Should we be using digital technologies in the treatment of psychotic disorders?

The use of technology in the treatment of common mental health disorders has grown exponentially over the last few years. There are a number of reasons for this including, wide availability and acceptability, especially for young people. Examples in an Australasian context, where technological approaches to mental health treatment has been pioneering, are the use of online cognitive behavioural therapy (CBT) programmes and mobile phone applications to treat depressive symptoms and other mental health difficulties. However, the use of technology in the treatment of psychotic disorders (including positive symptoms, negative symptoms, relapse prevention, cognitive impairments and social functioning) has been less well researched despite similar potential benefits. This may be because there are concerns that people with psychosis will not use technology or even that the technology may exacerbate symptoms, such as paranoia.

What types of technologies are available to use as treatments in psychosis? There is burgeoning research on a variety of platforms. Online treatments, mobile phone applications, short message service (SMS) contact, experience sampling methods, therapeutic games, virtual therapists and virtual or augmented reality have all been trialled as therapeutic approaches. In regard to the specific opportunities and challenges, we will discuss two particular technological interventions (social media and virtual reality [VR]) we are developing for young people with psychosis. Most technological approaches raise similar challenges and dilemmas, although there are some specifics for the type of technology used.

The first barrier to address in using technology to treat psychosis is the assumption that people with psychosis may not use or have access to technology. There is growing evidence to suggest both these assumptions are not true. Recent evidence suggests that people with psychosis are likely to engage with technological interventions; for example, rates of mobile phone ownership are high, views on mobile health interventions positive and social media habits similar to that of their peers. Indeed people with psychosis currently use technology to help with a number of issues related to their psychotic disorder (Gay et al., 2016).

The first specific technological application we will discuss is the use of social media platforms to monitor and deliver therapy. We have delivered a novel peer-to-peer social media intervention to young people with psychosis (Alvarez-Jimenez et al., 2013). The intervention (HORYZONS) merged psychosocial intervention modules, online social networking and peer-to-peer and professional moderation (‘coaches’) in an integrated platform. The effectiveness of HORYZONS in improving social functioning and reducing relapse rates beyond the timeframe of specialist first-episode psychosis services is currently being evaluated through a randomized controlled trial (RCT) (Alvarez-Jimenez et al., 2013). Social media platform have wider applications beyond delivering individual psychological interventions. For example, we have also begun an RCT using a similar social media platform to deliver interventions to carers in early psychosis (Gleeson et al., 2017).

Recent reviews have highlighted that while social media–based interventions provide a promising avenue to improve clinical and social outcomes in psychosis treatment, most prior studies consisted of pilot and preliminary evaluations of newly developed interventions. As a result, there is a need for well-powered, controlled studies, with iterative designs, which determine the benefits as well as the potential risks of online social media interventions in psychosis treatment.

Online social media interventions provide fascinating new prospects in psychosis treatment. Powerful analytic approaches such as natural language processing and machine learning are increasingly available. Using these techniques, real-time individualized predictions of disengagement, depression, social isolation and psychotic relapse make the delivery of tailored interventions a realistic prospect (Alvarez-Jimenez et al., 2013).

The second specific area where we are gathering experience is the use of VR to monitor symptoms and provide treatment. VR has been previously used in a number of mental health disorders, most notably the treatment of phobias and post traumatic stress disorder (PTSD). These environments include virtual worlds and computer-generated avatars although immersive or 360° video and photography are also within the scope of VR. This area of research is rapidly developing, especially as technology advances and the cost of such technology makes these devices more accessible and more widespread. Until recently psychosis researchers have focused on the use of VR as a research paradigm to study symptoms such as paranoia. Studies are now exploring using VR experiences to deliver CBT in patients with psychosis (Freeman et al., 2016) and ‘avatar therapy’ (where a created avatar is used to embody and challenge auditory hallucinations in patients) with promising results (Craig et al., 2015). We have started a research programme using virtual worlds and 360° videos to deliver social cognition training in early psychosis. It appears that VR can be used to investigate symptom development as an experimental paradigm and perhaps to deliver interventions. There is also the potential for such tools to be used to monitor symptom response and aide diagnosis.

While these areas of research are exciting, there are a number of potential concerns to consider. First, the use of technological devices, including VR, in vulnerable people with psychosis may exacerbate symptoms such as paranoia or specific delusional beliefs related to technology or surveillance. Colleagues working in this area have built into their protocols specific familiarization time for patients to ‘check out’ the equipment and make sure they are comfortable with its operation. In the limited trials in this area to date, there are no data to suggest that VR technology exposure worsens symptoms (Freeman et al., 2016).

Second, we currently do not have data regarding the necessary level of exposure to VR to affect change or to affect individuals adversely. It could be that exposure to VR worlds that are too like the real world is actually less effective that those that are ‘real enough’ but not too confronting. There is a well-cited hypothesis in VR and robotics research called the ‘unhappy valley’ effect which suggests that some people are less likely to interact (or even have a sense of unease) with VR characters if they are too lifelike or ‘almost human’. Contrary to that, it may be that exposure to VR environments that are not familiar to the individual or do not have any personal relevance may not be either therapeutic or harmful as they don’t ground an individual in their own reality. We are particularly interested in developing VR interventions that have the ability to be personalized by the therapist and the patient to provide the most relevant experience.

Finally, it is the contention that any intervention delivered to sufferers of complex disorders such as psychosis, and one that is essentially social in nature, needs to be delivered or at least complemented via human interaction and not rely on technology. Few interventions rely solely on the technology and most have an element of expert human therapeutic involvement from trained therapists or encourage interaction outside of the virtual world. The exact role and relevance of these human interactions in technological interventions in psychosis needs to be appropriately investigated in these types of trials.

The use of technology in psychosis has much potential to provide insights into symptom development, aide diagnosis and deliver therapy to patients with psychosis. However, care needs to be taken that technology is used as an adjunct to good clinical care and not in isolation or a cheap alternative. Researchers need to properly evaluate potential harms of technological interventions and develop established protocols to maintain patient safety, distinguish which patients are likely to engage with and benefit from such interventions and those in which it may be contraindicated as well as investigating generalization of any benefits to real-world settings. Methodologies to evaluate the outcomes of these interventions need to take into account the iterative and developing nature of these technologies.