Digital health technologies to help manage symptoms of psychosis and prevent relapse in adults and young people: early value assessment

Mental health services are in high demand and access varies widely across the NHS. Because of this high demand, many people are not getting the treatment and support they need. The clinical and patient experts advised that access to cognitive behavioural therapy for psychosis (CBTp) varies and is limited for some people. Most adults with psychosis who are having treatment outside of early intervention in psychosis (EIP) services do not have access to psychological interventions. Access to therapy may be limited by NHS workforce pressures, including not having enough trained staff to deliver CBTp in community mental health teams. The committee recognised that managing symptoms in young people may be different from adults. Digital health technologies offer another option for people with psychosis who may otherwise not have psychological interventions. Mental health professionals would not need specialised training in CBTp to deliver digital health technologies for managing symptoms of psychosis. This may then widen the range of appropriate mental health professionals who are able to deliver it, in comparison with standard care. Digital health technologies may also reduce the number of sessions needed when used with standard care CBTp.

Monitoring for relapse prevention varies across NHS services. People usually have regular follow-ups and psychiatric reviews, but the clinical experts advised that there is no formal relapse prevention process. They also noted that relapse prevention could reduce acute psychosis episodes and associated hospital admissions. Digital health technologies could help people to better monitor their symptoms of psychosis. They could also detect relapse earlier than standard care, so people can be offered treatment and support sooner.

All the technologies had relevant published evidence showing a potential benefit for adults with psychosis. The relevant evidence consisted of 12 studies reported in 13 publications, specifically 6 randomised controlled trials (RCTs) and 6 sub-studies. There was evidence comparing each technology with treatment as usual, and AVATAR Therapy with supportive counselling. But there was no evidence comparing AVATAR Therapy or SlowMo with CBTp. The EAG reported that there was good-quality evidence from large studies suggesting that AVATAR Therapy and SlowMo were effective at reducing the specific targeted symptoms. AVATAR Therapy was found to reduce auditory verbal hallucinations in adults with psychosis, while SlowMo reduced paranoia and delusions. The evidence suggested that these reductions can last up to 24 weeks after intervention. Both technologies also improved quality of life. The EAG considered that there was some evidence suggesting that CareLoop was effective at detecting and reducing relapses. People who used CareLoop were also less worried about having a relapse than people who had treatment as usual.

There was no evidence on the effects of using the technologies in young people. Clinical experts advised that evidence from adults, many of whom have had psychosis for years, was not generalisable to this younger population. So, the committee considered that the benefits and risks for young people were unknown. It concluded that research was needed in young people before the technologies could be used in the NHS for this age group.

There is some economic evidence, based on clinical trial data in adults, that shows cost effectiveness for AVATAR Therapy and CareLoop. Preliminary results of the EAG's early economic modelling for CareLoop suggested that the technology is more effective and less costly than standard care. The EAG said that there was not enough data to adequately populate an economic model for AVATAR Therapy and SlowMo. The cost consequence analysis for AVATAR Therapy and SlowMo showed that the staff time needed to deliver the intervention was the key driver of costs for both technologies. The committee concluded that more evidence was needed on healthcare resource use.

Digital health technologies to help manage symptoms of psychosis or prevent relapse must be delivered or supported by mental health professionals trained in the technology. All of the companies provide training to mental health professionals on how to use the technologies. For AVATAR Therapy, this includes self-directed training followed by supervised use of the technology in practice. For SlowMo, training is around 1 to 3 days depending on the person's level of experience. Training for both technologies has been designed for use by mental health professionals without specialist training in CBTp.

The evidence suggested that digital health technologies are acceptable and have good adherence in adults with psychosis who choose to engage with them. But the EAG considered that more evidence was needed on why people turned down or did not complete the interventions. There were also uncertainties around whether people would need extra sessions and the effectiveness of repeating the interventions after relapse. The company for AVATAR Therapy said that the trials were not designed to offer additional sessions beyond the protocol, but this could be considered in clinical practice. The company for SlowMo said that this was being explored in an implementation study. For SlowMo, people will be able to continue using the app after the sessions with a mental health professional have ended. The committee concluded that more evidence was needed on the long-term effects of the technologies, including the effects of repeat use.

For CareLoop, training is done at an individual NHS service level and includes onboarding, categorising early warning signs of relapse, how to support people with psychosis using the app, and what to do when early warning signs and relapse occur. The committee concluded that services should have a system in place that provides adequate and timely professional support in response to alerts and outputs from people using the technology.

The committee carefully considered the safety of using digital health technologies to help manage symptoms of psychosis or prevent relapse while further evidence is generated. Four studies reported adverse events, 1 study on AVATAR Therapy, 1 on SlowMo and 2 on CareLoop. Unpublished results from the AVATAR2 trial were also shared with the committee as academic-in-confidence evidence. For all the technologies, there were a few serious adverse events that were possibly related to the technology. The clinical experts advised that adverse events may occur with any treatment when managing severe mental illnesses such as psychosis, regardless of the delivery method. Services should have protocols for delivering digital health technologies, including for initial clinical assessment, matching the right treatment to people's needs and preferences, and ongoing monitoring and management of people's safety. CareLoop uses an algorithm to recognise worsening mental health and potential relapse. The committee considered that services would need staff and resources to monitor and respond to these alerts, and to escalate care when needed. It also agreed that adverse event data should be collected for these technologies as part of further evidence generation to help assess and understand which are related to the technologies.

The committee considered that it was important to identify who may benefit most from using digital health technologies to help manage symptoms of psychosis or prevent relapse. Each technology for managing symptoms of psychosis is indicated for specific symptoms, so mental health professionals should assess if each technology is suitable for each person's symptoms. The clinical and patient experts advised that some people with psychosis may have persecutory delusions or triggers related to digital technology. This was supported by the responses to a small patient survey. Some people who completed the survey said that they have difficulty using digital technology when their symptoms worsen. Only a few people who completed the survey had been offered digital health technologies to help manage their psychosis, with some not finding it helpful. The patient experts highlighted the need to have a range of treatment options available.

Digital health technologies could increase access to care by providing another option for people with psychosis. Patient experts advised that people with mental health conditions sometimes experience shame. They may face a lot of stigma and discrimination, and this is more prevalent in areas of social deprivation.

A clinical expert suggested that some people may have negative views of mental health services. They suggested various reasons for these views, for example people from ethnic minority backgrounds are more likely to have experienced restrictive interventions while in hospital. They may also be less likely to have psychological therapy, or when they do, are more likely to have fewer sessions. A patient expert advised that some people from ethnic minority backgrounds engage with services later and may prefer using digital health technologies. Some people may feel that services are not adapted to understand and accommodate some aspects of their culture. People's ethnic, religious, and cultural backgrounds may affect their views of digital health technologies. Some people would benefit from digital health technologies in languages other than English. Healthcare professionals should discuss the language and cultural content of the technologies with people before use.

Additional support and resources may also be needed for people with visual or hearing impairments, cognitive impairment, problems with manual dexterity, a learning disability, or who are unable to read.

Digital health technologies may not be suitable for everyone. They are delivered using a smartphone, tablet or computer. For monitoring technologies such as CareLoop, people need regular access to a device with internet access to use the technologies. People with limited access to these technologies or who are less comfortable or skilled at using digital technologies may be less likely to benefit. Hardy et al. (2022) found that people's experiences of using SlowMo were not affected by their level of digital literacy. But adherence was associated with people using smartphones more frequently at baseline and being more confident using them.

For all the technologies, there were evidence gaps related to the population, intervention, comparators and outcomes. The committee considered that there were uncertainties about the clinical and cost effectiveness of digital health technologies to help manage symptoms of psychosis or prevent relapse because of the limited evidence. There was enough evidence of potential benefits of all the technologies for adults with psychosis for them to be used in the NHS while further evidence is generated. Important evidence gaps for the technologies are:

• Population: the relevant evidence for all the technologies was in adults. There was no evidence of the effects of using the technologies in young people. So, research is needed on the benefits and risks of using digital health technologies for this age group. The EAG advised that evidence is also needed on using the technologies in people with newly diagnosed psychosis.

• Intervention: there was limited evidence for all the technologies. There was only 1 fully powered RCT for AVATAR Therapy and SlowMo, and 1 feasibility RCT for CareLoop. There are ongoing studies for all 3 technologies that may address their evidence gaps.

• Comparators: for AVATAR Therapy and SlowMo, a key evidence gap was the comparators used in the trials. The most common comparator was treatment as usual but there was no evidence comparing either technology with CBTp. Both technologies can also be used with CBTp. So, evidence is also needed on their effectiveness when used in addition to standard care.

• Outcomes: for all the technologies, evidence is needed on change in target psychosis symptoms managed by the technology, long-term impact, adverse events, relapse rates, real-world implementation including experiences of staff and people using the technologies, adherence, completion, clinical and functional outcomes and resource use. For CareLoop, this should include data on resource costs associated with relapse, such as hospital stay costs.