A qualitative exploration of adolescents’ experiences of digital Dialectical Behaviour Therapy during the COVID-19 pandemic

Abstract

The UK government implemented national social-distancing measures in response to the global COVID19 pandemic. As a result, many appointments in the National Health Service (NHS) took place virtually, including psychological interventions in out-patient settings. This study explored the experiences of adolescents participating in a dialectical behaviour therapy (DBT-A) programme via teletherapy (i.e. via video or telephone call) in a Children and Adolescent Mental Health Service (CAMHS). Thirteen adolescents with emotion dysregulation and related problems completed an online qualitative survey about their experience and acceptance of DBT-A delivered virtually. Thematic analysis was conducted on the survey data and generated three over-arching themes: (1) sense of loss; (2) feeling uncontained; and (3) benefits of virtual DBT. These over-arching themes were composed of eight subthemes (‘loss of connection with group and therapist’; ‘loss of skills-building opportunities’; ‘limited privacy’; ‘lack of safe therapy space’; ‘difficult endings’; ‘home comforts’; ‘convenience and accessibility’; and ‘easier to participate with others’). This study suggests that adolescents doing virtual DBT-A need approaches that acknowledge and address the additional relational, emotional and practical challenges of online therapy while maintaining fidelity to the evidence-based treatment model. Suggestions for further research and preliminary practice guidelines are discussed.

Key learning aims

1. (1) To learn about the experiences of adolescents participating in a DBT programme for adolescents (DBT-A) conducted virtually, including the challenges and benefits they identified.

2. (2) To learn about implications for clinical practice and future research directions.

Introduction

The COVID-19 pandemic, which escalated in early 2020, led to sudden shifts in the delivery of mental health services due to strict social distancing and lockdown measures (Gruber et al., 2020). In response to these challenges, many mental health services incorporated various forms of virtual therapy into their service delivery models to balance social distancing restrictions with clinical demand (Gruber et al., 2020). When it came to working with adolescents who suffer with severe difficulties with emotion regulation, there was limited research to inform the transition from face-to-face services to virtual service delivery. Specifically, the benefits and challenges of offering services virtually were unknown, and thus became the focus of the current study.

Recent evidence suggests that the pandemic has had a substantial influence on children and adolescents’ mental health (Ougrin et al., 2021; Rogers et al., 2020). Adolescents can experience a range of mental health challenges, including self-harm and suicidal behaviours (Hawton et al., 2012). Suicide is the fourth leading cause of death in children and adolescents (World Health Organization, 2020), and a history of self-harming behaviour is a critical risk factor for completed suicide in adolescents (Hawton and Harris, 2007). Since 2010, there has been a 68% increase in adolescent self-harm and a 67% increase in adolescent completed suicide rates in England and Wales (Office for National Statistics, 2019). In the past year, the proportion of children and adolescents presenting with self-harm at emergency departments increased from 50 to 57%, and from 58 to 66% for subgroups with self-harm and emotional disorders (Ougrin et al., 2021). Self-harm is considered to be a significant problem in several mental health disorders, including borderline personality disorder (Linehan, 1993).

Borderline personality disorder

The range of difficulties experienced by people with a diagnosis of borderline personality disorder (BPD) may present additional treatment challenges when therapy is delivered virtually. BPD is a diagnosis used to describe a pattern of instability in personal relationships, intense emotions, poor self-image, and impulsivity (American Psychiatric Association, 2013). The diagnosis is made based on the presence of at least five of the following problems: fear of abandonment/rejection; unstable or changing relationships; unstable self-image/struggles with identity or sense of self; impulsive or self-damaging behaviours; suicidal behaviour or self-injury; mood instability; chronic feelings of emptiness; problems with anger; and transitory, stress-related paranoia or loss of contact with reality (American Psychiatric Association, 2013). This same range of problems is also often diagnosed in the UK public health system as emotionally unstable personality disorder (EUPD) (World Health Organization, 2019). In this paper, we use the term BPD when referring to the literature, for consistency and in accordance with the diagnostic system and terminology most often employed by the authors we have referenced.

A growing body of literature supports the validity and value of using BPD diagnoses in adolescents (Chanen et al., 2008; Miller et al., 2008), although controversies around diagnosing BPD in adolescents have been highlighted (Bleiberg, 1994; Miller et al., 2008). Critiques of the BPD construct in both adults and adolescents have emphasised concerns relating to conceptual issues (Bleiberg, 1994), empirical support (Bondurant et al., 2004; Westen et al., 2003), and clinical utility (Bemporad et al., 1982). In addition, service users describe the experience of being diagnosed as feeling stigmatised, judged, marginalised and objectified (Castillo and Mind, 2003). In recognition of these controversies, particularly concerning the diagnosis of BPD in adolescents, and to be consistent with established practices in our clinical setting, we have chosen to describe our participants’ mental health difficulties without using diagnostic labels. Instead, we refer to adolescents with emotion dysregulation and related problems.

Adolescents with emotion dysregulation often experience related problems, including self-harm, suicidal behaviours, significant challenges in relationships (e.g. fear of abandonment and rejection), cognitive and neurodevelopmental vulnerabilities, and questioning of self-identity (Hawton et al., 2012).

Dialectical behaviour therapy

Dialectical behaviour therapy (DBT) is an intensive treatment, designed to address the complexity and severity of the difficulties associated with BPD, and the often-high levels of risk that can accompany these. ‘Standard’ DBT is usually delivered as a comprehensive treatment programme incorporating multiple modes that include weekly individual therapy sessions, a weekly skills training group, brief telephone coaching calls between sessions, and a weekly team consultation meeting for therapists (Linehan, 1993). Most of the treatment is delivered via the individual and group sessions, traditionally conducted in-person. Despite the potential challenges of providing these modes of the treatment virtually, there is a paucity of studies examining the use of teletherapy in DBT.

DBT is one of the most commonly used treatments for adults with BPD (Rizvi et al., 2011; Linehan et al., 2015). DBT has been demonstrated to be effective in numerous randomised control trials (RCTs) (Linehan et al., 2015; McMain et al., 2009). Meta-analyses of trials of DBT with adults diagnosed with BPD have demonstrated DBT to be superior to treatment as usual (TAU) in reducing suicidal and self-injurious behaviours (Kliem et al., 2010; Panos et al., 2014).

In the last decade, a number of studies have evaluated the effectiveness of DBT for adolescents experiencing emotion dysregulation and related problems (McCauley et al., 2018; Mehlum et al., 2014; Santamarina-Perez et al., 2020). However, there is a gap in the literature regarding the delivery of virtual DBT for this population. Miller et al. (2007) developed DBT for adolescents (DBT-A) experiencing emotion dysregulation and related problems and their parents or carers. Rathus and Miller (2002) adapted Linehan’s original model with some key differences in the structure. In DBT-A parents and carers can also access skills training groups, to improve family relationships and communication and help parents support adolescents’ ability to generalise their skills. The treatment is shorter, to help frame completion as achievable for a population with a typically high attrition rate. The volume of skills taught and the language used in group materials are reduced and simplified, to facilitate adolescents’ understanding of the skills training content (Rathus and Miller, 2002).

According to Koerner et al. (2007), adaption (as opposed to simple adoption) of DBT in different settings compromises its core therapeutic elements, as described by Linehan (1993), and the risk of reduced effectiveness, in addition to ethical issues, may have to be considered. Despite these concerns and controversies, DBT-A retains the core components of the standard DBT model and is supported by an emerging evidence base. DBT-A has been demonstrated to be more effective than TAU in reducing self-harm and suicidal ideation in adolescents in RCTs (Mehlum et al., 2014; McCauley et al., 2018; Santamarina-Perez et al., 2020) and meta-analyses (Bahji et al., 2021; Kothgassner et al., 2021). In the NHS, DBT-A is the recommended evidence-based treatment for children and young people with significant emotion dysregulation difficulties who have frequent episodes of self-harm (National Institute for Health and Care Excellence, 2022).

Despite the demonstrated effectiveness of DBT, over one in four clients in DBT leave treatment before completion (Dixon and Linardon, 2020). Studies exploring adolescents’ subjective experiences of DBT-A programmes are rare, yet these are critical for understanding the challenges they may encounter that can interfere with treatment engagement and effectiveness. Participants in a qualitative study of adolescents’ experiences in a DBT-A skills group (Pardo et al., 2020) stressed the importance of active therapist involvement in their motivation, engagement, learning and group cohesion. Furthermore, the study reported that participants felt less alone when they formed relationships with peers who shared similar struggles.

Internet-based mental health interventions

As the quality and accessibility of technological aids such as the internet and smartphone applications (apps) have improved, studies have increasingly been conducted to evaluate the efficacy of psychological therapies delivered virtually, including those based on cognitive behavioural therapy (CBT) (Andersson, 2009; Andersson et al., 2012; Donker et al., 2013; Spence et al., 2011), and DBT (Rizvi et al., 2011; Rizvi et al., 2016). Studies have also examined online self-help interventions with either low-level therapist involvement (such as reviewing homework assignments via email) or no therapist involvement (Andersson et al., 2012; Andersson and Titov, 2014; Storch et al., 2011).

There is extensive literature on the effectiveness of internet-based cognitive behavioural therapy (ICBT). ICBT consists of web-based modules with assigned homework or a task at the end of each module or the use of mobile self-help apps (Andersson, 2009). RCTs with adults have demonstrated the effectiveness of ICBT, with and without therapist support, for various psychiatric disorders, including panic disorder (Bergström et al., 2010), social anxiety disorder (Hedman et al., 2012; Hedman et al., 2013), mood disorder (Arnberg et al., 2014), and post-traumatic stress disorder (Sijbrandij et al., 2016), with effect sizes similar to face-to-face CBT delivery for several psychiatric disorders (Bergström et al., 2010). Studies evaluating the effectiveness of virtual therapy with a therapist via video for adults with psychiatric disorders have shown some encouraging preliminary results but are generally limited, with a distinct lack of RCTs so far (Gros et al., 2013).

Limited studies examine the effectiveness of ICBT for psychiatric disorders in children and adolescents. Smith et al.’s (2015) RCT of a computerised 8-week CBT intervention for adolescents found significant reductions in depression and anxiety. A qualitative study by Lenhard et al. (2016) described the experiences of adolescents with obsessive compulsive disorder (OCD) participating in ICBT. Participants described their experience participating in ICBT as supportive, flexible, and secure for expressing their feelings and thoughts (Lenhard et al., 2016). However, none of these studies explored interventions in which video conferencing with therapists was a core component. Existing studies also excluded participants with axis II disorders, whilst high levels of risk (i.e. suicidal ideation requiring face-to-face treatment) were also listed as an exclusion criteria (Andersson et al., 2012; Lenhard et al., 2016; Matsumoto et al., 2018; Storch et al., 2011). While the treatment success outlined in the existing literature is evident in its indication that there can be significant benefits to online therapeutic interventions, there are ultimately distinctions between participants engaging in ICBT in the existing literature and those requiring a DBT/DBT-A intervention, lending weight to the rationale for this study.

CBT and DBT-A are structured and manualised therapeutic approaches, which can be delivered via videoconferencing or other internet-based methods (Schroeder et al., 2018; Wilks et al., 2018). However, only a few studies have examined the effectiveness of internet-based DBT interventions. In these studies, smartphone apps were primarily used as an adjunct to a face-to-face standard DBT programme for adults with BPD (Rizvi et al., 2011; Rizvi et al., 2016; Wilks et al., 2018) as opposed to virtual therapy delivery of DBT core components. Rizvi et al. (2011) piloted using a smartphone app called ‘DBT Coach’ as an adjunct to a face-to-face standard DBT programme. This app includes content from all skills modules (mindfulness, distress tolerance, emotion regulation, and interpersonal effectiveness skills) and a diary card. Results indicated that emotional intensity and urges to use substances significantly decreased every time the app was accessed (Rizvi et al., 2011). Another study of the app within a standard DBT programme for adults found that it reduced subjective distress and urges to self-harm; however, it did not relate to any treatment outcomes (Rizvi et al., 2016). Schroeder et al. (2018) assessed the acceptability and feasibility of an app called ‘Pocket Skills’ for adults enrolled in psychotherapy (model undefined). Pocket Skills is similar to the DBT Coach app. However, it is web-based rather than a smartphone app (Schroeder et al., 2018). The study found that participants using Pocket Skills as an adjunct to their primary therapy reported increased self-efficacy, engagement, and significant improvements in depression, anxiety and DBT skills use (Schroeder et al., 2018).

Research regarding how virtual DBT/DBT-A is experienced is minimal. One recent qualitative study identified a range of challenges clinicians encountered offering standard DBT virtually for adults during the pandemic (Zalewski et al., 2021). In particular, participants noted difficulties assessing suicide risk and making clinical decisions in the moment without seeing someone in person. Many participants commented on novel therapy interfering behaviours (TIBs) their clients engaged in via telehealth, including clients logging off sessions early or turning off their video so they could not be seen. Participants reported challenges related to skills group participation and attendance, following standard individual therapy session procedures, responding effectively to client emotions, therapist burn-out, relationship challenges with consultation teams, and technology issues. Participants suggested addressing the range of challenges they had experienced when delivering DBT via virtual therapy, with a particular emphasis on managing the novel TIBs and maintaining fidelity to the evidence-based model. This study offered an opportunity to explore effective methods of delivering DBT via telehealth and highlighted the need for evidence-based clinical recommendations. However, the authors also acknowledged that the findings did not provide enough information to differentiate the lessons learned for delivering virtual DBT to adolescents (Zalewski et al., 2021). Furthermore, as the study only surveyed DBT providers, the authors recommended that future research further explore the experience of service users engaging in virtual DBT, the challenges they encounter, and their suggestions for improving the experience (Zalewski et al., 2021).

The current study aimed to explore adolescents’ perspectives on the challenges and benefits of virtual DBT-A and their suggestions for improving service delivery and service user experience. This study explored the following questions:

1. (1) What challenges and benefits have adolescents encountered in virtual DBT-A?

2. (2) How acceptable have they found doing virtual DBT-A?

3. (3) What suggestions do they have for improving the experience of virtual DBT-A?

Method

Given the aims of the current study, a qualitative paradigm was chosen. Qualitative methods are frequently used in psychosocial research to explore individual experiences, meaning and perspectives (Willig and Stainton Rogers, 2017). Qualitative methods are often employed in health settings for evaluation, policy development, implementation, and identifying meaningful outcomes (Berry et al., 2017; McBride et al., 2020;). In developing the study’s methodology, we drew on a realist/essentialist epistemology in which we aimed to capture the experiences, meanings and reality of participants expressed within their dataset (Braun and Clarke, 2006). As we were interested in exploring meaning at the more explicit level, we took an inductive orientation to our data in which theme development was driven by data rather than existing theoretical constructs. Thematic analysis (TA) is a commonly employed and flexible method in qualitative research (Braun and Clarke, 2022) and was the method of choice, given its compatibility with an essentialist paradigm and the philosophical underpinnings of this project (Braun and Clarke, 2006).

The data collection method was an online survey due to design and pragmatic reasons. According to critics of surveys, a failure to probe responses with follow-up questions ultimately leads to diluted, perfunctory answers (Braun and Clarke, 2021). However, qualitative surveys can also provide detailed and nuanced accounts of participants, which interests qualitative researchers with carefully thought-out designs and carefully piloted questions (Braun and Clarke, 2021). This study was initiated during the first wave of the pandemic when strict social distancing measures were in place, and the impact of the virus was not fully understood. In addition, we were conscious of the substantially increased demands on adolescents using our service at the time in terms of school/college and virtual therapy. Due to these practical considerations and limitations, we chose to use a qualitative survey as an easily accessible, relatively straightforward method of online participation instead of in-depth interviews or focus groups.

The first author created the survey using the online Qualtrics™ platform. The survey questions were informed by previous research (Zalewski et al., 2021) and input from the co-authors and a service user expert-by-experience (EBE), who was a graduate of the National and Specialist CAMHS DBT-A programme (see Appendix A in Supplementary material). The EBE then piloted the survey and following their feedback, final amendments were made.

In addition to some general demographics and preliminary questions, the key survey questions asked participants to provide free-text, written answers about various aspects of their experiences of virtual DBT-A, including individual and group therapy sessions, their relationships with therapists, the use of DBT-A procedures and resources, and the contrasting experience of virtual versus in-person therapy. Participants were encouraged to provide detailed responses.

Service setting

The National and Specialist CAMHS DBT Service is a Tier 4 out-patient programme offering DBT-A to adolescents up to 18 years and their parents/carers. Adolescents with emotion dysregulation and related difficulties are referred to the service from Tier 3 community CAMHS teams and Tier 4 adolescent psychiatric units across the country. The criteria for entry into the service include problems with emotion dysregulation, self-harm and/or suicidal behaviours, and at least three other areas of difficulties identified via the BPD module of the Structured Clinical Interview for DSM-IV (SCID-II; First et al., 1997). A diagnosis of BPD is not automatically given to adolescents who meet those inclusion criteria, nor is it required for them to access DBT-A. It may be explored collaboratively with adolescents over time if they feel this would be helpful to them. Pre-treatment is offered to adolescents who meet the inclusion criteria to help establish a commitment to DBT-A and therapy goals. At the end of pre-treatment, adolescents and their therapists decide whether to commence the DBT-A programme. Adolescents are then offered an initial 8-month programme with the option to extend to 12 months if they are stabilising and progressing. Adolescents generally only exit treatment early if they miss four consecutive group or individual sessions, known in DBT-A as the four-session rule (Miller et al., 2007).

The treatment consists of weekly individual DBT sessions, a weekly adolescent DBT-A skills group (for the first 6 months, including Distress Tolerance, Emotion Regulation, Interpersonal Effectiveness, Mindfulness, and Walking the Middle Path skills); telephone coaching for adolescents and parents/carers (Monday to Friday, 9 am to 5 pm, to support the generalisation of skills); a weekly parent and carer skills group (for the first 6 months); and parent/carer support and family sessions as required. Adherence to the DBT-A model is supported by weekly team consultation meetings, all clinicians in the service having completed at minimum a foundational level training in DBT with an accredited provider, and quarterly team consultation with an accredited DBT supervisor who is a recognised expert in the field.

Participants and recruitment

Participants were recruited from the National and Specialist CAMHS DBT-A Service at the Maudsley Hospital in the UK. All participants were currently enrolled in the DBT-A programme. Participants were 14–18 years old (mean age of 16 years). Among 30 eligible adolescents, 15 opted in after providing consent (for over-16s) and assent plus parental consent (for under-16s) (Table 1). Participants were included in the study if they were engaging in virtual DBT-A either through group therapy and/or individual therapy.

Table 1. Participant demographics

Procedure

The first author distributed the survey link via email to eligible, consenting adolescents within the service. The data from the surveys were stored electronically and securely by the first author.

Analysis

The survey data were manually coded and analysed by the first and second authors in accordance with the six-phases procedure of TA (Braun and Clarke, 2013):

1. (1) Becoming familiar with and making sense of the data by reading and re-reading each interview transcript, noting down ideas and patterns, and writing a summary.

2. (2) Generating initial codes by highlighting and labelling each section of the transcript that captures something meaningful and interesting about the data, collating and matching the relevant data to every code.

3. (3) Searching for themes by examining the coded data to generate potential themes and grouping them.

4. (4) Reviewing themes by checking them against data extracts to ensure that they are coherent, in a logical order and support each other.

5. (5) Defining and naming the themes to help convey their subject matter; and

6. (6) Writing up the report by refining and finalising the analysis and narrating a rich story of the data with generated themes and a sample of data extracts to demonstrate the findings.

The first author used a reflexive process throughout data collection and analysis, which included keeping a journal, reflecting on personal biases, and seeking consultation when needed. The themes developed were shared and discussed with the broader DBT-A service, stimulating ideas about how the programme could best support the practice of delivering adherent DBT-A virtually.

Results

Three over-arching themes (including eight key subthemes) described adolescents’ experience of virtual DBT-A (Table 2). Each theme is illustrated with survey response extracts as described below. The authors aimed to present representative excerpts from all participants; those selected were the most expressive and closely represented the themes.

Table 2. The three over-arching themes (and eight key subthemes) describing adolescents’ experience of virtual DBT-A

Over-arching theme 1 – Sense of loss

Many participants had been in the routine of attending in-person individual and group sessions before the pandemic. They described experiences of loss associated with attending DBT-A virtually, particularly in relationships with therapists and other service users, and decreased opportunities to practice being skilful.

Subtheme 1 – Loss of connection with group and therapists

Participants shared various challenges in feeling connected with their therapist via virtual DBT-A. For some, virtual therapy led to reduced motivation to engage in treatment than face-to-face sessions. P9 remarked on the change in their commitment and energy: ‘I feel like my therapist saw a side of me that was less willing to work on things over a video call, whereas when we had face-to-face sessions, I was always more enthusiastic and ready to talk and work’. P6 described the issues this created around reduced attendance, ‘I prefer to do face-to-face sessions as I’m more likely to do them. So I feel like I’ve missed a couple [of] sessions’. Another area of loss experienced by participants was their trust in their therapists. P10 remarked, ‘It was ok although harder to connect and trust the therapist’. Others discussed losses of confidence and feeling comfortable through having virtual therapy. P1 remarked, ‘I prefer person to person not over phone don’t have much confidence on the phone’. P6 highlighted increased difficulties talking about emotional material, ‘I think having to talk about more emotional things on the phone is more uncomfortable’. Participants also reported on the impact of teletherapy on their relationships with other group members during the virtual DBT-A skills training group. For some, there was something overwhelming about the experience of seeing multiple faces simultaneously on a video call. P1 remarked, ‘Didn’t like group coz [sic] there were too many people on my screen at once’. P4 highlighted the general ‘lost sense of connection to people’ as well as the specific challenges to developing ordinary social connections with other group members, ‘Haven’t got to know other people… could not talk to others normally (e.g. before and after group) so the social element is missing’.

Subtheme 2 – Loss of skills-building opportunities

Participants described how the shift to virtual therapy hindered skills development. Technical difficulties and feeling self-conscious about how they appeared on the screen caused emotional distress in many who attended DBT-A virtually, negatively impacting their ability to learn DBT-A skills. P4 highlighted how multiple challenges and distractions associated with video calls had negatively affected the effectiveness of sessions and left them frustrated: ‘My wifi at times played up so some sessions were cancelled/postponed due to the fact I couldn’t see or hear anything my therapist was saying. It is not the same as having virtual contact to physical contact. I get a lot more easily distracted on a video call because of everything around me and them not being in front of me physically. When I was really struggling, it did not help having video calls as it just seemed pointless to me’. P5 discussed the distraction of seeing themselves on screen and the time that can be lost resolving technical issues, ‘Sometimes the call cut out, and we had to use mobile, which took time away from the session. Also, being able to see your own face was sometimes stressful as I was constantly aware of what I looked like, which was a bit distracting to the session’. For others, there was something inherently distracting about trying to do virtual therapy in their bedroom. P1 remarked, ‘I feel like I lost skills for to [sic] it, did not find it helpful because I found it hard to focus in my room’. P6 described the difficulty of using DBT resources in a video session without the usual visual aids that would be used in the consulting room, ‘You can’t really see the sheets, for example the chain analysis, and because I’m more of a visual learner it was more difficult to see the patterns of my moods’. Participants also reported the loss of opportunity to build confidence and achievement by attending face-to-face sessions on days they felt distressed. P9 remarked, ‘I also felt less of a sense of accomplishment after having a session as with face-to-face sessions, I could feel good about the fact I was able to get out of bed and out of the house and see and talk to people and travel, which helped me a lot during very depressive episodes, but I couldn’t feel any of that with a video call’. P2 highlighted barriers to feeling and regulating emotions due to having sessions in the home, ‘We associate the DBT groups and one-to-one sessions as the place where we can expel emotion, home often has entirely different connotations’.

Over-arching theme 2 – Feeling uncontained

Many participants described difficulties associated with virtual DBT-A, which diminished the usual boundaries and structures of therapy sessions, leaving them stressed and uncomfortable. Several factors, including privacy, safe therapy space and endings, contributed to this feeling of being uncontained before, during and after sessions.

Subtheme 3 – Limited privacy

Most participants had to conduct their virtual DBT-A sessions in their homes. Participants expressed privacy concerns when engaging in virtual DBT-A sessions. P2 remarked, ‘It’s important to take into account that not everyone may have complete comfort or privacy at home’. P7 shared their worries that parents may hear their conversations with their therapist and how this restricted their ability to communicate freely during their sessions, ‘Felt uncomfortable being so close to parents for the call… I have really struggled with fear of people hearing, and I don’t think it’s good that I have to have therapy in my bedroom, so there’s no divide between my life and my therapy’.

Subtheme 4 – Lack of safe therapy space

For many participants, the physical separation between home and therapy space was crucial; in attempting virtual sessions at home, they felt a lack of safety. P2 described the difficulties of trying to do therapy in a home environment that does not feel safe at times compared with the structures of a therapy clinic, ‘Being able to go to a place separate from home becomes almost a safe place. It’s somewhere you know that people are dedicated to your safety and comfort, which is not always the case at home’. P9 illustrated how an in-person consulting room is superior to a virtual session at home for creating conditions conducive to good engagement, ‘Because I find doing therapy at home difficult when I’m not in the right space of mind and being in a room for therapy in an actual clinic helps me to be in that headspace and makes me more willing to talk about my issues and think of solutions’. For others, losing a segue between therapy and home proved difficult. P7 remarked, ‘I don’t like having to do therapy in my house bc [sic] it feels more stressful as I have to go straight from therapy to then seeing my family rather than having that gap between therapy in the home’. Safety in the skills group also appeared to be affected by virtual service delivery. P4 reported, ‘It feels so abnormal for everyone and it shows, harder to notice when someone is struggling over zoom rather than face to face. People sometimes left when a therapist was screen sharing, and it would not be immediately noticed’.

Subtheme 5 – Difficult endings

Participants shared their emotional challenges in ending virtual therapy sessions and their therapeutic relationship. They emphasised that the commute home helped them regulate their emotions following therapy sessions. However, at home, they had problems managing emotions after virtual sessions. P9 remarked, ‘Trying to manage emotions and urges was a little more difficult as although my therapist was able to help me stay in control and grounded during the session, once the session was over and we had hung up, it was more difficult to keep using those skills when I had no journey home to relax or distract me’. P2 described the distress of having the therapy process end via teletherapy, ‘Ending therapy online is abrupt – which is obviously difficult for people that have trouble processing emotions. Goodbyes tend to be difficult, and I think it’s the worst part of doing DBT online. As people that are so self-destructive it’s rare to have a safe place last so long, and to have it finished in such a mundane way is genuinely painful. It was also unsatisfying finishing the group online’. This subtheme raises the question of whether aspects of virtual DBT may at times exacerbate some of the vulnerabilities that people with emotion dysregulation and related problems might already have, e.g. chronic feelings of abandonment and rejection.

Over-arching theme 3 – Benefits of virtual DBT-A

Participants shared a variety of advantages of attending DBT-A virtually. They described how home-based sessions could be preferable at times, how the virtual medium could improve accessibility, and how online delivery facilitated group interaction.

Subtheme 6 – Home comforts

Participants described how virtual sessions from home could be more comfortable and less distressing. They emphasised the importance of having easy access to self-soothing objects when feeling overwhelmed during sessions. P2 described feeling less self-conscious about using self-soothing items while attending virtual therapy sessions: ‘There was some comfort in being in my own home that meant I had access to stress relievers such as a blanket, paper, and pens, toys etc. that I could easily use to distract me if things became overwhelming, without having the anxiety alongside of what others may think of me using these stress relievers as it was easy to keep it off-camera or even have it on camera without it being brought up or noticed’. The virtual medium felt more relaxed for others, leading to increased engagement. P3 remarked, ‘It was just like meeting in person, but it was better because I didn’t have to leave my room ahah [sic]. It was helpful. I felt more able to talk freely over video calls instead of in-person even though I knew my therapist wouldn’t have judged me either way. It made me more relaxed as I was doing therapy from the comfort of my own home’. In considering this subtheme alongside the previous subtheme, it was interesting to note that the experience of being in the home environment while doing virtual therapy could bring significant pros and cons for participants.

Subtheme 7 – Convenience and accessibility

Participants described other benefits in terms of ease of engagement and attendance. Not travelling to appointments brought advantages to many living far away and even relatively close people. P5 remarked, ‘It’s not a big issue having to meet online, I think it can be easier sometimes (no travel)’. The benefit of attending virtual therapy helped in avoiding longer journeys on public transport. P6 stated, ‘I liked having quick access to speaking to [my therapist] as I would usually have to take trains all the way up to London’. Family sessions are a crucial component of DBT-A. These appointments can be challenging to arrange due to family members’ other commitments and travel requirements. P9 highlighted how virtual family sessions can help address some of these challenges, ‘Group sessions with my family were definitely easier to organise as we weren’t all having to travel from different places because of work or school’. Participants also noted that having virtual sessions could make therapy more accessible on days when the journey would otherwise feel overwhelming. P9 remarked, ‘Video calling was helpful when I would’ve found it difficult or distressing to leave the house’. The online format was conducive to some of the standard DBT-A procedures and resources for many. P4 commented that ‘Worksheets could be virtually sent while in session’. P7 highlighted a more straightforward process for submitting diary cards, ‘Diary cards were easier because I just had to send them in rather than bring them’. This subtheme highlighted how a virtual DBT offering might increase access for adolescents and families, particularly in our programme being a national service. More effective use of DBT resources via the online medium might be more engaging and helpful in reducing TIBs for some.

Subtheme 8 – Easier to participate with others

While some participants reported challenges connecting with therapists and other service users during virtual sessions, they also experienced the virtual platform as manageable and effective, sometimes even preferring it to face-to-face meetings in some respects. P7 remarked, ‘Feels slightly less confrontational’. P2 noted that once they became familiar with the technology, it made virtual meetings more engaging, ‘Sometimes it also felt easier to communicate when asked questions due to our desensitisation of technology. It occasionally became easier to talk to a camera or even talk without looking directly at anyone, somehow having a more effective session as more speaking up happened, particularly for the group.’ P6 commented on video meetings helping their communication when multiple attendees are involved, ‘You can see each other, which I feel you can express your emotions a bit better, makes the conversation flow easier’. It appeared that the virtual medium might also help with more skills group members participating actively in group exercises. P8 remarked, ‘… we were all able to get involved in activities we did’.

Discussion

This study explored the experiences of adolescents participating in a DBT-A programme being delivered virtually, and highlighted some of the benefits and challenges they described.

Regarding ‘sense of loss,’ participants found that switching to virtual therapy presented changes and challenges in relationships, particularly with therapists (although also with peers in the programme), and reduced opportunities to develop and practise skills. They reported various problems linked to virtual DBT about commitment, trust, difficulties talking about and working on emotions, and less group cohesion. Contrary to Stubbing et al.’s (2013) findings that ICBT using video might be as acceptable as face-to-face therapy without compromising the therapeutic alliance, the current study indicated that adolescents with emotion dysregulation and related problems might have additional difficulties establishing rapport and connection with their therapists through virtual DBT. This finding is significant given that the therapeutic relationship is a key mechanism of change associated with improved treatment outcomes for people diagnosed with BPD (Barnicot et al., 2012). The potentially negative impact on participation and interpersonal connection when skills groups are conducted online resonates with Pardo et al.’s (2020) finding that the active presence of skills group therapists is key to promoting motivation, engagement and group cohesion, including how they use the room and the resources within it. Developing solutions that promote trust and connection between therapists and adolescents in virtual DBT-A may be crucial for dealing with problems arising from online working.

Behavioural rehearsal in DBT sessions is considered an essential therapeutic process in addressing the core deficits relating to emotion dysregulation and strengthening skills (Heard and Swales, 2016). The loss of skills-building opportunities reported by participants highlights potential challenges in adhering to the skills acquisition function of DBT-A.

Regarding ‘feeling uncontained,’ our adolescent participants highlighted a range of challenges of virtual DBT-A as the usual structure and boundaries of face-to-face therapy were altered, mirroring findings of virtual DBT in Zalewski et al.’s study focusing on clinicians’ experiences (Zalewski et al., 2021). Difficulties navigating online endings (of both sessions and the therapy itself), restricted privacy, and a lack of safe therapeutic space, were reported as contributing factors to feeling uncontained.

The concept of therapist ‘containment’ and the related concept of ‘holding’ have been drawn on in a variety of psychotherapeutic traditions (Lanman,1998), speaking to the integral role of the therapists’ active behaviour in creating a safe space in which clients can explore, experience, tolerate and respond more effectively to problematic emotions, including navigating the recurring presence and absence of the therapist through the therapy process. ‘Non-specific’ treatment factors, such as therapeutic alliance and therapist adherence to the specific treatment modality, have long been recognised as contributing significantly to treatment outcome and may account for more of the variance in the outcome than specific treatment approach (Chatoor and Krupnick, 2001). Participants in the current study reported that without the physical presence of their therapist in a dedicated therapy room, along with the lack of privacy at home, they felt more uncomfortable, less connected to their therapist, and less able to engage in the therapy process. Additionally, the abrupt and remote nature of session and therapy endings in online DBT may be challenging for adolescents, especially in light of the struggles with relationships (including feelings of abandonment and rejection) that many experience (Hawton et al., 2012).

Regarding the ‘benefits of virtual DBT-A,’ participants discussed ways in which virtual therapy could offer opportunities to access soothing/comforting aspects of the home environment, easier access to their sessions (not having to travel), increased convenience when organising family sessions, and using electronic resources for DBT session activities. The issue of increased accessibility seems particularly important to us given that our national service previously often required participants to travel at least 60 minutes, and sometimes even more, to attend appointments. Reductions in the need for long journeys may significantly increase service users’ access and engagement in treatment.

Strengths and limitations

This paper is the first to explore adolescent service users’ perspectives of virtual DBT-A and highlights some of the challenges and benefits they experienced. Previous research in this emerging area has described clinicians’ difficulties in delivering adherent DBT for adults online (Zalewski et al., 2021). The current study is the first to explore the relational and emotional challenges faced by adolescents in virtual DBT-A and their experience of the benefits of engaging in this therapy online. As virtual therapy will probably become an increasingly common part of mental healthcare delivery in the future, this study offers some recommendations to help clinicians anticipate and address the challenges adolescents may experience in online DBT-A while maintaining fidelity to the evidence-based model (see ‘Key practice points’ below). Based on these findings, our service has already applied these approaches, and clinicians have anecdotally indicated that these have generally been useful to follow when conducting DBT virtually with adolescents.

This study also has several limitations. Despite our rationale and efforts to make participation more accessible and manageable, some individuals’ responses remained limited within the online survey format. The multiple online demands on adolescents at the time (i.e. virtual school sessions) could have led to ‘screen fatigue’, as excessive screen time has been linked to poor psychological outcomes in adolescents (Oswald et al., 2020). In a time-limited academic project, the number of participants was less than would normally be recommended for this data collection method (Clarke and Braun, 2013). Although data saturation (Glaser and Strauss, 2017) may not have been reached, the number of participants and the richness of data may not always be predicted in advance (Malterud et al., 2016; Morse, 2000). In qualitative research, the concept of saturation as the ‘gold standard’ of data quality has been critiqued (Braun and Clarke, 2022; Malterud et al., 2016). While the current study would have benefited from a larger number of participants, we contend that it nevertheless had some positive attributes in terms of ‘information power’ (Malterud et al., 2016). For example, the aim was specific (exploring the experiences of adolescents with emotion dysregulation-related problems of DBT-A delivered virtually). The participants held characteristics that were highly specific to that aim. A high response rate was achieved within the sample invited to participate (43% of eligible adolescents within the service at the time took part), and the expert-by-experience was consulted in designing the survey questions. We note that there may have been other important experiences of adolescents participating in virtual DBT-A that were not captured in the data and are therefore not represented by the themes generated. As with all qualitative research, interpretation of participants’ responses can be influenced by researcher assumptions and expectations, which may lead to bias within the analysis.

Clinical implications, future directions, and conclusions

The present study is an important reminder of establishing rapport and relationship-building with emotionally dysregulated adolescents in virtual therapy, especially if online treatment combined with specific client vulnerabilities potentially renders that more difficult. Adolescent services delivering virtual DBT-A could also support and encourage service users to use larger screen devices wherever possible, promoting more effective monitoring and responses to emotional distress, therapeutic connection, and use of in-session resources and activities.

Safety and privacy seem especially important for online DBT-A conducted at home. Services could support service users in actively identifying the most suitable locations for virtual sessions. Various places, such as schools, health or local authority facilities, could be considered, or possibly less busy homes. As managing the urge to engage in high-risk behaviours is often a primary focus in DBT-A, thorough risk assessment and safety planning are paramount. Similar to Szlyk et al.’s (2020) recommendations regarding telehealth services, clinicians could consider developing collaborative ‘cope ahead plans’ for managing difficult emotions and risk urges that may arise during or after virtual sessions and service procedures for adolescents who leave online therapy sessions unexpectedly.

Due consideration should be given to the ending of therapy sessions and the ending of treatment as a whole when conducting DBT-A virtually, as it may complicate and intensify the emotional challenges associated with these processes. Proactive, open conversations about the difficulties of endings and collaboratively planning for skills used at such times could all be considered. Clinicians and adolescents can also explore ways to make final sessions more distinct and meaningful when therapy has been conducted online (Gilbert and Leahy, 2009).

In virtual DBT skills groups, facilitators could consider ways to increase positive connections among group participants and their engagement with group activities. Similar to recommendations made by Zalewski et al.’s study (2021) about clinicians working with adults, DBT-A group facilitators could also use the video conference platform’s functionality creatively, for example, by making the virtual room available early and during breaks to allow time for informal conversation, employing the chat function for shaping up participation, use of break-out rooms for smaller group activities, polling features for quizzes, etc.

Future research could build on the findings of the current study by exploring other key stakeholders’ experiences in delivering virtual DBT-A, including parents, carers and DBT-A therapists. To determine whether the model’s reported effectiveness extends to virtual adaptations, quantitative evaluations of outcomes for adolescents completing virtual DBT-A are essential.

While DBT-A in traditional face-to-face settings is developing an encouraging evidence base (Bhaji et al., 2021), it is important to consider how best to adapt service provision to maintain treatment adherence and effectiveness when working virtually (Hadjistavropoulos et al., 2017). The continuing endeavour to demonstrate the acceptability and efficacy of virtual DBT-A requires more developed, evidence-based clinical recommendations informed by clinicians, researchers and service users.