Mindset XR Module 20: Case studies

When highlighting the clinical need for a XR solution, it is important to understand how and why that need is not currently being met.

In this case, anhedonia - loss of pleasure - is a transdiagnostic problem. This means it is a problem across a group of mental disorders.

It is important to note that should a threshold for anhedonia be met, a diagnosis for severe conditions such as psychosis or depression is more likely to be made.

At the start of a study, it is important to identify gaps in the literature. For this study, it was found there were no first-hand accounts of anhedonia across depression and psychosis.

With no voice of the people with lived experience, we do not know what it feels like to have anhedonia.

Most of the existing tools:

This has led to:

To address these gaps, especially the lack of first-hand accounts, a qualitative interview study was undertaken to explore experiences of anhedonia.

To overcome the limits of existing tools, a VR scenario needs to be developed which can reliably evoke pleasure.

A VR solution is preferred as it can go further than standardised images and can appeal to a diverse group of people.

The new VR tool informed by lived experienced accounts, means the field can move forward because:

This pilot trial study looks to use XR for people with psychosis, a severe mental health condition. It can emerge in early adulthood and for most people it has lifelong implications.

People with psychosis experience different symptoms. Both pharmacological and non-pharmacological interventions are targeting symptoms such as hearing voices and delusions.

The majority of people with psychosis experience issues with motivation and experiencing pleasure, including social contact, and have restricted display of emotions.

These difficulties may lead to poor treatment outcomes as they are a major driver of:

We wanted to use VR as we believe this medium could:

When developing the VR product, we intended to use therapy with lived experience input from the start. We used that lived experience input to create the prototype of the environment and the script of interaction with the avatar.

The VR element was part of a therapy protocol that also required therapist input. This was due to the high level of disability often found in the clients that we were planning to offer this therapy to.

We created a therapy package called V-neST. This VR prototype went through different waves of usability testing by people with lived experience.

Over time, the VR software was refined to a product that our users, the research team, the technical team and the clinicians involved were all happy with.

After this, we felt the product was ready to be tested in a pilot randomised control trial.

It was agreed that there would be one session a week for 12 weeks lasting one hour.

The VR software supported various elements of the therapy work including psycho-education behavioural activation, pleasure forecasting, recover goal work, emotional literacy as well as real-life exposure to distressing symptoms – all supported by the therapist.

Five environments were created, each targeting one or more hypothesised mechanisms affecting motivation and pleasure difficulties including elements targeting:

The five environments were a:

The pilot trial was a single blind randomised control trial testing v-neST on 30 people with psychosis with poor motivation and loss of pleasure of daily activities.

Assessments were carried out at the outset and three months post-trial. Secondary outcomes were increase in pleasure and motivation and enhanced functioning.

They study aims were to see if:

The outcomes were:

After completed the therapy, we interviewed participants and recorded interviews to analyse the results qualitatively. This is an important step to make changes to the therapy and the research procedures.

From these interviews, the following areas emerged:

This is a pilot trial not powered to assess the efficacy of an intervention. However, it is good practice to estimate the effects of an intervention to see if they are going in the expected direction.

We can assess effects using standardised co-efficients so that they can be compared across outcomes and with other interventions.

We don’t just measure effect size but the likely range of an effect. For example, for the primary outcome of this study, personal goals attainment, there is a wide range indicating that improvement in personal goals is likely.

All other measures assessed problems in motivation or functioning. So negative results are an indication of improvement as it is a reduction in problems. Best practice recommends the results be available to the wider scientific community and general public.

Avatar therapy is designed for people who experience distressing voices in the context of psychosis. In this therapy, the person is supported to create an avatar who embodies the main distressing voice. They can customise how the avatar looks and sounds. They engage in series of face-to-face dialogues with the avatar facilitated by a therapist using voice-conversion technology.

This therapy aims to reduce the anxiety and distress associated with hearing voices and increase feelings of power, control and confidence.

The trial was developed on laptop screens but there is ongoing intern work looking at the impact of conducting the therapy in fully immersive VR environments.

The intervention was first developed in 2013. At this time, it established proof of concept for the method of speaking with digital avatars.

As a pilot study with only 26 participants, the first study was not designed to address questions of efficiency. However, validated measures were collected and the research identified encouraging signals for the therapy to be helpful for people who experience distressing voices.

The research then evolved into two randomised control trials (RCTs). At the first RCT called the Avatar 1 trial, 150 people were recruited and randomised to receive either avatar therapy or an active control condition called supportive counselling. Assessments were conduct at baseline, 12 weeks and 24 weeks.

The primary outcome was a reduction in voice-related distress at the therapy follow-up session at 12 weeks.

The key finding from the Avatar 1 trial was the therapy demonstrated a significant reduction in voice severity and distress at the primary time point compared to the control group.

The Avatar 1 trial had:

So, the next step was a multi-centred randomised control trial called the Avatar 2 trial. This aimed to optimise therapy delivery by comparing two versions of the therapy: a brief and an extended version.

The main research question centred on testing the effectiveness and cost-effectiveness of the two versions of Avatar therapy compared to standard care.

Importantly the trial tested effectiveness when the intervention is delivered across geographically and demographically diverse settings. This increases real world relevance of findings.

Data collection of the Avatar 2 trial has been completed but the results cannot be referenced until published in a journal. During the Avatar 2 trial, it was identified by the National Institute for Health and Care Excellence (NICE) Early Value Assessments as a promising digital technology.

The next steps will focus on the deployment of Avatar 2 within the NHS to gain real-world evidence.

Firstly, clinicians in routine NHS care settings would need to be trained, and it would be part of the therapy offer to service users.

Longer term follow-up data ideally over a 12-month period is needed on clinical and costs effectiveness at baseline, 12 months and 24 months.

This offer is not part of a randomised control trial but an open choice for people receiving therapy for distressing voices. The feedback could be compared to the experiences of those receiving standard therapy.

If an external control comparison was made, it could look at:

The results could include: