Feasibility and acceptability of contextually adapted AVATAR therapy for distressing voices in Ethiopia and India: a study protocol for the AVATAR3 study

This study protocol outlines the AVATAR3 initiative, which aims to contextually adapt and evaluate the feasibility, acceptability, and safety of the digital AVATAR therapy for distressing voices in Low- and Middle-Income countries through participatory implementation hubs in Ethiopia and India.

The Gist

Imagine your mind is a house, and for some people, there's a very loud, mean, and scary guest living in the attic. This guest (the "voice") talks constantly, insults the owner, and makes them feel small and trapped. For decades, doctors have tried to quiet this guest with medicine, but for many, the guest just won't shut up.

In the UK, researchers invented a high-tech tool called AVATAR therapy to help. It's like a video game where the person gets to create a digital "avatar" (a cartoon character) that looks and sounds exactly like that mean voice. Then, with a therapist's help, the person talks to the avatar. Instead of being scared, they learn to stand up to it, tell it to stop, and eventually, the voice becomes quieter and less scary. It's like teaching someone how to wrestle a bully until the bully gets tired and leaves.

This therapy works great in the UK, but the researchers realized: "What if we tried this in places where the 'bully' sounds different, and the house looks different?"

That is exactly what the AVATAR3 study is about.

The Mission: Taking the Tool to New Neighborhoods

The researchers are taking this digital therapy to two very different countries: Ethiopia and India.

Think of the UK as a specific neighborhood with its own rules, language, and culture. If you bring a tool designed for that neighborhood to a completely different village, it might not fit.

• In Ethiopia, people might believe the voice is a spirit or a curse, and they might first ask a religious leader for help.
• In India, people might think the voice is caused by "black magic" or family karma.

If you just copy-paste the UK therapy into these countries, it might feel strange or even offensive. So, the AVATAR3 team is doing something smarter: They are remodeling the tool.

How They Are Remodeling the Tool (The Two-Phase Plan)

Phase 1: The "Tailor" Phase
Before they give the therapy to anyone, they are acting like expert tailors. They are meeting with:

• People who hear voices (the "experts by experience").
• Families and caregivers.
• Local doctors and even religious leaders.

They are asking: "How do you describe this voice? What does it sound like? What would make you feel comfortable talking to a computer?"

They are using these answers to change the therapy.

• The Voice: They might change the language or the tone of the digital voice to sound more like a local elder or a familiar figure.
• The Look: They are updating the computer graphics so the avatar looks like a person from that specific culture (different skin tones, hairstyles, or even traditional clothing), so it doesn't look like a foreigner.
• The Story: They are changing the way they explain why the therapy works, using local stories and metaphors that make sense to the community.

Phase 2: The "Test Drive"
Once the tool is remodeled, they will try it out on a small group of people (about 15 in Ethiopia and 15 in India).

• They aren't trying to prove it's perfect yet; they are just checking: Does it work? Do people like it? Is it safe?
• They will watch to see if the "mean guest" in the attic gets quieter and if the person feels more in charge of their own house.

Why This Matters

Usually, when new medical inventions come from rich countries (like the UK or USA) to poorer countries, it's like dumping a pile of old clothes on someone and saying, "Wear this." It doesn't always fit.

This study is different. It's like saying, "We have a great coat design, but let's sit down with you, measure you, and sew a new coat that fits your body and your style."

They are also making sure that people from these countries aren't just "test subjects." They are leading the project. Local experts are helping design the study, training the therapists, and deciding how to share the results. It's a partnership, not a rescue mission.

The Big Picture

The goal isn't just to fix one person's voice. It's to see if we can build a bridge.

• Can we take a high-tech solution and make it work in places with very few doctors?
• Can we teach local communities how to use digital tools to heal their own minds?
• Can we learn from these cultures to make the therapy even better for everyone, back in the UK too?

If this works, it could be a game-changer. It could mean that millions of people in Ethiopia, India, and beyond could finally get a way to silence their inner bullies, using a tool that respects who they are and where they come from. It's about giving everyone a fair shot at peace of mind, no matter where they live.

Technical Summary

Based on the provided study protocol, here is a detailed technical summary of the AVATAR3 study.

1. Problem Statement

• Global Treatment Gap: In Low- and Middle-Income Countries (LMICs), access to psychological therapies for psychosis is extremely limited, creating significant treatment gaps and inequalities. Distressing voice-hearing affects approximately 70% of individuals with schizophrenia and often persists despite pharmacotherapy.
• Limitations of Current Interventions: While Cognitive Behavioural Therapy for psychosis (CBTp) is effective, it is resource-intensive and not scalable in LMIC settings.
• Cultural and Contextual Barriers: Existing digital interventions, such as the UK-developed AVATAR therapy, have not been tested in LMICs. Direct implementation without adaptation risks cultural dissonance, particularly regarding concepts of "self," empowerment, and the nature of voice-hearing (which is often attributed to spiritual or supernatural causes in Ethiopia and India rather than purely biomedical models).
• Need for Adaptation: There is an urgent need for interventions that are not only effective but also culturally acceptable, scalable, and contextually adapted to diverse global settings.

2. Methodology

The AVATAR3 study is a mixed-methods research project designed to adapt and test AVATAR therapy in Ethiopia (Addis Ababa) and India (New Delhi). The study is structured into two distinct phases:

Study Design & Framework

• Approach: A participatory, interdisciplinary approach led by local "Innovation and Implementation Hubs" (Addis Ababa University/CDT-Africa/MHSUA and AIIMS).
• Theoretical Frameworks: Adaptation is guided by the ADAPT guidance, the Cultural Adaptation of Scalable Psychological Interventions (CASPI) framework, and the Ecological Validity Model (EVM).
• Patient and Public Involvement (PPI): The study utilizes Experts by Experience (EbE) leadership. In Ethiopia, the Mental Health Service Users Association (MHSUA) is a formal research partner with budget control, ensuring equity and capacity building.

Phase 1: Contextual Adaptation (Formative Work)

• Desk Research: Scoping of cultural contexts, voice phenomenology, explanatory models, and health systems in both countries.
• Stakeholder Engagement: Conducted via focus groups, key informant interviews, and a Responsible Research and Innovation (RRI) workshop involving service users, caregivers, clinicians, and community/faith leaders.
• Therapy Training: Local "Champions" were trained in the UK AVATAR protocol to lead adaptation.
• Software Adaptation:
  • Voice Enrolment: Adapting language and content of voice recordings.
  • Face Enrolment: Augmenting the avatar gallery to reflect local diversity (ethnicity, hairstyles, cultural headwear) beyond the current UK/European/Asian groupings.
• Output: An adapted therapy protocol (Version 1) and refined software, iteratively refined through qualitative feedback.

Phase 2: Feasibility and Acceptability Testing

• Design: Parallel case series (no control group) at both sites.
• Sample Size: $n=15$ participants per site (targeting a 70% completion rate, resulting in ~11 completers per site).
• Intervention: Delivery of the contextually adapted AVATAR therapy.
• Data Collection Points: Baseline, 12-weeks, and 24-weeks follow-up.
• Qualitative Component: In-depth interviews with 10 participants and 8 therapists per site to explore experiences.
• Analysis:
  • Quantitative: Feasibility (completion rates), acceptability, safety (adverse events), and preliminary clinical impact (voice severity, distress).
  • Qualitative: Reflexive thematic analysis to understand the lived experience of the adapted therapy.

3. Key Contributions

• First LMIC Implementation: This is the first initiative to adapt and evaluate AVATAR therapy in LMIC settings (Ethiopia and India), moving beyond the UK and high-income country contexts.
• Equitable Partnership Model: The study shifts from a "deficit" model to an asset-based model, establishing local hubs as leaders rather than mere "test sites." It includes formal budgeting and leadership roles for local service user organizations (MHSUA-Ethiopia).
• Cultural Adaptation Tools: Development of specific tools (e.g., interactive Padlets, adapted TIDieR templates) to track and document the adaptation process, which can be shared for future digital mental health innovations.
• EbE Leadership: Demonstrates a robust model of "Experts by Experience" involvement, from conception to dissemination, including capacity building for local researchers.
• Software Localization: Technical adaptation of the avatar software to include culturally specific visual and auditory representations relevant to South Asian and East African populations.

4. Results

• Status: As this is a study protocol (preprint posted April 2026), no empirical results or clinical outcomes are available yet.
• Current Progress:
  • Phase 1 (Formative work) is reported as complete at both sites.
  • Qualitative research regarding stakeholder perspectives is ongoing.
  • Phase 2 (Case series testing) is scheduled to commence in October 2026.
• Anticipated Outcomes: The study aims to generate data on feasibility, acceptability, safety, and preliminary efficacy indicators to inform future scaling.

5. Significance

• Global Mental Health Impact: If successful, the study could provide a scalable, low-cost digital intervention for distressing voices in regions with severe shortages of mental health professionals (e.g., 0.01 psychiatrists per 100,000 in Ethiopia).
• Decolonizing Digital Health: The study addresses concerns of "digital colonialism" by ensuring LMIC partners lead the adaptation process, challenging the notion that Western interventions are universally applicable without modification.
• Policy and Implementation: Findings will inform the National Institute for Health and Care Excellence (NICE) and global health bodies regarding the deployment of digital therapies in diverse cultural contexts.
• Sustainable Legacy: The project aims to leave a sustainable legacy of peer-supported interventions, improved research accountability, and long-term funding for local peer roles in Ethiopia and India.

Conclusion: The AVATAR3 study represents a critical step in the globalization of digital mental health interventions. By prioritizing contextual adaptation and equitable partnership, it seeks to bridge the treatment gap for psychosis in LMICs while establishing a framework for responsible, culturally sensitive innovation.