AISHA: Exploratory Randomised Controlled Trial of an Artificial Intelligence Self Harm Application

Sponsor

Pakistan Institute of Living and Learning (Other)

Overall Status

Not yet recruiting

CT.gov ID

NCT06019663

Collaborator

(none)

312

Enrollment

Locations

Arms

Anticipated Duration (Months)

156

Patients Per Site

6.5

Patients Per Site Per Month

Study Details

Study Description

Brief Summary

This is a mixed-methods study. The quantitative component will comprise of a multicentre rater-blind, randomized controlled trial to evaluate the feasibility and acceptability of AISHA in addition to treatment as usual (TAU) compared to TAU alone in Pakistan.

In-depth interviews and focus groups with other stakeholders will take place to consider the wider implementation of AISHA in mental health services in Pakistan.

Condition or Disease	Intervention/Treatment	Phase
Self-harm Suicide	Behavioral: The Artificial Intelligence intervention for Self-Harm prevention Application (AISHA) is a mobile intervention based on Y-CMAP	N/A

Detailed Description

Mental health resources in LMICs, including Pakistan, do not match the burden of mental illness, resulting in a massive mental health treatment gap. Interventions delivered on mobile platforms (M-Health) have the potential to overcome these barriers. Evidence supports the acceptability and efficacy of interventions delivered on digital platforms for a variety of mental disorders. Mobile-based interventions have been developed to target self-harm, however, there is paucity of research supporting their efficacy in LMIC.

To our knowledge there are currently no clinical trials examining the feasibility, acceptability and preliminary efficacy of a culturally adapted CBT informed problem solving intervention delivered on a digital platform for self-harm prevention.We have developed an Artificial Intelligence Self-Harm prevention Application (AISHA) to deliver an evidence-based problem solving intervention (CMAP) for people presenting with self-harm.

Study Design

Study Type:

Interventional

Anticipated Enrollment :

312 participants

Allocation:

Randomized

Intervention Model:

Parallel Assignment

Masking:

Single (Outcomes Assessor)

Primary Purpose:

Other

Official Title:

Exploratory Randomised Controlled Trial of an Artificial Intelligence Self Harm Application

Anticipated Study Start Date :

Oct 30, 2023

Anticipated Primary Completion Date :

Oct 30, 2024

Anticipated Study Completion Date :

Oct 30, 2025

Arms and Interventions

Arm	Intervention/Treatment
Experimental: Intervention	Behavioral: The Artificial Intelligence intervention for Self-Harm prevention Application (AISHA) is a mobile intervention based on Y-CMAP YCMAP is a "Youth culturally adapted manual assisted psychological intervention" based on CBT principles. The therapy focuses on current problems that contributed to the self-harm episode. Therapists and adolescent clients choose from a list of techniques those which are most relevant to the client's problems. Therapy is therefore adapted to fit with the client's problems and primarily utilises problem solving, CBT, and dialectical therapy strategies to bring about change.
No Intervention: Standard Routine Care Standard care for self-ham in Pakistan does not usually involve psychiatric or psychotherapy follow-up. Clinical teams in psychiatric departments or family medicine practices provide standard routine care according to their clinical judgment and available resources. We will obtain details of any treatment received by each participant

Arm

Intervention/Treatment

Experimental: Intervention

Behavioral: The Artificial Intelligence intervention for Self-Harm prevention Application (AISHA) is a mobile intervention based on Y-CMAP

YCMAP is a "Youth culturally adapted manual assisted psychological intervention" based on CBT principles. The therapy focuses on current problems that contributed to the self-harm episode. Therapists and adolescent clients choose from a list of techniques those which are most relevant to the client's problems. Therapy is therefore adapted to fit with the client's problems and primarily utilises problem solving, CBT, and dialectical therapy strategies to bring about change.

No Intervention: Standard Routine Care

Standard care for self-ham in Pakistan does not usually involve psychiatric or psychotherapy follow-up. Clinical teams in psychiatric departments or family medicine practices provide standard routine care according to their clinical judgment and available resources. We will obtain details of any treatment received by each participant

Outcome Measures

Primary Outcome Measures

Beck Scale for Suicide Ideation [Change in scores from baseline to 3-month follow up (end of intervention)]
This is a self-report measure of current suicidal ideation. The 19-item instrument indicates the severity of suicidal ideation in the previous week. Higher scores (≥6) on the scale indicate greater risk. When measured against other instruments of suicidal ideation, the convergent validity of the BSI has been demonstrated, r= 0.41(de Beurs, Fokkema, & O'Connor, 2016). We have used the Urdu translated version of this instrument in prior work from Pakistan and found the Cronbach's alpha to be 0.89 (M. O. Husain et al., 2019). Higher scores indicate greater severity of suicidal ideation

Secondary Outcome Measures

Beck Hopelessness Scale [Change in scores from baseline to 3-month follow up (end of intervention)]
This is a 20-item self-report instrument designed to measure three aspects of hopelessness, feelings about the future, loss of motivation and expectations. Higher scores indicate increasing severity of hopelessness; normal (0-3), mild hopelessness (4-8), moderate hopelessness (9-14), severe hopelessness (>14). The test-retest scores of the scale are found to be good (r=0.81) (Ayub, 2009). The Urdu version of the BHS has been used in Pakistan and reported to have a reliability coefficient (Kuder-Richardson Index) of 0.93 (M. O. Husain et al., 2019).
Beck Depression Inventory [Change in scores from baseline to 3-month follow up (end of intervention)]
This is a 21-item scale of depressive symptoms. Higher scores on the scale indicate greater severity of depression. Mild depression is indicated by scores of 14 -19; Moderate depression scores of 20 - 28 and Severe depression scores of 29-63. The BDI has concurrent validity, correlating highly with the Hamilton Rating Scale for Depression (r=0.72 - 0.73) (A. T. Beck, Steer, R.A., Carbin, M.G., 1988). The test-retest reliability is also high (r= 0.60) (Jo, Park, Jo, Ryu, & Han, 2007). The Cronbach's Alpha coefficients of the Urdu translated BDI are good (0.75-0.92) (Khan, Marwat, Noor, & Fatima, 2015). We have used the Urdu translated version of this instrument in our previous work, the Cronbach's alpha was 0.97 (M. O. Husain et al., 2019).
Coping resource inventory [Change in scores from baseline to 3-month follow up (end of intervention)]
The CRI is a structured instrument to assess the coping resources to manage stress available to an individual. We have used Urdu translated version in our previous exploratory trial (N. Husain et al., 2014). Higher scores indicate better coping skills
The Resilience Scale [Change in scores from baseline to 3-month follow up (end of intervention)]
The RS is a 25-item self-report scale that aims to assess an individual's level of resilience. This is a 7 point Likert Scale and provides a total score of resilience, ranging from 25 to 175, with higher scores indicative of higher levels of resilience.
The Problem Solving Inventory [Change in scores from baseline to 3-month follow up (end of intervention)]
The PSI is a 35-item instrument that measures the individual's perceptions regarding one's problem-solving style and problem-solving abilities in the everyday life. It consists of three subscales. Higher scores indicate higher level of problem solving abilities
Euro-Qol Quality of Life Scale [Change in scores from baseline to 3-month follow up (end of intervention)]
This is an instrument which is used to measure quality of life or health profiles of participants. Higher scores indicate higher health related quality of life
Client Satisfaction Questionnaire [Level of satisfaction with the services received at end of intervention i.e., 3-month post-randomization]
This is an eight item self-report questionnaire used to assess participant's satisfaction with the services using a 4-point likert scale. Higher scores indicate greater satisfaction.
Client Service Receipt Inventory (Beecham & Knapp, 2001): [Change in scores from baseline to 3-month follow up (end of intervention)]
This instrument is used to collect information on the use of health services (both formal and informal sector such as physician, faith healers/Imams)

Eligibility Criteria

Criteria

Ages Eligible for Study:

16 Years to 25 Years

Sexes Eligible for Study:

All

Accepts Healthy Volunteers:

Inclusion Criteria:

All patients presented to the participating family physicians, community health centres, emergency departments or admitted to a general hospital after an episode of self-harm at participating sites.
Age 16 to 25 years
Participants must be familiar with functioning of handheld mobile devices (mobile phones, tablets).
Able to read Urdu or English language.
Individuals able to provide written informed consent.
Have an android mobile phone/device

Exclusion Criteria:

Participants with ICD-I0 mental disorder; due to a general medical condition or substance misuse, dementia, delirium, alcohol or drug dependence, schizophrenia, bipolar disorder, learning disability which prevents participation.
Participants needing inpatient psychiatric treatment as determined by their clinical teams