The Impact of Digital Technology on Psychological Treatments and Their Dissemination

The psychological treatment of mental health problems is beginning to undergo a fundamental change. This change is being driven by the widespread availability of “digital technology” by which we mean computers, the internet, mobile devices such as smartphones, and mobile software applications (apps). In this paper we describe the various uses of digital interventions and consider their likely impact on clinical practice, clinical services and the global dissemination of psychological treatments.

There are well established digital treatments for depression and most of the anxiety disorders, and for problems such as insomnia (Andersson & Titov, 2014). The great majority are self-help programmes designed either to be used on their own or with some form of support. These treatments vary markedly in their content, clinical range, format, functionality and mode of delivery.

Digital technology provides new means of assessing and tracking psychopathology. At the simplest level, it can improve both the administration and interpretation of assessment questionnaires which, until recently, have been largely in pencil-and-paper format and manually scored. Many questionnaires are now available in digital form and can be completed via a website or app. This allows them to be automatically scored and interpreted with reference to established norms, and in some instances the scores can be transmitted directly to the clinician, the user and the user's clinical record.

The psychometric performance of digital questionnaires is a matter of concern as their presentation often differs from that of the source instrument. Overall, their performance appears to be similar to that of their pencil-and-paper counterparts but exceptions have been found (Alfonsson, Maathz, & Hursti, 2014; van Ballegooijen, Riper, Cuijpers, van Oppen, & Smit, 2016). Little is known about the performance of app-based questionnaires. Some degree of standardisation of digital questionnaires seems desirable, perhaps by instrument developers creating “approved” digital versions, ones which have a consistent presentation across different devices.

Digital technology opens up new modes of assessment. Virtual reality procedures can assess sensitivity to particular environments (Freeman, 2008) and the presence of sensors in smartphones makes it possible to track many phenomena on an ongoing basis including sleep, movement, physical activity, speech, device usage and the person's location (Abdullah et al., 2016; Saeb et al., 2015). How to use this information is only just beginning to be explored. It may prove possible to catch new episodes at a very early stage and supply interventions at the very time that users might benefit most (Faurholt-Jepsen et al., 2014). “Real time” intervention might be especially relevant to suicide prevention if markers of imminent risk could be identified (Christensen, Cuijpers, & Reynolds, 2016). However, psychopathology tracking is not necessarily benign. It can magnify rumination and self-focus, and it has been reported to trigger emotional instability (Murnane et al., 2016). Similarly, the immediate delivery of interventions might create reliance upon them which could interfere with the acquisition of selfmanagement skills.

Digital technology is increasingly being used for educational purposes. Some “massive online open courses” (MOOCs) attract huge numbers of students from around the world (Eichhorn & Matkin, 2016). If it were possible to train therapists in a similar way, this would be an important advance as the conventional form of training is not scalable and is therefore a barrier to treatment dissemination (Fairburn & Cooper, 2011).

Clinician training websites are starting to be developed and evaluated (McHugh & Barlow, 2012). Some are basic in form and constitute little more than brief lectures or segments of text combined with video clips of therapy. Others describe the treatment in great detail and illustrate it with extensive libraries of video clips. The more sophisticated training websites also incorporate learning exercises and formative tests to help trainees grasp key concepts and master challenging procedures. Digital training has many advantages. In addition to letting trainees see therapy actually being delivered, it can be accessed whenever and wherever it suits the trainee, and it can easily be updated. Digital training programmes can be used on their own or supplemented with support.

Therapist training has been neglected as a research area (Fairburn & Cooper, 2011; McHugh & Barlow, 2012). Digital training is changing this. Training research can now be conducted on a much larger scale than before as digital training allows large numbers of geographically dispersed clinicians to be trained simultaneously. For example, a training website for enhanced cognitive behaviour therapy for eating disorders has been tested in a proof-of-concept study across an entire country; a RCT across North America; and in a global cohort study involving over 750 clinicians.

Digital treatment has been the focus of an impressive amount of research given that this is a young field. Many of the website-based interventions have been tested in randomised controlled trials (RCTs) and their findings have been the focus of numerous systematic reviews (Andersson, Cuijpers, Carlbring, Riper, & Hedman, 2014; Richards & Richardson, 2012). Surprisingly, app-based interventions have barely been studied (Donker, et al., 2013b). Here is a brief summary of the main findings.

Not surprisingly, many important questions have yet to be answered. Here are some examples. First, as there have been few head-to-head comparisons of different digital interventions for the same mental health problem, it is not clear which ones are the most effective ones nor is their relative cost-effectiveness known (Donker et al., 2015). These are significant omissions since this information is needed by policy makers. Second, it is not known whether the functionality of a digital intervention has a bearing on its effectiveness. This too is an important omission as the answer has major implications for the design of future treatments. It is possible that the age of the user may be relevant with younger users complying better with more interactive treatments. The nature of the psychopathology being addressed may also need to be taken into account when designing interventions; for example, users with depression may struggle to complete interventions which require sustained concentration. In addition, there is a need for research on how these interventions work; who is accessing them; who benefits most; and whether the changes last. Also, more needs to be known about any negative effects of digital treatment (Rozental et al., 2014).

The evaluation of digital interventions poses intriguing challenges (Glasgow, Phillips, & Sanchez, 2014; Mohr et al., 2015). New interventions appear almost weekly, particularly app-based ones. Do all of them need to be evaluated or can one generalise from similar interventions? A related problem stems from the fact that digital interventions are easy to tweak. This raises the matter of at what point does a tweak result in what is essentially a “new intervention” that requires its own evaluation? The place of the conventional, somewhat ponderous, randomised controlled trial (RCT) is a particular concern in this fluid and fast-moving field. In the early stages of developing, testing and refining a digital intervention a variety of research designs may be of relevance (Collins, Dziak, Kugler, & Trail, 2014; Collins, Nahum-Shani, & Almirall, 2014; Glasgow et al., 2014; Lei, Nahum-Shani, Lynch, Oslin, & Murphy, 2012), but once an intervention has been reasonably well defined RCTs are likely to be needed to establish its utility.

It is far from clear how best to use digital treatments. To date they have been implemented in many different ways and in numerous different settings. Most have been accompanied by some form of support, but the nature of the support has varied greatly as have the qualifications and training of the person delivering it (who has been variously termed a “guide”, “coach”, “facilitator”, “therapist” or “clinician”). As a result, the research findings are difficult to integrate and interpret.

In Table 1 we propose a framework for classifying the ways that digital treatments may be used. It is based on whether the intervention is programme-led or clinician-led and, if there is external input, whether it is from a non-specialist practitioner or clinician.

Direct-to-user digital treatments have considerable potential as public health interventions (Fairburn & Patel, 2014; Munoz et al., 2016). Whether they are autonomous or accompanied by non-specialist support, they are scalable. If they are directly accessed via the internet, their potential reach is enormous and they may well circumvent many of the barriers to the receipt of help including stigma, shame, scarcity of local treatment resources and cost (Munoz et al., 2016).

A recent innovation is the online clinic, MindSpot being a leading example. It is a government-funded national online treatment service for Australian adults with anxiety or depression (Titov et al., 2015). It provides information and advice via email or over the telephone and, if indicated, it supplies digital treatments with or without external input. Online clinics like MindSpot are likely to proliferate. They have many advantages for users (for example, ease of access, and the convenience of having treatment sessions when and where it suits them) and for healthcare systems (for example, high patient throughput at low cost).

In standard clinical settings digital interventions may be used in a range of ways. Unguided and supported digital treatments are well suited for non-specialist settings such as primary care, although in practice uptake and use can be poor (Gilbody et al., 2015). In specialist settings blended treatment is likely to be the preferred option as the patients will be expecting input from a clinician.

A barrier to the use of blended treatment is the absence of guidance regarding when and how to implement it (Wentzel et al., 2016). This is important as blended treatment can take many forms, not all of which are advantageous (Kenter et al., 2015). For blended treatment to be rationally used and well implemented, it needs better operationalisation and further study.

The challenges to disseminating empirically supported psychological treatments are global; to our knowledge, their effective coverage does not exceed 50% in any country. Even this figure is likely to be a significant over-estimate. The coverage in low and middle income countries is much lower. A recent estimate from India and China, two relatively well resourced middle income countries, revealed a treatment gap exceeding 90% for common mental disorders and alcohol use disorders (Patel et al., 2016), the two mental health conditions for which psychological treatments are recommended as first line interventions by the mhGAP program (WHO, 2008).

Global mental health innovators have attempted to address two major barriers to reduce this gap, viz. their low acceptability due to contextual differences between the settings in which psychological treatments were developed and those in which they are to be used, and their low feasibility due to the lack of mental health professionals to deliver them (van Ginneken et al., 2013). This body of research has shown that psychological treatments are effective in widely different cultural and social contexts even when delivered by people with little or no prior mental health training (van Ginneken et al., 2013). However, there remain two significant barriers: the continuing reliance on face-to-face formats for training and supervision, and the low demand for mental health care in formal health care settings, not least due to the high levels of stigma attached to mental health problems. Digital technologies offer a genuine opportunity to leap-frog both barriers.

The Mental Health Care Innovation Network (www.mhinnovation.net), the largest repository of innovations in global mental health, has identified 38 innovations that are using digital technology to disseminate empirically supported interventions (24th April 2016). These include a wide range of approaches with distinct goals including the direct delivery of treatments such as cognitive behaviour therapy; social networking for mental health care provider supervision; digital platforms for service users to share experiences and support one another; case management systems for monitoring the delivery of care; telemedicine systems to provide mental health care to remote populations or supervision to front-line workers; screening and decision-support tools for front-line workers; video game interfaces to address psychopathology targets; and text messaging to motivate patients.

A systematic review of the evidence base on interventions to disseminate psychological treatments (Naslund et al., in preparation; protocol registration number: CRD42015027179) identified 44 studies that used digital technology for the treatment, diagnosis, or management of mental disorders, or for providing mental health training and education to health workers. The review found that these methods were well accepted and while it was concluded “that mobile and online technologies may hold significant potential to extend workforce capacity, reach forcibly displaced persons and refugees, support young people with mental disorders who are high users of mobile devices, and empower patients and families to take charge of their own care and to support each other through online communities” it was also noted that there has been little formal evaluation of their effectiveness. This is possibly the most important immediate priority for the field of global mental health, i.e., to identify best practices in each of the domains of technology-assisted delivery of psychological treatments.

There are a number of other challenges which will need to be addressed in the process of dissemination. Foremost amongst these are language (the vast majority of digital innovations are in English); the restricted coverage of internet-enabled devices; limited internet literacy in vast sections of the global population; and the lack of clearly defined regulatory procedures to ensure privacy and confidentiality of digital health data. Notwithstanding these limitations and challenges, we are bullish about the prospect of digital technology being transformative in improving the global availability of psychological treatments. Our enthusiasm is influenced by several factors: the ingenuity with which digital technology is being applied in diverse ways for diverse goals; the demonstrated successes of digital technology in a variety of other health care domains; the rapid growth in internet coverage and internet literacy, in particular among young people who are potentially the most important group for targeting psychological treatments for common mental health problems; and the ever-increasing speed of data access and the reducing cost of internet-enabled devices.

While the “digital divide” undoubtedly remains a problem, particularly in low-resource settings, the divide is closing and there is no reason to think that this will not continue. Digital interventions that can be used without support are of particular importance as they have enormous potential to improve access, and additionally they have the value of being inherently empowering. They need to be optimised and “task sharing” needs to be expanded to embrace digital self-help. National and international organisations concerned with mental health need to endorse and support digital technologies as they are likely to be transformative. Above all, the international psychological treatment community must strive to engage digital entrepreneurs and innovators, particularly those who are championing initiatives in global health, to partner with them to exploit the many opportunities for using digital technology to transform mental health care worldwide.

Over the next decade or two much is likely to change. Digital interventions will gradually find their place within mental healthcare systems, and online clinics will become more commonplace. Digital assessment and treatment are likely to merge. Blended treatment may displace some conventional face-to-face treatment, and the limitations and negative effects of these innovations are likely to become evident. It is to be hoped that systems for evaluating, regulating and promoting these interventions will be developed thereby accelerating their appropriate use.