Trial Review

Technical difficulties have been reported by some users of the search function and is being investigated by technical staff. Thank you for your patience and apologies for any inconvenience caused.

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
ACTRN12621001223820
Ethics application status
Approved
Date submitted
16/08/2021
Date registered
13/09/2021
Date last updated
21/04/2023
Date data sharing statement initially provided
13/09/2021
Date results information initially provided
21/04/2023
Type of registration
Prospectively registered

Titles & IDs
Public title
Using AI to investigate the effectiveness of smartphone delivered self-care strategies for psychological distress in university students.
Scientific title
An AI-response-adaptive randomised controlled trial of smartphone app-delivered self-care strategies for psychological distress in university students
Secondary ID [1] 305013 0
Nil known
Universal Trial Number (UTN)
U1111-1269-1762
Trial acronym
Not applicable.
Linked study record
This study is a follow-up to the pilot study registered under the record ACTRN12621001092886.

Health condition
Health condition(s) or problem(s) studied:
Psychological distress 323184 0
Condition category
Condition code
Mental Health 320767 320767 0 0
Anxiety
Mental Health 320768 320768 0 0
Depression

Intervention/exposure
Study type
Interventional
Description of intervention(s) / exposure
This study will include a 10-day baseline measurement period incorporating questionnaire batteries on Day 1 (baseline) and Day 10 (mid-point), plus Ecological Momentary Assessment (EMA) from Day 1 to Day 9.

The intervention period is for a duration of 14 days. All interventions in this study will be delivered via a smartphone app. Each study participant will receive one study intervention, once, over a period of 2 weeks. Interventions are self-administered.

Each arm of the trial is administered via a common mobile app. During the intervention period, participants will only have access to the allocated intervention. All of the app content is designed specifically for this study. Participants are instructed to complete tasks within the app each day, however the time spent interacting with the app is at their discretion (however, see below for suggested time spent engaging with intervention activities).

This study is an AI-adaptive group sequential RCT. The study proceeds as a series of discrete “mini-trials”, each recruiting a sample of participants who are allocated to study arms under the control of an AI algorithm. More details on this can be found in Study Design section.

Participants who complete the study (i.e. finish post-intervention questionnairs) will receive a $30 electronic gift card in recognition of their contribution to the study. In addition, if participants complete the set of follow-up questionnaires, they will go into a draw to win one of three $50 electronic gift cards available to their mini-trial group.

Arm 1 Mindfulness

The mindfulness intervention starts with an introductory video (3 minutes), followed by five guided mindfulness practices delivered as audio recordings (3-5 minutes in duration each). The five guided mindfulness practices are: 1) Mindful breathing, 2) Unhooking from thoughts, 3) Body scan, 4) Mindful eating, and 5) Mindful walking.

The introductory video is available to participants immediately after allocation to the intervention. The first mindfulness audio guide becomes available immediately after the introductory video has been completed. The subsequent mindfulness audio guides are sequentially released, at one day intervals (regardless of participant engagement). Participants are expected to complete a minimum of one mindfulness exercise daily at their convenience and for all modules to be completed at least once by end of the intervention period.

Arm 2 Physical activity

The physical activity intervention starts with an introductory infographic (i.e. illustrated text that participants read), then prompts participants to choose a goal each day to increase their physical activity that day. The introductory infographic demonstrates benefits of physical activity, Australian guidelines for physical activity, setting realistic goals, and tips to increase physical activity. For the daily goal setting, participants can choose to increase their step count, follow an evidence-based 7-minute high intensity circuit training (HICT) workout, engage in other activities such as social sport, gardening, yoga etc., or opt for a rest day. Participants can review the introductory infographic and goal setting information as many times as they please throughout the intervention, however new/additional information will not be presented as the intervention period proceeds.

The HICT protocol includes the following exercises: Jumping jacks, Wall sit, Push up, Abdominal crunch, Step up onto a chair, Squat, Tricep dip on chair, Plank, High knees, Lunge, Pushup and rotation, Side plank.

Participants will be invited to complete a daily log of physical activity (i.e. number of minutes they spent being physically active). Participants will not receive feedback on this log during the intervention; it is intended for the research team to monitor adherence. Time spent interacting with the app during the physical activity intervention is at the discretion of the participant and depends on the physical activity chosen for each day; however, participants are expected to log their activity and set a goal each day, which is expected to take at minimum approximately 3 minutes. It is expected that physical activity will be completed daily at each participant’s convenience, including 7-minute work out three times weekly or, alternatively, increasing steps by 10% weekly.

Arm 3 Sleep Hygiene

The sleep hygiene intervention centres around four brief, sequential modules covering key sleep hygiene concepts which are delivered via infographics. Each module takes up to five minutes to read. Modules include: 1. Why sleep (benefits of sleep), 2. Sleep habits (regular wake/sleep time, reducing naps, wind down routine), 3. Sleep environment (environmental adjustment including impact of electronics and stimulus control information), and 4. Daily activities for sleep (diet, reducing stimulants, physical activity).

The modules are made available to participants as follows: module 1 is available to participants immediately after allocation to the intervention; module 2 becomes available immediately after a participant has completed Module 1. Module 3 and Module 4 are subsequently made available at two day intervals (regardless of participant engagement).

Adherence to the interventions will be measured by accessing app analytics (i.e. module completion, number of times mindfulness audio is played, number of times physical activity goal for the day is registered), and self-reported log of engagement with the intervention (i.e. daily log of time spent practising mindfulness, being physically active, or sleeping).
Intervention code [1] 321410 0
Lifestyle
Intervention code [2] 321411 0
Behaviour
Comparator / control treatment
The active control condition consists of two weeks of Ecological Momentary Assessment (EMA) measurements, The control condition is also delivered via the same app as the active interventions (participants will only have access to the form to enter EMA measurements during the intervention period).

EMA involves repeated sampling of participants' current behaviours and experiences in real time, in participants' natural environment. Modified Positive and Negative Affect Schedule, Short Form (PANAS) will be used to collect EMA data. The measurement schedule will be twice daily signals (participants complete EMA in response to app generated notifications) and unlimited self-initiated measurements (participants can elect to complete EMA within the app at any point, in addition to when prompted).

It is anticipated that duration of data entry for participants in the control group will take a maximum of 5 minutes per day.
Control group
Active

Outcomes
Primary outcome [1] 328579 0
Self-reported psychological distress as measured by Depression Anxiety and Stress Scale - 21 item version (DASS-21)
Timepoint [1] 328579 0
Mid-point (10 days after app download, following baseline EMA period and prior to intervention commencement), Post-intervention (two weeks after intervention commencement; primary time point), & 8-week follow-up (10 weeks after intervention commencement).
Secondary outcome [1] 399510 0
Level of physical activity measured by the Modified Physical Activity Vital Sign
Timepoint [1] 399510 0
Mid-point (10 days after app download, following baseline EMA period and prior to intervention commencement), Post-intervention (two weeks after intervention commencement), & 8-week follow-up (10 weeks after intervention commencement).
Secondary outcome [2] 399511 0
Sleep quality measured by Abridged Pittsburgh Sleep Quality Index
Timepoint [2] 399511 0
Mid-point (10 days after app download, following baseline EMA period and prior to intervention commencement), Post-intervention (two weeks after intervention commencement), & 8-week follow-up (10 weeks after intervention commencement).
Secondary outcome [3] 399512 0
Level of mindfulness measured by Mindfulness single item questionnaire
Timepoint [3] 399512 0
Mid-point (10 days after app download, following baseline EMA period and prior to intervention commencement), Post-intervention (two weeks after intervention commencement), & 8-week follow-up (10 weeks after intervention commencement).
Secondary outcome [4] 399514 0
Productivity Costs Questionnaire; a standardised questionnaire including three modules measuring productivity losses related to paid work due to absenteeism and presenteeism, as well as productivity losses related to unpaid work.
Timepoint [4] 399514 0
8-week follow-up (10 weeks after intervention commencement).
Secondary outcome [5] 399516 0
Use of Mental Health Care Services, assessed via self-report items related to hospitalization, outpatient services, online self-help, community based services and pharmacotherapy.
Timepoint [5] 399516 0
8-week follow-up (10 weeks after intervention commencement).
Secondary outcome [6] 399517 0
Quality of life measured by EQ-5D-5L & Recovering Quality of Life (ReQoL-10)
Timepoint [6] 399517 0
8-week follow-up (10 weeks after intervention commencement).
Secondary outcome [7] 399518 0
GPS location data, sampled on the basis of significant location change (as defined by the underlying operating system).

Note: only for participants who do not opt-out of digital phenotyping at consent or disable the required permissions on their smartphone.
Timepoint [7] 399518 0
Day 1 -30 of the app experience. That is, from when the app is installed, through baseline EMA period, and until post-intervention is completed (2 weeks after intervention commencement).
Secondary outcome [8] 399519 0
Gravity-adjusted accelerometery (Android devices and iPhone 7 or later) and gyroscope (Android devices only) data, sampled with best-effort at 50Hz for 1 minute every 5 minutes.

Note: only for participants who do not opt-out of digital phenotyping at consent or disable the required permissions on their smartphone.
Timepoint [8] 399519 0
Day 1 -30 of the app experience. That is, from when the app is installed, through baseline EMA period, and until post-intervention is completed (2 weeks after intervention commencement).
Secondary outcome [9] 399520 0
Operating system-generated activity label plus confidence (e.g. in vehicle, walking, stationary with low/medium/high confidence), step count and total distance travelled sampled with best-effort once every 5 minutes.

Note: only for participants who do not opt-out of digital phenotyping at consent or disable the required permissions on their smartphone.
Timepoint [9] 399520 0
Day 1 -30 of the app experience. That is, from when the app is installed, through baseline EMA period, and until post-intervention is completed (2 weeks after intervention commencement).
Secondary outcome [10] 399521 0
Device power state and battery percentage, sampled with best effort once every 5 minutes.

Note: only for participants who do not opt-out of digital phenotyping at consent or disable the required permissions on their smartphone.
Timepoint [10] 399521 0
Day 1 -30 of the app experience. That is, from when the app is installed, through baseline EMA period, and until post-intervention is completed (2 weeks after intervention commencement).
Secondary outcome [11] 399522 0
Device model, make, platform and app software versions, submitted with every sensor payload.

Note: only for participants who do not opt-out of digital phenotyping at consent or disable the required permissions on their smartphone.
Timepoint [11] 399522 0
Day 1 -30 of the app experience. That is, from when the app is installed, through baseline EMA period, and until post-intervention is completed (2 weeks after intervention commencement).
Secondary outcome [12] 399523 0
Adherence to the interventions, measured by accessing app analytics (i.e. module completion, number of times mindfulness audio is played, number of times physical activity goal for the day is registered), and self-reported log of engagement with the intervention (i.e. daily log of time spent practising mindfulness, being physically active, or sleeping).
Timepoint [12] 399523 0
Throughout the intervention period (day 11-24 of the app experience).
Secondary outcome [13] 399524 0
Usability of the app, measured via study-specific self-report questionnaire. This 'UX questionnaire' asked participants to rate usability, usefulness and satisfaction with the app on a 5-point likert scale.
Timepoint [13] 399524 0
Post-intervention (two weeks after intervention commencement).

Eligibility
Key inclusion criteria
- 18 years or older;
- Living in Australia;
- Currently studying at an Australian university, TAFE or other higher education institution;
- Fluent in English;
- Own an up-to-date smartphone, with active mobile number and internet access; and
- Are experiencing elevated psychological distress, indicated by a score of 20 or higher on the Kessler Psychological Distress Scale, 10 item version (K-10) at screening.
Minimum age
18 Years
Maximum age
No limit
Sex
Both males and females
Can healthy volunteers participate?
No
Key exclusion criteria
- People who have reported significant thoughts of suicide in the past month, indicated by a score of greater than or equal to 21 on the Suicide Ideation Attributes Scale (SIDAS).
- People who self-report a current diagnosis of psychosis or bipolar disorder;
- People who have been previously recruited to the study (screened eligible and consented);
- People who self-report that they expect major events or disruptions in the subsequent two months that would make it difficult to participate in the study;
- People who indicate they cannot safely undertake a self-guided physical activity intervention (i.e. in which they control the intensity and type of activity involved) for any reason.

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 access the link to the automated online screening via the study website. They will read and sign the electronic participant information statement and consent form, then complete the self-report screening questions. If eligible to take part, at the end of the online screening, participants will see an onscreen message and receive an automated email with the start date for the next mini-trial. On the first day of the mini-trial period, participants will be sent a link to download the study app, and complete the baseline questionnaires. After completing the baseline questionnaires, baseline EMA period and mid-point questionnaires (10 days after app download), participants will be randomly allocated by the app to their intervention. The app will only reveal the allocated condition to participants after the baseline period and mid-point questionnaires are completed (i.e. on Day 11).

Allocation concealment will be guaranteed by:
- Preventing access by blinded study staff to the computer system holding randomisation information; and
- Breaking randomization codes only once primary data analysis is complete.
Methods used to generate the sequence in which subjects will be randomised (sequence generation)
Computerised allocation will be performed automatically by the Optimise platform.

An initial allocation ratio (mini-trial 1) of 1:1:1:1 (three interventions plus active control) will be used. Sequence generation for random assignment will be based on computer generated random numbers.

Subsequent allocation will be determined on a per-individual basis, conditioned on the context offered by baseline DASS-21 scores, by the Multi-arm bandit algorithm as part of its designed function. In subsequent mini-trials, allocation is effectively deterministic based on the Upper Confidence Bound (UCB) of the observed context for each new participant. There will be no minimum per-arm allocation, as one of the consequences of variable allocation under the multi-arm bandit scheme is that allocation to any given arm may tend to zero.
Masking / blinding
Blinded (masking used)
Who is / are masked / blinded?



The people analysing the results/data
Intervention assignment
Parallel
Other design features
This is an AI-adaptive group sequential RCT. The study proceeds as a series of discrete “mini-trials”, each recruiting a restricted sample of participants who are allocated to study arms under the control of an AI algorithm. In a contextual AI-adaptive study, learning is conditioned on a pre-specified set of characteristics (e.g. demographic properties) so that trial analyses can explore questions such as “What works for whom?”.

After each mini-trial, the algorithm uses the observed data to update an underlying model of effects attributable to each intervention and determine how the study will proceed. The trial can then either:
- Stop, if interim analyses confirm that all optimisation goals have been satisfied.
- Continue with another mini-trial. The updated model is used to allocate participants to each subsequent trial consistent with the pre-specified goals.
Phase
Not Applicable
Type of endpoint/s
Efficacy
Statistical methods / analysis
The algorithm used in this AI-adaptive trial is a contextual, Gaussian process-based multi-arm bandit (cMAB).

OPTIMISATION GOALS:
After blindly rank-ordering all study arms by their observed mean primary endpoint scores (where lower is better) such that arm A < arm B < arm C < arm D , the optimisation goals are:
- Goal 1: Find the best. Using the smallest number of mini-trials/participants possible, find the best performing arm, such that for all pairwise comparisons (A<B, A<C and A<D) a statistically significant difference is detected.
- Goal 2: Find the next best. Should Goal 1 be satisfied (best-performing intervention having been identified and fixed), then with the smallest number of mini-trials/participants possible, find the next best arm, such that for both pairwise comparisons (B<C and B<D) a statistically significant difference is detected.
- Goal 3: Find the third best. Should Goal 2 be satisfied (second best-performing intervention having been identified and fixed), and further mini-trials remain, trial will continue to find with the smallest number of mini-trials/participants possible the next best intervention arm, such that for the pairwise comparison C<D the null hypothesis of no difference can be rejected.

Optimisation goals will be formally evaluated as part of planned interim analyses, to identify robust estimates of the effect sizes of the most effective intervention(s) as quickly as possible while simultaneously minimising the numbers of participants allocated to likely-ineffective interventions (consistent with the pragmatic effectiveness orientation of the trial).

ANALYSIS OF PRIMARY ENDPOINT:
For analysis/reporting, the sample will be split into three clinically-relevant severity groups according to their baseline normalised DASS-21 total score (following the procedure outlined in the DASS manual): Normal/mild symptoms, Moderate symptoms, Severe/extremely severe symptoms. This approach will allow exploration of whether the most effective intervention(s) differ according to severity. For each group, planned contrasts will compare arms using the DASS-21 total score at pre-intervention (mid) and post intervention. After blindly rank-ordering all study arms (where lower is better) such that arm A < arm B < arm C < arm D, we will compare:
- The best performing intervention versus the other lower-performing interventions; A vs B, A vs C and A vs D.
- The second-best performing intervention versus the other lower-performing interventions; B vs C and B vs D.
- The third-best performing intervention versus the other lower-performing intervention: C vs D.
These contrasts are intended to be explicitly aligned with the cMAB optimisation goals (see above).

We will use a mixed-effects repeated measures (MMRM) model adjusted for multiple comparisons. Pre-intervention (mid) scores will be incorporated as a predictor variable in all models. An unconstrained variance–covariance matrix will model within-individual dependencies. Transformation of scores, including categorisation, may be undertaken to satisfy distributional assumptions and accommodate outliers.
Analysis of the primary endpoint will be based on an intention-to-treat analysis strategy, analysing all participants starting the study regardless of the intervention received, under the assumption that missing data are Missing At Random.

INTERIM ANALYSIS:
We will conduct interim analyses three times during the trial period, after mini-trials 4, 8 and 12 (evenly spaced). Interim analyses will be used to:
- Decide for each context-related clinical severity group defined above, whether Optimisation Goal 1 has been satisfied and therefore whether the cMAB can be reconfigured for that group to focus on Optimisation Goal 2.
- Decide whether Optimization Goal 2 has been satisfied for all clinical severity groups, in which case the trial will seek to achieve Goal 3.

ADDITIONAL ANALYSES:
Mediation analyses will be explored using structural equation modelling. We will use mixed effect logistic and Poisson regression models to assess if self-reported psychological distress and suicidality at the baseline are associated with the likelihood to respond to feelings measured using EMA. We will assess the effect of time-varying responses to feelings, momentary affect, and changes in self-reported psychological distress and suicidality using mixed effects regression models. The relationship between momentary affect, psychological distress, exercise, and sleep quality at a given time interval will be assessed by mixed effect regression models. Descriptive analyses will be used to examine compliance and reactivity of the EMA.

Machine learning will be used to analyse digital phenotyping data to:
- explore whether any novel behavioural factors predict the study primary endpoint; and
- investigate within-individual behavioural signals that predict individual changes in self-reported distress or affect measured using EMA.

SIGNIFICANCE LEVEL:
The study significance level for individual comparisons, alpha, is 0.05. The risk of Type 1 error inflation associated with repeated planned comparisons that will be performed for the interim and trial analyses will be managed by applying a Benjamini-Hochberg adjustment to each set of comparisons in order to control the False Discovery Rate for each clinical severity group at 5%.

PLANNED SAMPLE SIZE:
Up to 1200 participants across 12 mini-trials. To allow for attrition between screening and mini-trial commencement, recruitment for each mini-trial will continue until 120 individuals have screened eligible. Assuming that up to a third do not respond to the subsequent invitation to install the study app and complete baseline questionnaires, this will yield at least 80 participants starting the mini-trial (i.e. 20 per arm, assuming the mini-trial 1 allocation ratio of 1:1:1:1). Attrition after baseline is assumed at 20%, resulting in expected completion (i.e. having mid and post assessments) of 64 (i.e. n=16 per arm, assuming the initial allocation ratio of 1:1:1:1). Assumptions will be updated based on observed attrition. Recruitment procedures may be adjusted to recruit more/fewer participants to ensure sufficient numbers of participants completing the mid and post assessments.

SELECTION OF SUBJECTS FOR ANALYSIS:
All randomized subjects who download the study app and complete the baseline questionnaire will be analysed.

MISSING DATA:
We intend to analyse data from all participants who start the study. To qualify as having started the study, participants must install the app and complete the baseline questionnaires within the 72-hour window. Participants who do not do this will be excluded from all study analyses.

To reduce bias in the assessment of a treatment’s effectiveness, intention to treat based analysis of the data will be conducted. For this, all data from participants who had been recommended a treatment by the optimiser (following their completed onboarding DASS survey), who had also completed their mid-DASS survey but failed to complete the post-DASS survey, are included in the analysis.

To accommodate the results of participants with incomplete data, ‘last observation carried forward’ (LOCF) is avoided as this has been shown to produce potentially biased estimates. Instead, an observation-wise ANOVA-type model of group by time effects will be fitted (mixed-model repeated measures), allowing the variance of observations to vary between occasions of measurement (mid-DASS and post-DASS time points) and the residuals of individuals to correlate freely over occasions. This approach assumes that missing assessments are missing at random, an assumption that allows missingness to be dependent on observed information (intervention assignment and previous scores) but not on the unobserved values themselves.

As our analysis would be conducted separately for each cohort, the amount of data points will be divided and may lead to a small number of samples for a particular treatment within a cohort’s analysis. To accommodate for this, an adjustment to the degrees of freedom of the mixed-model would be applied as per the Satterthwaite method.

CONFOUNDING:
The possibility of confounding will be assessed by looking for differences between mini-trials/arms across a range of measured characteristics/behaviours (e.g. engagement/compliance data, self-reported life events collected at post, and intervention compliance measured automatically by the study app).

Recruitment
Recruitment status
Completed
Date of first participant enrolment
Anticipated
11/11/2021
Actual
9/11/2021
Date of last participant enrolment
Anticipated
15/02/2023
Actual
3/11/2022
Date of last data collection
Anticipated
28/04/2023
Actual
17/02/2023
Sample size
Target
1200
Accrual to date
Final
1683
Recruitment in Australia
Recruitment state(s)
ACT,NSW,NT,QLD,SA,TAS,WA,VIC

Funding & Sponsors
Funding source category [1] 309402 0
Government body
Name [1] 309402 0
Department of Health, Australian Government
Country [1] 309402 0
Australia
Primary sponsor type
University
Name
University of New South Wales
Address
Chancellery UNSW
High Street Kensington NSW 2033
Country
Australia
Secondary sponsor category [1] 310371 0
None
Name [1] 310371 0
Address [1] 310371 0
Country [1] 310371 0

Ethics approval
Ethics application status
Approved
Ethics committee name [1] 309208 0
University of New South Wales Human Research Ethics Committee
Ethics committee address [1] 309208 0
UNSW Research Ethics & Compliance Support
The University of New South Wales
Sydney NSW 2052
Ethics committee country [1] 309208 0
Australia
Date submitted for ethics approval [1] 309208 0
25/01/2021
Approval date [1] 309208 0
01/02/2021
Ethics approval number [1] 309208 0
HC200466

Summary
Brief summary
The purpose of the study is to evaluate whether a specific AI technique (multi-arm contextual bandit combined with an adaptive group sequential study trial design) can be used to efficiently identify how well interventions work in a multi-arm randomised controlled trial in a mental health context. To test this idea, the study is comparing the effectiveness of three brief self-guided smartphone interventions based on mindfulness, physical activity and sleep hygiene, and an active control of mood monitoring (i.e. ecological momentary assessment), in reducing self-reported psychological distress in university students.

Based on past research, we hypothesise that the mindfulness and physical activity conditions will be more effective than the sleep hygiene and active control conditions. We hypothesise that the novel method used in this study will be able to efficiently determine the effectiveness of the interventions.
Trial website
https://www.blackdoginstitute.org.au/research-studies/vibe-up/
Trial related presentations / publications
Public notes
This project will use Artificial Intelligence (AI) and adaptive trial methods to discover the most effective treatments to provide to Australian university students.

There is strong evidence that psychological distress can be reduced through psychological interventions, including those targeting mindfulness, physical activity and sleep. More recent evidence suggests that these interventions can effectively be delivered via digital websites and apps. However, most psychological interventions are only partially effective and we have little information to inform treatment selection between various effective interventions.

Although the randomised controlled trial (RCT) is the ‘gold standard’ method for exploring intervention options, it is time consuming, expensive, and inefficient for comparing multiple alternative interventions. New developments in Artificial Intelligence (AI) offer a way to address the limitations of traditional RCTs. By adapting the design of RCTs and running a series of ‘mini-trials’, it is possible to compare multiple treatment strategies more efficiently and examine what treatment options work best for whom. This kind of approach has only recently been introduced in health settings (i.e., intervention design in skin cancer and improving physical activity) and has not yet been applied to mental health.

The current study has two principal aims:

1) To evaluate whether a specific AI technique (multi-arm contextual bandit combined with an adaptive group sequential study trial design) is effective in identifying estimates of how well each intervention works in a multi-arm randomised controlled trial, using the specific example of digital treatments for distress.

2) To identify the most effective brief self-guided digital treatment for psychological distress in university students, and to see if the most effective treatment differs based on initial severity of symptoms.

To do this, we will recruit up to 1200 Australian university students to participate in a four-arm AI-driven adaptive randomised controlled trial (3 interventions + active control).

Contacts
Principal investigator
Name 113346 0
Prof Helen Christensen
Address 113346 0
Black Dog Institute
Hospital Rd, Randwick NSW 2031
Country 113346 0
Australia
Phone 113346 0
+61 2 9065 9099
Fax 113346 0
Email 113346 0
h.christensen@blackdog.org.au
Contact person for public queries
Name 113347 0
Dr Wu Yi Zheng
Address 113347 0
Black Dog Institute
Hospital Rd, Randwick NSW 2031
Country 113347 0
Australia
Phone 113347 0
+61 2 9065 9248
Fax 113347 0
Email 113347 0
wuyi.zheng@blackdog.org.au
Contact person for scientific queries
Name 113348 0
A/Prof Jill Newby
Address 113348 0
Black Dog Institute
Hospital Rd, Randwick NSW 2031
Country 113348 0
Australia
Phone 113348 0
+61 2 9065 9108
Fax 113348 0
Email 113348 0
j.newby@unsw.edu.au

Data sharing statement
Will individual participant data (IPD) for this trial be available (including data dictionaries)?
Yes
What data in particular will be shared?
Any individual participant data collected during the trial, after de-identification, is potentially available, pending review of data requests by the Study Publications Committee
When will data be available (start and end dates)?
Immediately following publication, no end date
Available to whom?
Researchers who provide a methodologically sound proposal, following approval on a case-by-case basis, at the discretion of the Study Publications Committee
Available for what types of analyses?
Only to achieve the aims of approved research proposals
How or where can data be obtained?
Access is subject to approval by the Study Publications Committee, via the Project Manager by emailing vibeup@blackdog.org.au


What supporting documents are/will be available?

Doc. No.TypeCitationLinkEmailOther DetailsAttachment
12852Informed consent form    382572-(Uploaded-26-10-2021-12-53-58)-Study-related document.docx
12853Ethical approval    382572-(Uploaded-12-08-2021-16-56-43)-Study-related document.Pdf



Results publications and other study-related documents

Documents added manually
TypeIs Peer Reviewed?DOICitations or Other DetailsAttachment
Plain language summaryNo N/A

Documents added automatically
SourceTitleYear of PublicationDOI
EmbaseProtocol for a bandit-based response adaptive trial to evaluate the effectiveness of brief self-guided digital interventions for reducing psychological distress in university students: the Vibe Up study.2023https://dx.doi.org/10.1136/bmjopen-2022-066249
N.B. These documents automatically identified may not have been verified by the study sponsor.