Trial Review

The safety and scientific validity of this study is the responsibility of the study sponsor and investigators. Listing a study does not mean it has been endorsed by the ANZCTR. Before participating in a study, talk to your health care provider and refer to this information for consumers
Trial registered on ANZCTR


Registration number
Help: Registration number
ACTRN12626000908336
Ethics application status
Help: Ethics application status
Approved
Date submitted
Help: Date submitted
30/06/2026
Date registered
Help: Date registered
23/07/2026
Date last updated
Help: Date last updated
23/07/2026
Date data sharing statement initially provided
Help: Data sharing statement date
23/07/2026
Type of registration
Help: Type of registration
Prospectively registered

Titles & IDs
Public title
Evaluating an Artificial Intelligence Mental Health Intervention (AIMHI) for Adults with Anxiety and Depression Symptoms: A Randomised Controlled Trial.
Scientific title
Evaluating the Cara Plus App: A Randomised Controlled Trial of an Artificial Intelligence Mental Health Intervention (AIMHI) for Adults with Generalised Anxiety and Depression Symptoms.
Additional trial identifiers [1] 317514 0
Nil
Universal Trial Number (UTN)
Trial acronym
AIMHI
Related trial records

Health condition
Health condition(s) studied:
Generalised Anxiety 342254 0
Depression 342255 0
Condition category
Condition code
Mental Health 338366 338366 0 0
Anxiety
Mental Health 338369 338369 0 0
Depression

Intervention/exposure
Study type
Interventional
Description of the intervention or exposure
Intervention:

The intervention is Cara Plus, an AI-enabled mental health smartphone application developed by Spectrum Wellness Limited (Spectrum.Life), intended to support the management of mild to moderate symptoms of anxiety and/or depression in adults. The app delivers support through secure, text-based chat with an AI companion. Users can choose a preferred personality style for the AI companion, which adjusts the conversational tone while keeping the underlying model unchanged. During chat, the AI companion responds to users and guides the conversation, offering short interactive evidence-based techniques (such as breathing exercises or guided meditations) when appropriate. The AI retains memory of prior sessions, enabling personalised support over time. There is no human involvement in the chat. If any message suggests possible risk or distress, the chat pauses and the user is prompted to contact appropriate crisis support services. Participants will install the app through a link to Google Play. They can initiate text conversations with the app at any time that is convenient for them. Participants will also receive reminder messages through the Prolific platform twice per week.

Delivery of the Intervention:
The intervention is delivered autonomously by the Cara Plus AI, which was developed with input from clinical psychologists, lived experience experts, and independent specialists to ensure a therapeutically sound and user-centred experience. The trial is led by Senior Lecturers in Clinical Psychology at the University of the Sunshine Coast, Australia. The Cara Plus application was developed and tested by Spectrum Wellness Limited, who will provide technical support throughout the trial.

The intervention is delivered entirely online, on an individual basis, via the Cara Plus smartphone application on Android devices (operating system version 8 or higher). All study procedures, including recruitment, screening, consent, randomisation, assessments, and the intervention itself are conducted entirely online. Participants are recruited via the Prolific research platform, with recruitment occurring across Australia and the United States. Participants access the app using a direct download link and a unique login code provided at enrolment. Participants engage with the AI through interactive text chat (with optional voice-to-text), supplemented by in-app exercises. No face-to-face, group, or video-based components are included.

The intervention is delivered over 8 weeks. Participants in the intervention group are advised to use the app for a minimum of 5 minutes per day, 5 days per week, amounting to a minimum of 200 minutes of application use across the study period. Assessments are conducted at four time points: baseline (Week 0), Week 2, Week 4, and Week 8.

Adherence:
Participants are provided with a clear usage goal at enrolment (minimum 5 minutes/day, 5 days/week for 8 weeks) and receive a brief schedule of assessments so they know when surveys are due. Adherence is supported through reminders sent twice a week via the Prolific platform, prompting participants to continue engaging with the app between assessment time points.

App usage data, including frequency, duration, and engagement with specific features are collected by the development team and provided to the research team in de-identified form, enabling monitoring of engagement patterns.

Participants in the intervention arm also receive a safety disclaimer at onboarding, and automated safety messaging is triggered if distress indicators are detected during app use.
If distress or risk is detected, participants will be directed to the relevant crisis support services in their country:
Australia: Lifeline on 13 11 14. In an emergency, participants will be advised to call 000.
United States: Suicide and Crisis Lifeline by calling or texting 988. In an emergency, participants will be advised to call 911.

The same progressive compensation structure applied to both groups (£8/hour at baseline to Week 2; £9/hour at Weeks 2–4; £10/hour at Weeks 4–8), providing a financial incentive to remain engaged with the study across the full 8-week period.
Compensation will be provided through Prolific. The researcher sets the payment amount in British pounds. Participants receive the payment in their local currency, such as Australian dollars or United States dollars, using the applicable exchange rate on the date of conversion.
Compensation for time spent using the app will be paid in addition to compensation for the time spent completing study assessments.
Intervention code [1] 334310 0
Behaviour
Description of the comparator or control
Control Group:
The control condition is a static psychoeducational resource in which participants are provided with a link to reputable, publicly available online mental health self-help materials. Australian participants are directed to the Beyond Blue anxiety and depression information pages (https://www.beyondblue.org.au/mental-health/anxiety; https://www.beyondblue.org.au/mental-health/depression), with equivalent resources provided to participants in the United States via the National Institute of Mental Health (NIMH) website (https://www.nimh.nih.gov/health/topics/anxiety-disorders; https://www.nimh.nih.gov/health/publications/depression). These materials cover psychoeducation about anxiety and depression but involve no interactive, adaptive, or AI-driven components, no chatbot, and no personalised guidance. This condition is designed to control for the general effect of receiving health information and study attention, while isolating the active components of the Cara Plus app.

Delivery of the Control:
The control resources are pre-existing, publicly available materials produced and maintained by established mental health organisations (Beyond Blue) and government research institutes (NIMH). No member of the research team delivers content to control participants. Participants access the materials independently, without facilitation by a clinician, coach, or researcher.

The control resource is delivered entirely online, on an individual, self-directed basis. Participants receive links to the relevant psychoeducational pages and directions on recommended use. There is no interactive, adaptive, or conversational component, and no face-to-face, group, or video-based contact occurs at any point.

The control resource is used over the same 8-week period as the intervention. Participants are asked to engage with the resources for a minimum of 10 minutes from Week 0 to Week 2, 10 minutes from Week 2 to Week 4, and 10 minutes from Week 4 to Week 8 — a minimum total of 30 minutes of resource use across the study period. Assessments are completed at the same four time points as the intervention group: baseline (Week 0), Week 2, Week 4, and Week 8.

The control resources are accessed remotely via participants' own internet-connected devices, in any location of their choosing. As with the intervention arm, all study procedures are conducted entirely online.

Adherence:
No formal monitoring of control participants' engagement with the psychoeducational resources is conducted, consistent with a treatment-as-usual design and standard practice in digital mental health RCTs. This approach preserves ecological validity by allowing participants to engage with the materials as they would in real-world settings.

The same progressive compensation structure applied to both groups (£8/hour at baseline to Week 2; £9/hour at Weeks 2–4; £10/hour at Weeks 4–8), providing a financial incentive to remain engaged with the study across the full 8-week period.
Compensation will be provided through Prolific. The researcher sets the payment amount in British pounds. Participants receive the payment in their local currency, such as Australian dollars or United States dollars, using the applicable exchange rate on the date of conversion.
Compensation for time spent using the website will be paid in addition to compensation for the time spent completing study assessments.

Following completion of the 8-week study period, participants in the control group will be offered access to the Cara Plus app.
Control group
Active

Outcomes
Primary outcome [1] 345682 0
Generalised Anxiety
Timepoint [1] 345682 0
Week 0 (baseline), Week 2, Week 4, Week 8.
Secondary outcome [1] 462896 0
Depression
Timepoint [1] 462896 0
Week 0 (baseline), Week 2, Week 4, Week 8.
Secondary outcome [2] 464999 0
Worry
Timepoint [2] 464999 0
Week 0 (baseline), Week 4, Week 8
Secondary outcome [3] 465001 0
Rumination
Timepoint [3] 465001 0
Week 0 (baseline), Week 4, Week 8
Secondary outcome [4] 465002 0
Cognitive Attentional Syndrome
Timepoint [4] 465002 0
Week 0 (baseline), Week 4, Week 8
Secondary outcome [5] 465003 0
User experience
Timepoint [5] 465003 0
Week 8

Eligibility
Key inclusion criteria
Aged 18 years or above.

Proficient in English.

Residing in Australia or the United States.

Access to an Android mobile device with operating system version 8 or higher and reliable internet access.

Mild to moderate symptoms of anxiety and/or depression, defined as a GAD-7 score of 5–14 and/or PHQ-9 score of 5–14.

No changes to psychiatric medication in the past four weeks and no planned changes during the 8-week study period.

No initiation of structured psychotherapy in the past four weeks and no planned changes to psychotherapy during the 8-week study period.
Minimum age
18 Years
Maximum age
No limit
Sex
Both males and females
Healthy volunteers
No
Key exclusion criteria
Severe anxiety symptoms, defined as a GAD-7 score of 15 or higher

Severe depressive symptoms, defined as a PHQ-9 score of 15 or higher

Self-reported suicidal ideation, indicated by a PHQ-9 item 9 response greater than 0

Self-reported diagnosis of, or symptoms of bipolar disorder or psychotic disorders

Self-reported experience of hallucinations, delusions, or loss of contact with reality

Self-reported acute mental health crisis requiring urgent or emergency intervention

Self-reported current substance dependence

Current structured psychotherapy started within past 4 weeks

Planned therapy changes during the 8-week intervention window

Psychiatric medication changes within the past 4 weeks or planned during the 8-week intervention window

Self-reported neurological disorders, including dementia, epilepsy, or intellectual disability

Self-reported serious or unstable medical conditions

Self-reported use of benzodiazepines

Self-reported hearing or vision impairments that would interfere with the app use

Current treatment with transcranial magnetic stimulation, electroconvulsive therapy, or ketamine

Lack of reliable internet access

Prior use of a digital mental health self-help app or structured online self-help programme within the past six months

Study design
Purpose of the study
Educational / counselling / training
Allocation to intervention
Randomised controlled trial
Procedure for enrolling a subject and allocating the treatment (allocation concealment procedures)
Participants will be randomised to one of two study groups using a computer-generated randomisation schedule.
Methods used to generate the sequence in which subjects will be randomised (sequence generation)
Computer-generated randomisation schedule. Stratified by:
o baseline GAD-7 severity band
o sex
Masking / blinding
Blinded (masking used)
Who is / are masked / blinded?
The people receiving the treatment/s


Intervention assignment
Parallel
Other design features
Phase
Not Applicable
Type of endpoint/s
Efficacy
Statistical methods / analysis
Outcome Measures:

Primary: Anxiety (GAD-7)
Secondary: Depression (PHQ-9)
Exploratory: Worry (PSWQ-10)
Cognitive Attentional Syndrome (CAS-I, item 1)
Rumination (BSRI)
User engagement, usability, satisfaction, and likelihood of recommendation.

Research Questions:

Main Research Question

Do participants allocated to the Cara Plus app differ from those allocated to the psychoeducational control condition on primary and secondary outcomes at Week 8 (post intervention)?

Exploratory Research Questions

To explore the impact of the Cara Plus app on worry, cognitive attentional syndrome, and rumination across the trial period.

To explore changes in anxiety and depressive symptoms among participants with anxiety dominant, depression dominant, and minimal symptom profiles at baseline.

To explore user engagement, usability, and satisfaction with the Cara Plus app among intervention group participants.

Hypotheses for Each Outcome Variable:

Anxiety

H1a: Participants allocated to the Cara Plus app will report significantly lower anxiety symptoms (GAD-7) at Week 8 compared to those allocated to the psychoeducational control condition.

Depression

H1b: Participants allocated to the Cara Plus app will report significantly lower depressive symptoms (PHQ-9) at Week 8 compared to those allocated to the psychoeducational control condition.

Worry, Cognitive Attentional Syndrome, and Rumination

H1c: It is hypothesised that participants allocated to the Cara Plus app will report lower levels of worry (PSWQ-10) at Week 4 and Week 8 compared to those allocated to the psychoeducational control condition. This will be examined as an exploratory analysis.

H1d: It is hypothesised that participants allocated to the Cara Plus app will report lower cognitive attentional syndrome (CAS-I, item 1) at Week 4 and Week 8 compared to those allocated to the psychoeducational control condition. This will be examined as an exploratory analysis.

H1e: It is hypothesised that participants allocated to the Cara Plus app will report lower levels of rumination (BSRI) at Week 4 and Week 8 compared to those allocated to the psychoeducational control condition. This will be examined as an exploratory analysis.

Statistical Analysis of Primary and Secondary Outcome Measures:

The primary endpoint is the between group difference in change in GAD-7 from baseline to Week 8. The trial is powered for a small to medium effect size (f = 0.20), with a two sided alpha = 0.05 and power of 80%, requiring approximately 100 participants per group for the main RCT ITT analysis using ANCOVA.

The planned efficacy sample comprises two main RCT groups of 100 participants each (N = 200). To allow for 25% attrition and retain adequate power for the planned per protocol sensitivity analysis, the main RCT cohort will be intentionally over recruited to approximately N = 300. A separate single arm exploratory cohort of approximately N = 100 will also be recruited independently of the main RCT. The total overall recruitment target is therefore approximately 400 participants.

A per protocol sensitivity analysis and linear mixed modelling across all time points (baseline, Week 2, Week 4, Week 8) will also be conducted.

The primary and secondary analyses will be conducted in participants with mild to moderate symptoms on both GAD-7 and PHQ-9 at baseline (GAD-7 5–14 and PHQ-9 5–14).

The primary analysis uses ANCOVA with change in GAD-7 as the dependent variable, treatment arm as the fixed factor, and baseline GAD-7 score and sex as covariates, conducted under the ITT principle with missing data addressed via Multiple Imputation.

The secondary analysis mirrors this approach using change in PHQ-9 as the outcome, with treatment arm as the fixed factor and baseline PHQ-9 score and sex as covariates.

For the exploratory GAD-7 and PHQ-9 outcomes, the analyses will be conducted in three cohorts independent of the main RCT randomisation: anxiety dominant with minimal depressive symptoms (GAD-7 5–14 and PHQ-9 0–4), depression dominant with minimal anxiety symptoms (PHQ-9 5–14 and GAD-7 0–4), and minimal symptoms on both measures (GAD-7 0–4 and PHQ-9 0–4). Participants in these exploratory cohorts will not be randomised and will all be assigned to receive the Cara Plus app. Analyses will focus on safety, symptom change, and symptom stability, using within group change from baseline to Week 8, with paired t tests, effect sizes (Cohen’s d), and 95% confidence intervals reported.

For the exploratory worry and rumination outcomes (PSWQ-10, CAS-I item 1, BSRI), between group differences will be examined at Week 4 and Week 8 using independent samples t tests, with effect sizes (Cohen's d) and 95% confidence intervals reported. These analyses are hypothesis generating and will be interpreted accordingly, without adjustment for multiple comparisons.

Qualitative Analysis

At Week 8, participants in the intervention group will complete two open ended qualitative items assessing their experience of using the Cara Plus app: a series of questions about their overall app use experience, assessing usability, feature usage, satisfaction, and likelihood of recommendation, and the following item: “How has your experience of worrying, dwelling (ruminating), or over thinking changed since using the Cara Plus?”

Qualitative data will be analysed using Braun and Clarke's (2006) six phase framework for thematic analysis. This approach was selected for its flexibility and suitability for exploring participants' subjective experiences of a novel digital mental health intervention.

Given the online, text based nature of data collection, responses are anticipated to vary in length and depth. Themes will be reported alongside illustrative participant quotes and interpreted in the context of the quantitative findings.

Descriptive statistics will also be reported for structured app experience items assessing usability, feature usage, satisfaction, and likelihood of recommendation.

Recruitment
Recruitment status
Not yet recruiting
Date of first participant enrolment
Anticipated
23/07/2026
Actual
Date of last participant enrolment
Anticipated
31/07/2026
Actual
Date of last data collection
Anticipated
1/10/2026
Actual
Sample size
Target
400
Accrual to date
Final
Recruitment in Australia
Recruitment state(s)
ACT,NSW,NT,QLD,SA,TAS,WA,VIC
Recruitment outside Australia
Country [1] 28191 0
United States of America
State/province [1] 28191 0

Funding & Sponsors
Funding source category [1] 322043 0
Commercial sector/Industry
Name [1] 322043 0
Spectrum Wellness Limited (Spectrum.Life)
Country [1] 322043 0
Ireland
Primary sponsor type
University
Name
University of the Sunshine Coast
Address
Country
Australia
Secondary sponsor category [1] 324967 0
None
Name [1] 324967 0
Address [1] 324967 0
Country [1] 324967 0

Ethics approval
Ethics application status
Approved
Ethics committee name [1] 320574 0
Human Research Ethics Committee of the University of the Sunshine Coast
Ethics committee address [1] 320574 0
Ethics committee country [1] 320574 0
Australia
Date submitted for ethics approval [1] 320574 0
08/05/2026
Approval date [1] 320574 0
26/05/2026
Ethics approval number [1] 320574 0
A252742

Summary
Brief summary
Trial website
Trial related presentations / publications
Public notes

Contacts
Principal investigator
Name 150934 0
Dr Lee Kannis-Dymand
Address 150934 0
University of the Sunshine Coast, 90 Sippy Downs Drive, Sippy Downs, Queensland 4556, Australia
Country 150934 0
Australia
Phone 150934 0
+61 7 5459 4879
Fax 150934 0
Email 150934 0
[email protected]
Contact person for public queries
Name 150935 0
Lee Kannis-Dymand
Address 150935 0
University of the Sunshine Coast, 90 Sippy Downs Drive, Sippy Downs, Queensland 4556, Australia
Country 150935 0
Australia
Phone 150935 0
+61 7 5459 4879
Fax 150935 0
Email 150935 0
[email protected]
Contact person for scientific queries
Name 150936 0
Lee Kannis-Dymand
Address 150936 0
University of the Sunshine Coast, 90 Sippy Downs Drive, Sippy Downs, Queensland 4556, Australia
Country 150936 0
Australia
Phone 150936 0
+61 7 5459 4879
Fax 150936 0
Email 150936 0
[email protected]

Data sharing statement
Will the study consider sharing individual participant data?
No


What supporting documents are/will be available?

TypeCitationLinkEmailOther DetailsAttachment
Statistical analysis plan    AIMHI_Statistical Analysis Plan.pdf


Results publications and other study-related documents

Documents added manually
No documents have been uploaded by study researchers.

Documents added automatically
No additional documents have been identified.