

Healthy volunteer

Age group

Adult

Gender

Both

Target number of participants

115

Participant exclusion criteria

Does not meet inclusion criteria

Recruitment start date

01/08/2017

Recruitment end date

29/08/2017

Locations

Countries of recruitment

United States of America

Trial participating centre

Northwestern University
Evanston
60201
United States of America

Trial participating centre

X2 AI Inc.
995 Market St. Unit 209
San Francisco
94103
United States of America

Sponsor information

Organisation

X2 AI Inc.

Sponsor details

995 Market St. Suite 209
San Francisco
94103
United States of America
+1 (0)2489356366
angie@x2ai.com

Sponsor type

Industry

Website

https://www.x2ai.com/

Organisation

Northwestern University

Sponsor details

633 Clark Street
Evanston
60208
United States of America

Sponsor type

University/education

Website

https://www.northwestern.edu/

Funders

Funder type

Industry

Funder name

X2 AI Inc

Alternative name(s)

Funding Body Type

Funding Body Subtype

Location

Results and Publications

Publication and dissemination plan

This study has been accepted for publication through JMIR and is available for download here: https://preprints.jmir.org/preprint/9782/accepted

IPD sharing statement
The datasets generated during and/or analysed during the current study are/will be available upon request from Angie Joerin (angie@x2ai.com). Type of data: SPSS output. Data requests may take up to 90 days to process and may be shared only to those who meet the criteria defined below. Data may be shared only through secure, password protected platforms and only to researchers or organizations who submit a written request with aim to further evaluate the feasibility, generalizability, efficacy or other of using the artificial intelligence system called Tess to deliver emotional support. The X2 ethical board and Northwestern University primary investigator maintain the right to refuse access to data. Participants signed an electronic consent form and agreed to the services privacy policy (https://www.x2ai.com/privacy) before participating in the study. Participants were randomly selected into three groups and automatically assigned a unique user ID. The system automatically de-identified transcripts, which were stored on a HIPAA compliant server in which only a select few individuals from the core research team were granted security level access during the duration of the study. The terms of use for the system used to deliver support during this study are described in this link: https://www.x2ai.com/terms.

Intention to publish date

29/08/2018

Participant level data

Available on request

Basic results (scientific)

Publication list

Publication citations

Additional files

Editorial Notes