Apps and adjuncts to treatment

One clear advance reflected in this group of articles is in the development of apps, specifically in what some apps can now do in terms of delivering CBT techniques (Erhardt et al., Reference Erhardt, Bunyi, Dodge-Rice, Neary and Schueller2022; Li et al., Reference Li, Achilles, Spanos, Habak, Werner-Seidler and O’Dea2022), or in their use as an adjunct to therapist-led treatment formats (Pacheco and Scheeringa, Reference Pacheco and Scheeringa2022). This represents a valuable route towards making CBT more accessible and widely available, particularly for people or groups who may be less likely to access traditional mental health services. However, although high-quality mental health apps exist, with engaging content and strong fidelity to evidence-based treatment principles, the app reviews (Erhardt et al., Reference Erhardt, Bunyi, Dodge-Rice, Neary and Schueller2022; Pacheco and Scheeringa, Reference Pacheco and Scheeringa2022) demonstrate how these may be hard to find given the sheer volume of mental health apps available. Pacheco and Scheeringa (Reference Pacheco and Scheeringa2022) found 722 apps in their initial search for mental health apps, of which only 163 met their criteria for having a substantive clinical element. Erhardt et al. (Reference Erhardt, Bunyi, Dodge-Rice, Neary and Schueller2022) aimed to identify apps implementing cognitive restructuring, but even this more targeted search returned 256 apps, of which only 15 implemented the cognitive restructuring technique. These studies also note that the information provided about apps is often insufficient to fully understand what the app provides and how. Independent platforms that rate and evaluate mental health apps (such as One Mind PsyberGuide, and the mHealth Index and Navigation Database ‘MIND’) are available, though may not be exhaustive.

Reviews such as Pacheco and Scheeringa (Reference Pacheco and Scheeringa2022) and Erhardt et al. (Reference Erhardt, Bunyi, Dodge-Rice, Neary and Schueller2022) are valuable in that they provide thorough and fine-grained analyses of what specific apps provide. The study of Erhardt et al. (Reference Erhardt, Bunyi, Dodge-Rice, Neary and Schueller2022) highlights that cognitive restructuring apps include most core elements of the technique, although they vary regarding additional elements such as asking for emotion intensity ratings. They make the interesting point that different apps may be more helpful depending on the specific needs of the user and note that improvements are needed due to variability in privacy protections and features to support user engagement.

Studies such as Browne et al. (Reference Browne, Carragher, O’Toole, Pimm, Ryder and Thew2022), Pacheco and Scheeringa (Reference Pacheco and Scheeringa2022) and Patterson et al. (Reference Patterson, Peynenburg and Hadjistavropoulos2022) reflect advances in how apps and other digital tools can be used as an adjunct to clinician-led interventions. CBT practitioners may be interested in knowing that apps are available to support clients’ self-assessment of mental health problems, implementation of treatment techniques, taking fidelity ratings of therapy sessions, and monitoring of clinical outcomes over time (Pacheco and Scheeringa, Reference Pacheco and Scheeringa2022). Importantly, treatment adjuncts do not always have to occur concurrently with therapy. Digital tools such as online peer support platforms (Browne et al., Reference Browne, Carragher, O’Toole, Pimm, Ryder and Thew2022) and self-directed online booster sessions (Patterson et al., Reference Patterson, Peynenburg and Hadjistavropoulos2022) can be used after treatment ends. Browne et al. (Reference Browne, Carragher, O’Toole, Pimm, Ryder and Thew2022) found that while not all participants reported regular use of the peer support platform, participants reported a benefit of simply knowing it is there. Such tools may therefore provide a sense of support which may be particularly valuable after treatment ends and clients are no longer receiving direct support from their therapist. Patterson et al. (Reference Patterson, Peynenburg and Hadjistavropoulos2022) found that the offer of a post-treatment booster session was accepted by around a third of participants, who on average found the session helpful and felt it would help them manage symptoms of anxiety and depression in future. The use of digital technology in this way may open new possibilities to support clients in consolidating their therapy gains and maintaining their wellbeing after treatment ends, which previously would not have been practical.

The current challenge regarding digital adjuncts to treatment is a significant lack of studies that evaluate whether these ‘add-ons’ confer a benefit over usual care. Of the 59 apps reviewed in Pacheco and Scheeringa (Reference Pacheco and Scheeringa2022), only 12 had supporting evidence from randomised controlled trials. The authors highlighted a need for more studies, and clearer information about empirical evaluations in app stores and descriptions. They also reported that few apps identified in their searches were in languages other than English. The development of innovations to be used alongside or following therapy is hugely valuable, but using robust study designs to evaluate their value is a vital next step before any wider implementation or dissemination can be considered. Where possible, these studies should include materials translated into other languages and people from groups under-represented in mental health services, including those who may be less familiar with or less confident in using digital tools.

Remote treatment delivery

Articles in this special issue also show advances in our understanding of whether remote treatment via videoconferencing or telephone is effective. Examining data from two IAPT services, Nguyen et al. (Reference Nguyen, McNulty, Grant, Martland, Dowling, King, Neely, Ball and Dom2022) found that the shift to remote treatment delivery during the pandemic was not linked to significant worsening of clinical outcomes across any demographic group or presenting problem, although there were some indications that older clients, and clients with social anxiety or health anxiety, may have achieved less benefit from remote therapy. One IAPT service showed a significant increase in their overall recovery rate after moving to remote treatment provision. Porter et al. (Reference Porter, Galloghly and Burbach2022) also found promising clinical outcomes in their evaluation of videoconferencing based therapy for young people in routine clinical practice, which showed medium to large effect sizes in anxiety and depression symptoms reduction and in progress towards personalised goals.

Given these positive findings, the studies also perhaps reflect advances in the skills and experience of practitioners who are delivering therapies using these formats. It is likely that many CBT practitioners had little or no experience of remote therapy delivery prior to the pandemic, so the present results are a credit to practitioners’ willingness to adapt in the face of significant adversity and change. Song and Foster (Reference Song and Foster2022) document a positive shift in some therapists’ attitudes towards remote treatment. They do, however, also suggest that further therapist training in remote treatment delivery may still be needed. Although some guidance on adapting CBT procedures for remote sessions is available (e.g. Cromarty et al., Reference Cromarty, Gallagher and Watson2020; Jassi et al., Reference Jassi, Shahriyarmolki, Taylor, Peile, Challacombe, Clark and Veale2020; Murphy et al., Reference Murphy, Calugi, Cooper and Grave2020; Warnock-Parkes et al., Reference Warnock-Parkes, Wild, Thew, Kerr, Grey, Stott, Ehlers and Clark2020; Wild et al., Reference Wild, Warnock-Parkes, Murray, Kerr, Thew, Grey, Clark and Ehlers2020), there remain gaps, and further work to document and evaluate best practice would be beneficial, including disorder-specific or transdiagnostic CBT protocols, working with groups or family units, or specific client groups, such as adolescents or older adults. Behavioural experiments may pose the biggest challenge and require more substantial adaptation for remote delivery. Perhaps it is then reassuring to read the positive experiences of patients and therapists with regard to remote experiments (Song and Foster, Reference Song and Foster2022). However, this study also highlighted that some therapists found experiments harder to do. It would be interesting to explore how therapists deal with this, whether this results in fewer behavioural components in treatment, and perhaps even whether remote administration could facilitate ‘therapist drift’ (Waller and Turner, Reference Waller and Turner2016) away from evidence-based therapy procedures.

The main evidence gap at present regarding our understanding of remote treatment is the lack of studies with appropriate control conditions. More robust study designs are needed, for example direct comparisons of in-person versus remote administration using the same therapeutic content. Clinicians need to better understand and be able to predict who does or does not benefit from in-person versus remote delivery. Telephone-delivered treatment was included in the Nguyen et al. (Reference Nguyen, McNulty, Grant, Martland, Dowling, King, Neely, Ball and Dom2022) study, and although it was not examined directly, the results from Service B, which replaced in-person work primarily with telephone sessions during the pandemic, indicated that clinical outcomes improved for low-intensity CBT interventions, but declined for high-intensity interventions. This warrants further investigation as it may aid the interpretation of reviews of telephone treatments (e.g. Coughtrey and Pistrang, Reference Coughtrey and Pistrang2018). Comparisons of telephone versus videoconference delivery would also be valuable.

User experiences

The present articles also demonstrate advances in our knowledge of, and approaches to, user experiences of digital CBT. Many of the studies (Browne et al., Reference Browne, Carragher, O’Toole, Pimm, Ryder and Thew2022; Li et al., Reference Li, Achilles, Spanos, Habak, Werner-Seidler and O’Dea2022; Patterson et al., Reference Patterson, Peynenburg and Hadjistavropoulos2022; Porter et al., Reference Porter, Galloghly and Burbach2022; Song and Foster, Reference Song and Foster2022) have incorporated at least an element of user feedback, which is perhaps a positive reflection of the value placed on this information as well as its importance in the development, refinement or delivery of interventions. Taking this one stage further, Li et al. (Reference Li, Achilles, Spanos, Habak, Werner-Seidler and O’Dea2022) demonstrate in their study advances in how co-design principles are now being used to involve the ‘end-users’ of a digital product in its development from the outset. This study describes the iterative process used, involving adolescents, parents and mental health professionals in the design of a CBT app, and highlights differences in preferences among end-users that need to be accounted for in app development (e.g. adolescents prefer an app they can use autonomously, while mental health professionals desire an app they can use as an adjunct to therapy). Li et al. (Reference Li, Achilles, Spanos, Habak, Werner-Seidler and O’Dea2022) highlight that co-design does not appear to be common practice, or at least that the reporting of co-design or user involvement procedures is lacking, citing a recent review that found this occurred in less than 30% of preventative mental health interventions for young people (Bergin et al., Reference Bergin, Vallejos, Davies, Daley, Ford, Harold, Hetrick, Kidner, Long, Merry, Morriss, Sayal, Sonuga-Barke, Robinson, Torous and Hollis2020). There is clearly scope for further improvement in the extent of meaningful user involvement as well as how this is described in the literature.

Another significant challenge is to ensure that the data gathered from evaluations of user experiences is representative. We must continue to improve our efforts to avoid sampling biases such that only the experiences of users who completed an intervention, or who were sufficiently motivated to attend user groups or complete surveys, are heard. Browne et al. (Reference Browne, Carragher, O’Toole, Pimm, Ryder and Thew2022) and Patterson et al. (Reference Patterson, Peynenburg and Hadjistavropoulos2022) made specific efforts to obtain feedback from those who had dropped out or were no longer using the intervention. This was informative in revealing the range of reasons that might be behind drop-out (such as forgetting about it, lack of time, not finding it helpful, or no longer needing it due to feeling better), but more could be done in future studies to examine specifically the experiences of those who disengage or stop using a particular intervention. This is critical to ensure that studies of user experiences or client satisfaction are not biased towards positive responses or missing the views of under-represented groups.

CBT research

The advances we have considered so far focus primarily on how digital methods can be used to deliver or enhance CBT interventions; but this set of articles perhaps also reflects advances in the use of digital tools to support service evaluation and research in CBT. Online administration of outcome questionnaires, surveys and focus groups are all demonstrated within this set of articles, which illustrate progress in how technology is being used to facilitate studies, and the collection of routine outcome data. It is promising to see that many of the present articles report on work conducted in routine clinical practice (Browne et al., Reference Browne, Carragher, O’Toole, Pimm, Ryder and Thew2022; Nguyen et al., Reference Nguyen, McNulty, Grant, Martland, Dowling, King, Neely, Ball and Dom2022; Porter et al., Reference Porter, Galloghly and Burbach2022; Song and Foster, Reference Song and Foster2022), and it may well be the case that these studies have been made more achievable due to the technological advances that support data collection. Beyond advancing digital CBT, the studies serve as valuable examples of how to collect data in routine care and use this data collection to iteratively improve services.

However, as we make greater use of digital technology to develop, deliver and research CBT interventions, it is critical to ensure that these methods do not exclude people. With regard to psychological therapies, ‘digital exclusion’ occurs when people are not equally able to access mental health services, receive psychological support, or participate in research, because they do not have access to the internet or relevant devices, or they do not have the skills or confidence to use this technology. There is evidence that the latter may be worsened by mental health problems (Greer et al., Reference Greer, Robotham, Simblett, Curtis, Griffiths and Wykes2019). This issue also applies to CBT practitioners, who may feel hesitant or lacking in skills to use digital technology effectively in their work. Developing a fuller understanding of digital exclusion in relation to CBT practice and research, and finding ways to minimise this inequality, is an advance much needed in the field. Lastly, another topic that warrants further research is the potential negative effects of delivering CBT methods digitally, which should be explored and reported more extensively to ensure patient safety and minimise unwanted and adverse events (Rozental et al., Reference Rozental, Andersson, Boettcher, Ebert, Cuijpers, Knaevelsrud, Ljótsson, Kaldo, Titov and Carlbring2014).