Accessibility at What Price? Therapists’ Experiences of Remote Psychotherapy with Children and Adolescents During the COVID-19 Pandemic

ABSTRACT Psychotherapy has traditionally been delivered in person, but recent technological advances have made it possible to conduct remote treatments. There is currently strong evidence for the efficacy of guided self-help with online support from a therapist, but less is known about video-mediated psychotherapy. The COVID-19 pandemic has however forced many therapists to provide remote treatments. This transition might be especially trying for therapists of children and adolescents, but their experiences are underexplored. This study aimed to investigate their perceptions of video-mediated psychotherapy. Semi-structured interviews were conducted with 16 therapists and analyzed using thematic analysis. The therapists described how they struggled with technical and ethical issues and tried to overcome the loss of their usual therapeutic tools. They were concerned that the online format led to less effective treatments or could have negative effects, even if it might increase care availability. Generally, they felt frustrated, inadequate, and stressed, and experienced less job satisfaction. The therapists concluded that video-mediated sessions might be a good alternative for children and adolescents – provided the therapists themselves could determine for whom and when to offer video sessions. Implications of their experiences are discussed, including how psychotherapy training might have to incorporate issues related to remote psychotherapy.


Introduction
Communication technology was used in remote psychotherapy long before the outbreak of the COVID-19 pandemic. According to Markowitz et al. (2021), we can expect that online counseling and remote psychotherapy will eventually be the most common form of psychotherapeutic treatment. Globalisation and the availability of communication technology have a great influence on human thinking and lifestyle. Psychotherapists thus need to adapt themselves and their treatment methods accordingly (Scharff, 2012). Many therapists around the world nowadays offer remote psychotherapy as an option for their patients (McMullin et al., 2020). In Sweden, internet-based treatments, such as guided self-help, have been used in primary care settings for a long time. Accessibility for different patient groups increases when the patient does not have to travel to the therapist's office. Patients living far away and those with special needs or diagnoses may have access to specialized care that may not otherwise be available locally. The cost and inconvenience are reduced for patients when the need for travel time or time off from work is minimized. The expenses for therapists and care facilities are reduced when the need for office space decreases. The relative anonymity of remote psychotherapy can reduce the individual's resistance to seeking help and fear of stigma. This may imply that groups that otherwise refrain from seeking care will look for psychotherapy to a greater extent. To sum up, remote psychotherapy can contribute to increased accessibility and costeffectiveness in care (c.f., Backhaus et al., 2012). Likewise, in the field of child and adolescent psychotherapy (CAT), the use of remote communication technology has been put forward as a potential and effective alternative or complement to in-person sessions (Slone et al., 2012), and the public acceptance and demand for remote psychotherapy has increased over the past two decades (Wade et al., 2020).
However, conceptual confusion makes it difficult to know what type of format we have evidence for. Different terms are used synonymously but in practice may include different variants of communication technology in the performance of psychotherapy. Synchronous remote communication can be provided by telephone, online audio-or video-link, chat, or e-mail. Different studies refer to different terms, such as telepsychotherapy, internet psychotherapy, online psychotherapy, videoconferencing, iCBT, IPBT, et cetera. Furthermore, remote psychotherapy can be designed as guided selfhelp with minimal and asynchronous communication with the therapist, in most cases via a website supplemented with the use of text, video, and audio files (Fernandez et al., 2021;Smith et al., 2022) Research evidence for guided self-help for children and adolescents is still limited (c.f., Mortimer et al., 2022). However, there are some randomized controlled studies of online treatments, for example, for depression .
Video-mediated psychotherapy is based on synchronous remote communication and thus is more like face-to-face sessions in the therapist's office. Often called "videoconferencing," this form of remote psychotherapy has been found to have effects equivalent to traditional treatment for anxiety and depression, PTSD, OCD, panic disorder, and social phobia (c.f., Backhaus et al., 2012). The few studies of video-mediated child and adolescent psychotherapy show good effects when children and parents participate together in the sessions (Slone et al., 2012).
The extensive restrictions following the World Health Organization's declaration on 11 March 2020 of the COVID-19 outbreak as a pandemic accelerated transitions to remote psychotherapy (c.f., Aafjes-van Doorn et al., 2022). Therapists working with children and adolescents in Swedish public health services were forced to quickly switch to remote sessions, mainly video calls. There is some evidence that the pandemic and lockdown measures had a negative impact on mental health in the general population, and especially among children and young people (c.f., Cielo et al., 2021). Increasing the availability of psychotherapy for these age groups by providing remote alternatives will probably continue to be an important issue all over the world.
Following the outbreak of the COVID-19 pandemic, a growing number of studies focused on the therapists' experiences of forced transition to remote work (Humer et al., 2020;Khan et al., 2022). Generally, the therapists seemed to be hesitant to use audio or video online sessions, concerned with issues of patient security, working alliance, and how efficient they could be when working remotely (Ahlström et al., 2022;Békés & Aafjes-van Doorn, 2020;Békés et al., 2021;McMullin et al., 2020). Video-mediated sessions tend to be especially susceptible to technical problems with audio and video quality, the internet connection, and other technical difficulties, leading to stress, breaks, and disturbances in the therapeutic relationship (Békés & Aafjes-van Doorn, 2020;McMullin et al., 2020). It becomes difficult for the therapist to protect patient integrity when the sessions take place outside of the therapist's office and wherever the patient has chosen to be (Khan et al., 2022;Knott et al., 2020;McMullin et al., 2020). The boundaries between private and public are blurred when therapists are allowed to experience the patient's home environment and the therapists experience the conversations as less serious (Békés & Aafjes-van Doorn, 2020;Isaacs Russell, 2021). The therapists find it more difficult to make assessments of mental health and risks, and to understand and answer the patients' feelings; the probability of misunderstandings increases; and the emotional contact and non-verbal communication are more restricted. All this results in the therapists' fatigue and concentration problems, and feeling inadequate, uncertain and less competent (Aafjes-van Doorn et al., 2022;Ahlström et al., 2022;Békés & Aafjes-van Doorn, 2020;Békés et al., 2021;Isaacs Russell, 2021;Khan et al., 2022;Knott et al., 2020;McMullin et al., 2020;Williams, 2021).
As far as we know, there are only a few studies examining experiences of therapists conducting videomediated child and adolescent treatments (Bate & Malberg, 2020;Bomba et al., 2021;Elford et al., 2000;Pozzi Monzo & Micotti, 2020;Trub, 2021;Udwin et al., 2021;Wade et al., 2020;Widdershoven, 2017). The growing need for psychiatric care for children and young people and the demand for increased care accessibility and digitalization may imply increased demands for video-mediated treatments. Therapists' experiences and feelings in relation to remote work with children and adolescents may therefore become an important work environment issue in the future. The present study aims to fill an essential gap in clinically relevant and research-based knowledge in this field.

Aim
The aim of this study was to examine the experiences of child and adolescent therapists when faced with the transition from face-to-face to video-mediated psychotherapy during the COVID-19 pandemic.

Method
This study is part of a larger research project, Transitions to telepsychotherapy, personality orientation and attachment style: Learning from the COVID-19 pandemic and its effects on provision of psychotherapy, intended to explore which patients benefit the most from psychotherapy administered face-toface and under which circumstances, and for which patients remotely delivered psychotherapy may be most beneficial. In the present investigation, the focus was on the experiences of child and adolescent therapists (working with patients within the age range of 6-17 years old), and who had shifted all or some of their treatments to video-mediated psychotherapy during the COVID-19 pandemic.

Participants
Recruitment was initially carried out by the directors of different clinics for child and adolescent patients in Gothenburg (second largest city in Sweden, approximately 600,000 inhabitants), Malmö (third largest city, approximately 325,000 inhabitants), and Uppsala (fourth largest city, approximately 165,000 inhabitants). Clinic directors disseminated the information about the study to their employees. A total of three therapists were recruited using this strategy. To reach therapists in Stockholm (largest city, approximately 1 million inhabitants), one major private care provider was contacted; however, none of their therapists were interested in participating. Lastly, the first author of this study, working at a child and adolescent psychiatry clinic in Västerås (sixth largest city, approximately 130,000 inhabitants), recruited ten therapists. Three further therapists from Stockholm, Uppsala, and Västerås were recruited using snowball sampling, i.e. they received contact information from other people interested in participating. The recruitment strategy and the choice of heterogenous sample was aimed at capturing different facets of the phenomenon in focus.
A total of 16 therapists were eventually included in this study. They were in the age range of 30-64; 30-39 years (eight therapists), 40-49 years (three therapists), and 50-64 years (five therapists), with a median age of 39.5. All had at least two years of experience working as child and adolescent therapists.
In terms of their professions, nine were clinical psychologists, six were social workers with basic training in psychotherapy, and one was a behavioral scientist. Two of the therapists also had a psychotherapist license. Three identified themselves as men and 13 as women. None of the therapists were asked to specify their therapeutic orientation or their ethnicity. Fourteen therapists were employed in primary care clinics for children and adolescents, one in child and adolescent psychiatry, and one was in private practice. None of the therapists had previous experience of using video-mediated psychotherapy.

Ethics
This study was approved by the Swedish Ethical Review Authority (2020-06819, 2021-03728). All participants received written information about the research project prior to the interview and provided both verbal and written informed consent. All interviews have been anonymized and are presented in such a way that the identities of specific patients, therapists, or clinics are not disclosed.

Interviews
Semi-structured interviews focused on the subjective experiences of working with video-mediated psychotherapy with child and adolescent patients. A preliminary version of the interview guide was based on the first author's own reflections on video therapy and the final version was developed in consultations with the other authors with long experience of interview studies. Following the recommendations by Hill et al. (2005), three pilot interviews were conducted (later included in this study), demonstrating that no further changes to the interview guide were required. The interviews were scheduled using the telephone, e-mail, or personal communication, and were performed in Swedish from August 2021 to January 2022 (during the third and fourth waves of the COVID-19 pandemic in Sweden). Nine interviews were conducted face-to-face, and seven by telephone. All interviews were audio-recorded and lasted between 18 and 44 minutes. All questions in the interview guide were asked and were answered in all interviews.
All interviews began with the first author reiterating information about the present study and checking the participant's informed consent. The therapists were then asked to specify their profession and how many of their treatments were delivered in person or remotely prior to and during the COVID-19 pandemic. Further questions focused on how the therapists perceived their ability to make assessments, deliver psychotherapy, and form a therapeutic alliance via video-mediated sessions. The therapists were also encouraged to think about the pros and cons of administering psychotherapy in this way, if and how they experienced ethical concerns throughout the transition from face-to-face to video-mediated psychotherapy, and how they felt they had been affected by having to perform their work remotely. The interviews ended by allowing the participants to discuss other topics related to the subject matter and to ask questions about the research project. All interviews were transcribed verbatim by the first author. The quotations included in the final version of this article were then translated into English by one of the authors and revised by all coauthors.

Analysis
The interview transcripts were analyzed by applying inductive thematic analysis. According to Braun and Clarke (2006), this qualitative method is particularly suitable to gain a multifaceted understanding of people's subjective experiences of the phenomenon in focus. This procedure involved reading the material in its entirety to understand its content and register preliminary ideas about its meaning. The material was then coded to identify meaningful units, which involved an iterative process going back and forward between codes and the interview transcripts to ensure that nothing was missed, and to refine the codes. All interviews were analysed following the same procedure, regardless of how they were conducted.
A semantic approach was used throughout the analytic process, focusing on explicit or manifest material when looking for recurrent patterns (Braun & Clarke, 2006). The codes and emerging themes were subsequently grouped together using different colors on a mind-map, i.e. a graphic representation of interconnections between the emerging themes. Codes derived from only one interview were removed, as recommended by Hill et al. (2005). The interviews were then reviewed once again to make sure that the themes were representative. Themes that overlapped each other were aggregated, whereas themes that seemed inconsistent were grouped into distinct themes. Using the mind-map, the main themes were divided into different sub-themes, where applicable. Excerpts from the interview which captured the essence of the themes and sub-themes were then selected. The frequencies of each theme and sub-theme were labeled following Hill et al. (2005): "general" indicates themes occurring in all interviews, or all but one, "typical" implies themes present in half or more of the interviews, and "variant" themes are represented in at least two interviews.

Researchers
The first author of this study (AE) conducted the interviews and performed the data analysis as part of her diploma thesis at an advanced psychotherapy training programme, leading to a Swedish psychotherapist license. She is also a social worker, trained in cognitive behavior therapy, with 11 years of experience of working in child and adolescent psychiatry. The second author (DF) is a licensed psychologist and associate professor in psychology with experience in conducting qualitative research in the field of gambling and psychotherapy research. He is trained in psychodynamic psychotherapy. He has a few years of clinical experience in an outpatient psychiatric treatment setting. The third author (AR) is a licensed psychologist and associate professor in psychology, with comprehensive experience of conducting research on guided self-help for common mental disorders. He is trained in cognitive behaviour therapy and works clinically at a private practice with adults, both face-to-face and remotely, and has more than 10 years of clinical experience, mainly in outpatient psychiatric care. The fourth author (AW) is a senior researcher and psychoanalyst, affiliate professor in clinical psychology, licensed psychologist and licensed psychotherapist. He works in private practice with children, adolescents, and adults, and has extensive experience in remote psychoanalytic psychotherapy and supervision.

Results
The thematic analysis yielded 16 sub-themes that are gathered under five overarching themes, representing different aspects of transition to remote treatment with children and adolescents (Table 1). All themes are presented below in order of their frequencies and illustrated by verbatim quotations from the interviews. In order to elucidate the individual differences in the therapists' experiences, each direct quote is followed by the therapist's alias.

Issues with patient safety
All therapists described their worries about patient safety when meeting someone online, such as breach of confidentiality and unreliable assessments leading to incorrect or insufficient treatment. Trying to reduce the risk of negative effects the therapists placed greater responsibility on the parents.

Patient difficulties in protecting privacy
Typically, the therapists experienced that they were unable to ensure confidential exchanges when the patient was at home and others could hear the conversation, enter the room, or otherwise disturb the session. The patients could look for privacy in the bathroom or in the car, or signal that sensitive topics should be avoided. Thus, both felt a loss of control over the therapeutic boundaries.
Here, I know that nobody listens, but I don't know that on video. I don't know if a mother or a little brother is behind the door and can hear everything. So, you don't know if it's private when you talk in that way. [Peter]

Therapist difficulties in making assessments
Typically, the therapists felt unable to make reliable assessments of the patients' problems and tailor the treatment accordingly. Lacking the non-verbal cues in the remote encounters, it was more difficult to understand the patient. They felt that being bodily co-present was vital when meeting depressed patients. Consequently, ten of them decided to meet their new patients in the office prior to remote work, despite the prevailing recommendations on social distancing.
What I could feel sometimes when I talked to someone was, no, this one I just have to bring here, this is not possible, I cannot! When I got someone to assess . . . I felt that I couldn't do this assessment, it was like my own solution. As this cannot be done online. [Gunilla]

Concerns about less effective treatments and negative effects
Typically, the therapists showed doubt about whether the online treatments could be good enough: "Yeah, but there will be some doubts about -is this helpful?" [Jessica]. The therapists could doubt their efficacy and feel stressed about not being able to meet the patients' needs. Video meetings were often experienced as less meaningful, and it could be difficult to work in depth with the patient's problems or cope with the patient's strong emotions. "It is much more difficult to activate and deepen [the therapeutic work], even if I try. But I can also feel that I become much more cautious" [Kathrine]. Therapists could feel reluctant to ask questions about suicide risk as they did not really know how to act on the information in a remote setting. They could worry that remote work on the parent-child relationship could make the child more vulnerable: "When I have witnessed conflicts with a parent involved . . . what happens when I hang up?" [Agnetha]. Therapists were anxious that patients might feel worse after remote sessions when nobody would be there to manage the patients' emotions. Altogether, this contributed to the therapists feeling powerless.

Shift of responsibility to the caregivers
Accordingly, the therapists typically felt that they had to place greater responsibility on the parents for both the treatment and for monitoring the child's mood. Parents were asked to pay greater attention to the child's mood shifts and to act if they perceived risk of a self-harm or of a suicide attempt.
When you felt unsure, you had to do it some other way, like via parents, 'You can ask this,' or 'Can he or she fill in this questionnaire?' trying to ensure that this is not a child who feels too bad and whom you may need to refer . . . so, it's important to ensure that the parent knows how to proceed. [Brigitte] For various reasons, it was considered difficult for children in remote encounters to depict their symptoms. Some lacked adequate verbal capacity, others were unable to share their perspective on their functioning, and some difficulties were related to shy and anxious children having to sit in front of the camera. Accordingly, therapists often had to gather the medical history through their parents during the initial consultations. Especially with younger children, parents had to take greater responsibility for their treatment. Meanwhile, the therapists described sessions with parents as particularly suitable for video calls and said that this generally worked very well.

Restricted therapeutic repertoire
All therapists described the loss of practical aids used in therapy which are usually accessible in their offices. They also experienced difficulties providing age-appropriate psychoeducation, establishing a working alliance, and adjusting communication to the patient's age, maturity, or functional level. Furthermore, they missed the opportunity to pick up non-verbal messages, clarify verbal utterances, and signaling closeness or space to the patient. Losing access to the therapists' body expressions, the patients became more dependent upon their verbal and self-observing abilities. The restricted information channels evoked insecurity among the therapists and feelings of inadequacy and incompetence.

Loss of practical tools in the room
All therapists described their frustration at losing the practical aids commonly used in child therapy, such as toys, whiteboards, and creative materials such as plasticine, animal and human figures, colored crayons, handouts, etcetera. When such aids were lost, it became more difficult to establish a working alliance and provide psychoeducation. The therapists were inventive in their attempts to overcome this limitation, trying to use the therapeutic material in front of the camera or sending handouts in advance to the patient. However, when it was no longer possible to make drawings or play with toys together, the space for a fruitful collaboration was restricted. The sessions contained more passive "talking" and less active "doing," and the therapist became more of a teacher who checked up on how the patient was performing tasks or who was teaching the patient about symptoms. "But this therapeutic . . . where you work in a family-oriented way and with creative material with the children and young people, it fell apart" [Brigitte].

Loss of bodily cues and non-verbal communication
Typically, the therapists experienced difficulties establishing a rapport with the patient when the therapists were unable to use their body to convey verbal messages and emotions. Furthermore, it became more difficult to mirror the patient's feelings, show empathy, motivate the patient, and establish a working alliance. The work became more verbal, and focus shifted toward expressing oneself clearly and making oneself understood, which demanded more energy from the therapists. It became more problematic to grasp and recognize the patient's inner state when only the face was visible on screen or when the patient was not visible at all. Taken together with the patients' difficulties in verbally describing their symptoms, emotions, or events this became an obstacle to understanding the patient, and the therapists reacted with frustration and by feeling incompetent.
It is perhaps a bit more difficult sometimes to read what is happening, especially with those who do not want to be seen on screen. Then you become obtrusive or risk stirring up feelings that you don't even notice. And it would not happen in the office. [Emma]

High demands on the patient
Typically, the therapists listed several patient characteristics that could affect the prospects of online therapy, such as age, maturity, personality, verbal ability, level of functioning, and motivation. Furthermore, the demands on the parents were also higher.

A certain level of functioning and age, low problem complexity, and good verbal ability is required
Typically, the therapists experienced that teenagers could profit more from video-mediated sessions, whereas younger children had difficulty sitting still in front of the screen and had a greater need for a structured environment, practical tools, as well as clear boundaries between play and conversation.
With younger children, the sessions had to take place via the parents. However, apart from chronological age, the patient's level of maturity and the complexity of the patient's problems mattered. The therapists experienced that obsessive-compulsive, phobic, or anxiety symptoms were relatively wellsuited for video-mediated psychotherapy, whereas depression, relationship problems, or comorbid symptoms were more difficult to treat remotely. For some patients, the remote format could be advantageous: "It has been interesting to see, for example, those with social phobia or autistic diagnoses. Then it has actually worked better sometimes." [Katrine]

The patient's own motivation is required
Typically, the therapists mentioned patients' high motivation, and the parents' motivation, as preconditions for working remotely: "Young people who lack self-motivation . . . they sit there but it doesn't feel like they listen so much" [Petra]. Motivated patients could however put up with and tolerate the limitations of such a format.

Personality matters
As a variant, the patients' personality traits were described as influencing the efficacy of the remote treatment. Despite their age or diagnosis, open and outgoing patients could profit more from remote therapy than patients who did not want to be seen on camera, who instead were shy, introverted, and not so social. According to Agnetha: "Even the younger ones, if it is a very playful person, maybe they show up and say "Hey, you should meet my teddy bear," then it goes a little easier of course. Those who invite you and are a bit more open." According to Brigitte: "It has to do with personality, that is, if you see someone who's verbal and outgoing, in terms of personality, that has nothing really to do with being healthy or sick, then it is much easier . . . Thus, I think a lot about personality."

Pros and cons of the new normal
Typically, the therapists experienced both limitations and advantages of video-mediated sessions. They wanted to preserve the possibility of remote treatment, but also to decide for themselves with whom and when they would use it.

The therapist is less emotionally affected
Typically, the therapists experienced that conversations via video did not have the same emotional impact as in-person treatments. Emotions were not transmitted the same way, the therapists felt less emotionally committed to their patients and their families, and the restricted non-verbal feedback made conversations less engaging. According to Frida: "You are not influenced by each other in the same way. Emotions are not as contagious online." It could also be more difficult to remember the sessions and the therapists needed more time to take notes and review the medical records. On the other hand, the emotional distance helped the therapists to be calm and think rationally. Not being so emotionally involved several times a day was also considered less draining.

Decreased job satisfaction and feelings of inadequacy
Typically, the therapists indicated less job satisfaction due to the restricted access to their therapeutic tools, feeling less competent, having concerns about less effective treatments, and being frustrated by technical issues. The therapists yearned for social exchanges, contact, and collaboration with patients, and they lacked opportunities for being creative and playful together. Sitting still in front of a screen and being unable to become active also made it more difficult to stay focused. Video-mediated sessions required more energy, were more tiresome, and the therapists could lose commitment and get less pleasure out of their work. It was also considered stressful and frustrating to feel the patient's dissatisfaction with remote sessions and to worry that the patient was not receiving adequate treatment.
When you felt that this was not going as well as you wished for or knowing everything you could do if the patient was allowed to come here. Then you were affected by it. You did your job the best you could, but these restrictions stopped you from helping patients in some cases, it was difficult to cope with, I think. [Emma]

More efficient conversations
As a variant, the therapists described increased efficiency. This did not concern increased availability, but rather referred to more effective and shorter conversations. Loss of social chatter made you focus more on what you had to work on in treatment, and the therapists could spend more time on other assignments. Taking anamnesis or following up on homework could also be more effective online. According to the therapists, remote sessions are especially suited for less serious diagnoses and more structured, manual-based treatments, such as CBT.

Less serious conversations
Video sessions can be carried out from almost any location the patient chooses, putting the therapeutic room beyond the therapist's control. As a variant, the therapists experienced this as disturbing and irritating, and the sessions as less serious. Some patients and parents were keen on finding a quiet and secluded place and being prepared, whereas others made no such arrangements prior to their sessions. According to Gunilla, it could be a sensitive issue to address when the arrangements were not optimal: "One mom who sat there in the bra, sort of." Hanna experienced that teenagers were often less prepared for the sessions: "I know someone has set the alarm five minutes before, and then they wake up and jump directly from sleep to the session." Both patients and parents could take the opportunity to do something else during the sessions, for example eat a meal or drive a car.

Possibilities and limitations of communication technology
All therapists reflected on both the possibilities and limitations of using video-mediated therapy. The price of increased availability was recurring technical problems, but the restrictions due to the COVID-19 pandemic left no other choice.

Increased availability for some patients
All therapists underlined that the greatest advantage of remote therapy is the possibility for children and parents to participate in treatments independently of their place of residence, education, or work, thus contributing to good service. Especially for parents, it saves a lot of stress and energy not having to take a day off to travel or accompany children to the therapist's office. People with just a cold do not need to cancel their sessions and rescheduling is reduced. According to Helene: "For my own part, I think it has been an advantage to the extent that when you haven't been able to meet in person you have still been able to keep the process going digitally." Seven therapists however underlined that depending on communication technology might lead to reduced accessibility for certain families or groups. People who live in places with an unstable internet connection or none at all, those with limited economic resources, and children of more socially disadvantaged parents might have difficulty attending video-mediated sessions. Another concern was the restriction of space when many people worked from home during the pandemic and there was no safe place for some patients to have their sessions. None of the therapists mentioned having had remote sessions with families who needed an interpreter.

The freedom of choice is important
Typically, the therapists stressed the importance of being able to decide for themselves which patients and in what phase of treatment are suitable for video-mediated sessions. Ideally, they wanted to be able to alternate in-person and online sessions in a flexible way and thus benefit from both therapy formats. They also preferred to meet in person initially to establish a working alliance and make assessments prior to a remote contact. Nine of the therapists unambiguously preferred working with the patients present in their office: "Just because I think it's easier for me to establish a relationship in the room. I find it easier to see and assess and understand and ask about things. And I'm not bothered by technology that lags and jams" [Lena].

Technical problems and work environment
Typically, technical issues were a clear stress factor for the therapists. Problems with the internet connection on the part of both the therapist and the patient led to interruptions or inability to start the session. Patients could be forced to walk around with their computer or mobile to find a stable connection, and it took a lot of time from the sessions to try to reconnect. Problems with sound and image quality led to irritation and insecurity in the therapists when they did not see the patient properly, missed information, had to repeat questions, or did not dare to deepen the conversation. Tiresome interruptions disrupted the dialogue and interfered with the working alliance.
It may mean that you don't go as deeply into different things because it becomes stressful, as it can be broken at any time. Now we need to check this off pretty quickly, you know, or you switch directly to the phone and then you are lost, completely. [Petra] According to six therapists, remote sessions also led to an impaired work environment. The therapists spent much of their time in front of a computer screen, forcing themselves to hear and see the patient, causing increased tension in the body and more fatigue as a result. On the other hand, remote work also had positive health effects. Alternating between standing and sitting led to less pain in the body and better ergonomics. Working with different things during the day also gave a feeling of variation: "Since you are not stuck in the chair, because it is so easy to end up there, and then it should be in this particular way. I am positive about the variation and trying things out!" [Katrine].

Discussion
Taken together, all themes in the therapists' experiences convey their feeling of loss of their ordinary therapeutic tools, loss of emotional contact with their patients, frustration, and restricted competence, resulting from the transition from face-to-face to remote treatment. The therapists and their patients were in a more vulnerable situation, making them susceptible to stress and other types of negative events during therapy. However, some positive aspects also emerged, mainly regarding the accessibility of care. The therapists seemed to have learned a lot from their unexpected lived experiences. Accordingly, they considered remote therapy and video sessions with children and adolescents as a complement to traditional in-person sessions. Generally, the therapists expressed a view that remote therapy might be a good alternative in child and adolescent psychotherapy, as long as the therapists themselves could determine when and for whom to offer video sessions.

Issues with patient privacy and safety
Our study indicates that the therapists, in connection with loss of control over the therapeutic setting, experienced that the patients were at increased risk for harm. Difficulties in safeguarding the patient's integrity were described as a widespread problem, and similar findings have been reported in prior research on transitions to remote treatments with adults (c.f., Ahlström et al., 2022;Békés et al., 2021;Isaacs Russell, 2021). As stressed by Bate and Malberg (2020), the child therapists had to rethink and renegotiate the frame in a concrete way. The responsibility to ensure treatment integrity and safety for the treatment situation seemed to shift from the therapist to the patient (c.f., Trub, 2021). For example, children might have difficulty finding privacy at home and feeling secure in relation to other family members. The child's age and level of maturity can also play a decisive role, as younger children need parental support to be able to participate in remote sessions at home. In addition, the therapists are obliged to assure confidentiality while at the same time they are forced to adhere to other Swedish regulations that are intended to protect the children's right to treatment. When forced to shift to video-mediated psychotherapy, these regulations can conflict with each other, thereby inflicting stress on the therapists. The lack of privacy might also result in children being less prone to talk about their problems. The therapists in our study said that the video format itself made it more difficult to make assessments of mental health, and especially of suicide risk. This dilemma has been previously noted in other studies of remote treatments of adults (Knott et al., 2020) as well as children and adolescents (Pozzi Monzo & Micotti, 2020;Widdershoven, 2017), and it becomes less problematic when remote sessions are conducted in an arranged and controlled space (Elford et al., 2000). Similarly, the therapists' concerns of not being able to understand the patients' feelings and manage affect-loaded insession situations has previously been observed in remote treatments of adults (Khan et al., 2022;Knott et al., 2020) and children and adolescents (Trub, 2021). All this could have adverse consequences for the therapists, in the form of compassion fatigue and feelings of inadequacy.
Another dilemma described by therapists in our study was the greater parent involvement, perceived as distinguishing remote treatments for children and adolescents from remote work with adults (c.f., Békés et al., 2021;Trub, 2021). When the non-verbal cues were lost, the parents could be given a role as a spokesperson for the child and as an interpreter of the child's inner life. This could restrict the child's autonomy and place increased responsibility on the parent, the effects depending on the quality of the parent's relationship with the child, ability to empathize, and the parent's own difficulties.

Therapists' attempts to compensate for the restricted therapeutic repertoire
For the therapists, the transition to video-mediated sessions implied the loss of several tools facilitating communication and interaction with children, such as play and creative materials, as also noted in previous studies (Trub, 2021;Widdershoven, 2017). Furthermore, the therapists lost access to body language and the non-verbal communication when only the face was visible on the screen (c.f., Fernandez et al., 2021;Isaacs Russell, 2021;Khan et al., 2022;Knott et al., 2020;McMullin et al., 2020). This increased the risk of misunderstanding, impaired the quality of the mutual dialogue and the therapeutic relationship, and potentially the treatment outcome. Having access to only two senses (sight and hearing), both parties had restricted access to communication channels, thus altering the transference and countertransference. Like therapists in other studies (c.f., Ahlström et al., 2022;Bomba et al., 2021;Fisher et al., 2021;Isaacs Russell, 2021;Khan et al., 2022;Trub, 2021;Udwin et al., 2021), the child and adolescent therapists tried to find new ways for compensating for the limitations of video-mediated sessions and loss of embodied communication. They described focusing more on verbal communication, facial expressions, and tonality, which reinforced their feeling of being less focused and more stressed and tired.

Patient characteristics determine for whom video-mediated psychotherapy is suitable
The therapists in our study said that the remote format imposed higher demands on the patient's functioning and the patient's ability to take responsibility for their own treatment, as also noted in previous studies (Khan et al., 2022;Pozzi Monzo & Micotti, 2020). One important factor is the child's age (c.f., Trub, 2021). Video-mediated sessions with younger children had to be assisted by a parent, thus influencing the therapeutic exchanges (c.f., Trub, 2021;Widdershoven, 2017). On the other hand, the therapists found that parental support worked very well (Wade et al., 2020). Another important patient characteristic influencing the therapists' judgment of conducting psychotherapy remotely was the complexity of the disorder. The therapists found that patients with low comorbidity were a good fit, e.g., patients exhibiting only obsessive-compulsive disorder or specific phobia, as well as patients with a relatively limited anxiety level. Overall, difficulties in dealing with comorbidity appear to be more challenging for the child and adolescent therapists working remotely than when in a face-to-face setting. This finding is consistent with previous studies of the adult population (Backhaus et al., 2012;Fernandez et al., 2021;Khan et al., 2022;Smith et al., 2022). However, the child and adolescent therapists, in contrast to some adult therapists (Backhaus et al., 2012;Smith et al., 2022), regarded depression as less suitable for remote treatment, due to the perceived risk of misjudgments and increased difficulty in establishing a working alliance with the child. On the other hand, they experienced neuropsychiatric disorders and other communication difficulties, such as language impairment or disability, as less suitable for video-mediated sessions. Additionally, the therapists emphasized the patients' own motivation as crucial for successful video-mediated psychotherapy, as previously found in internet-based treatment for depressed adolescents (Mortimer et al., 2022). In our study, therapists working with younger children might have to put greater emphasis on determining and increasing the patient's motivation.

Technology influences the therapeutic work
According to the therapists in our study, the greatest advantage of video-mediated psychotherapy is that it can be accessible for patients living far from the treatment center and can be performed on the patient's own terms. This finding, consistent with a large body of previous research, is especially relevant in the time of restrictions imposed by the COVID-19 pandemic. However, therapists also brought up the risk of reduced access for certain patient groups, i.e. those not having access to a stable internet connection, lacking the required equipment, or having difficulty using this technology (c.f., Khan et al., 2022;Knott et al., 2020).
On the other hand, the therapists in our study suffered from technical problems to varying degrees, influencing their experience of the patient-therapist relationship. Being aware that the session could be interrupted at any time and being ready to quickly change the communication medium interfered with their attentiveness and caused them additional stress. Still, they regarded remote psychotherapy as a good alternative to face-to-face sessions with children and adolescents. However, they also emphasized the importance of an initial face-to-face meeting into order to make an assessment and to decide what format to use. Rather than seeing remote psychotherapy as a general alternative to traditional face-to-face sessions, the therapists thought it was crucial to determine for themselves when and for whom to offer video-mediated sessions.
Most importantly, the therapists in the present study experienced that the transition to the remote format changed different essential features of their work. As mentioned also by Bomba et al. (2021), they needed to spend more time writing and reading medical records in connection with the sessions to remember the patients and the focus of treatment. They also experienced remote sessions as emotionally draining.

Possibilities and limitations of video-mediated psychotherapy
One unique finding in our study was that the therapists could perceive video-mediated sessions as more effective, since less time was spent on small talk. However, similarly to other studies Békés & Aafjes-van Doorn, 2020;Isaacs Russell, 2021;Trub, 2021;Widdershoven, 2017), the therapists reported impaired job satisfaction due to the loss of control over the therapeutic setting and loss of their usual therapeutic tools. They could doubt if they were competent enough to provide adequate remote treatment, they were stressed by ethical dilemmas and technical difficulties, and they felt left behind and frustrated. Furthermore, they reported concentration difficulties and fatigue, previously described by Williams (2021) as "Zoom fatigue," common in video meetings in general. Patients usually engaged in video-mediated sessions at home, which led to disturbances caused by the home environment and impaired the working alliance. Therapists were embarrassed about getting an unexpected and unwanted glimpse into the private space of patients, their parents, and siblings to a greater degree than with adult patients (Isaacs Russell, 2021;McMullin et al., 2020;Trub, 2021), causing involuntary participation in boundary crossing, especially with younger children with externalizing problems. On the other hand, the therapists found that parental support worked as well remotely as if conducted face-to-face (Wade et al., 2020).

Clinical implications
The findings from this study illustrate some of the complexities surrounding the use of video-mediated psychotherapy with children and adolescents. Our study demonstrates that results from studies of remote psychotherapy with adults cannot be easily generalized to younger patients. Even if there are several similarities between experiences of therapists working with children and adolescents and those working with adult patients, it seems that the ethical and clinical issues are amplified and more trying in remote child and adolescent therapy. Ideally, therapists might have to conduct a proper face-to-face assessment prior to matching the patient with the format and initiating remote treatment. Videomediated therapy brings about specific issues related to the lack of bodily co-presence and of sheltered time and space, the restricted therapeutic repertoire, limited communication channels, and greater dependence on parental support and assistance. As we can expect an increased use of remote treatments with children and adolescents, these issues have to be included in the therapeutic training programmes as well as the therapists' continuing education. Furthermore, the therapists' professional organizations, together with healthcare providers, have to develop guidelines and recommendations for the practice and policies of remote treatment of children and adolescents, following the idea of "what works for whom."

Limitations
This study puts forward many important findings in relation to the use of video-mediated child and adolescent psychotherapy, but several limitations need to be considered when reviewing the results. First, there might exist systematic biases when it comes to the sampling procedure, such as only having recruited therapists who were highly motivated to share their experiences of remote psychotherapy. However, great effort was made to ensure that participants worked in different regions and contexts in Sweden, and the sample exhibits great variation regarding years of experience and professions. In addition, none of the participants had prior experience of working with video-mediated psychotherapy, which should be representative for how most therapists in Sweden worked prior to the COVID-19 pandemic. A sample size of 16 participants is regarded as a medium to large sample in qualitative studies applying thematic analysis (Braun & Clarke, 2006). Second, all interviews as well as the qualitative analysis were carried out entirely by the first author, even if audited by the coauthors, which may have affected the findings. However, the fact that someone with a similar professional background and clinical experience as the participants performed the interviews could also be regarded as a strength and increased the therapists' willingness to share their experiences. Additionally, the entire process, from construction of the interview guide, transcription, initial coding, and thematic analysis to interpretation of the results, was carried out in collaboration with the other authors in regular meetings, ensuring there was a shared understanding and agreement on how to present the empirical data. It should also be noted that the data collection took place 1.5 to 2 years after the restrictions following the pandemic were implemented in Sweden. Thus, the present study does not reflect the therapists' more acute reaction to the transition but rather their more long-term adaptation to this working format.
Still, it is unclear whether the findings from this study are representative or transferable to similar settings or other populations. However, providing information about the rigorous step-by-step qualitative data analysis and presenting representative excerpts from the interviews is intended to increase transparency. As argued by Levitt (2021), the logic of qualitative inquiry is about justifying generalization to the phenomenon in question, not the population.

Future research
Video-mediated psychotherapy is seemingly becoming increasingly available in healthcare, and the COVID-19 pandemic has demonstrated that delivering treatment remotely can sometimes be a viable option for many patients. However, research regarding which children and adolescents, in which circumstances, might benefit the most from this transition is still limited. Accordingly, further studies are needed to elucidate various facets of the phenomenon in focus for the present study, for example, by means of triangulation, using surveys or interviewing patients or stakeholders. In addition, some of our findings warrant deeper investigation, such as what specific tools and strategies therapists use to ensure that treatment resembles what is done face-to-face, even though the patient is not physically present. Similarly, the function of the intermediate space and time between everyday life and therapy sessions may be a topic worth pursuing, that is, how do patients prepare themselves for and engage in psychotherapy when it no longer requires traveling or being at a clinic. Also, hybrid forms of treatment, a mix of face-to-face and remote sessions, could be explored to understand for whom and when the video-mediated psychotherapy might be most efficient. Lastly, given the novelty of remote psychotherapy, the development and use of guidelines and policies for its use in different settings is crucial, both from the perspective of the patient and the therapist.

Conclusions
This study reveals that therapists who had to shift to remote sessions with children and adolescents to some extent have faced similar challenges and difficulties as has been previously described for therapists working with adult patients. The therapists described both positive and negative aspects of this transition. In accordance with prior research, therapists reported that video-mediated psychotherapy could have many advantages for those patients and their families who otherwise might find it difficult to come to the clinic. Meanwhile, ethical dilemmas linked to patient safety and risk assessments were frequently brought up as a downside of the format, and the perception of delivering treatment of poorer quality had a significant impact on the therapists' mood, which led to increased feelings of stress, inadequacy, and loss of skills. Factors such as disturbances from the patient's home environment, technical difficulties, loss of body language and other communicative tools, also affected the therapists negatively and might have contributed to concentration problems, frustration, fatigue, and reduced job satisfaction. All this reflects the therapists' long-term process of digesting the new therapy format that could be summarized as a journey from a necessary evil to an opportunity -for some young people.
One lesson from this journey is that therapists who work with children and adolescents remotely need to address specific issues that are associated with the specific characteristics of each individual patient. Difficulties with protecting the patient's integrity become more pronounced, as does the lack of communicative tools. The greater involvement of parents is common in remote sessions in regard not only to technical assistance but also becoming a spokesperson for the child. Unique to the present study is the therapists' experience of transferring the risk assessment for the children to the parents, who could become responsible for determining the child's mood and suicidal ideation. Such issues should be addressed in an ageappropriate way with the patient, and when applicable, also with the parent. Above all, our study reveals the need for professional support and preparation time for forced transitions to new forms of treatment and new patient groups in times of crisis.