Mindfulness-Based Cognitive Therapy for Cancer Patients Delivered via Internet: Qualitative Study of Patient and Therapist Barriers and Facilitators

Background The number of patients living with cancer is growing, and a substantial number of patients suffer from psychological distress. Mindfulness-based interventions (MBIs) seem effective in alleviating psychological distress. Unfortunately, several cancer patients find it difficult, if not impossible, to attend a group-based course. Internet-based MBIs (eMBIs) such as Internet-based mindfulness-based cognitive therapy (eMBCT) may offer solutions. However, it is yet to be studied what facilitators and barriers cancer patients experience during eMBCT. Objective This study aimed to explore facilitators and barriers of individual asynchronous therapist-assisted eMBCT as experienced by both patients and therapists. Methods Patients with heterogeneous cancer diagnoses suffering from psychological distress were offered eMBCT. This 9-week intervention mirrored the group-based MBCT protocol and included weekly asynchronous written therapist feedback. Patients were granted access to a website that contained the eMBCT protocol and a secured inbox, and they were asked to practice and fill out diaries on which the therapist provided feedback. In total, 31 patients participated in an individual posttreatment interview on experienced facilitators and barriers during eMBCT. Moreover, eight therapists were interviewed. The data were analyzed with qualitative content analysis to identify barriers and facilitators in eMBCT. Results Both patients and therapists mentioned four overarching themes as facilitators and barriers: treatment setting (the individual and Internet-based nature of the treatment), treatment format (how the treatment and its guidance were organized and delivered), role of the therapist, and individual patient characteristics. Conclusions The eMBCT provided flexibility in when, where, and how patients and therapists engage in MBCT. Future studies should assess how different eMBCT designs could further improve barriers that were found.


Introduction
Cancer poses a major psychological challenge for individuals. A meta-analysis of psychiatric disorder in oncological and hematological settings yielded a prevalence of psychiatric disorder of 30% to 40% [1]. In the coming decades, a great increase is expected in the number of people living with cancer [2]. This means that a growing number of cancer patients are in need of effective and accessible psychological treatment.
Mindfulness-based interventions (MBIs) such as mindfulness-based stress reduction (MBSR) [3] and mindfulness-based cognitive therapy (MBCT) [4], the latter more oriented toward those in need of clinical psychological treatment, could be viable intervention options for cancer patients. Mindfulness is defined as follows: "paying attention, on purpose, in the present moment and nonjudgmentally" [5]. Its practice enables participants to recognize habitual, conditioned modes of reacting, and to make a radical shift in how they relate to their thoughts, feelings, and body sensations, as well as to outer circumstances [6], such as when coping with cancer.
Evidence for the effectiveness of MBIs for cancer patients has rapidly expanded. In 2015, an overview including six systematic reviews in heterogeneous cancer patients demonstrated significant small to moderate effects on various psychosocial outcomes in cancer patients [7]. In addition, studies confirmed these effects at longer-term follow-up [8,9]. Moreover, studies demonstrated that the effect of MBIs in breast cancer patients might be mediated by nonreactivity, reduced catastrophizing, and increased self-kindness [10,11]. Notwithstanding the potential of MBIs, several cancer patients encounter practical and psychosocial barriers that hamper access and participation in psychological treatments such as MBIs. These barriers include cancer-related illness, fatigue, limited mobility or disability, limited transportation options, and time constraints [12,13]. Internet-based interventions, such as Internet-based MBIs (eMBIs), may offer solutions to these problems. Mobile MBI apps have already demonstrated their potential [14]. Internet-based interventions are easily accessible, available 24×7, save traveling time, and could be less costly [15]. Evidence of over 100 well-controlled trials suggests that Internet-based treatments can be as effective as group-based psychological treatments for a wide range of psychiatric and somatic conditions [16].
Moreover, a previous review suggests that eMBIs may be helpful in alleviating symptom burden of patients with physical health conditions, particularly when the eMBI is tailored to specific symptoms [17]. A total of 16 studies examining eMBIs for people with chronic physical health conditions were reviewed, of which two specifically targeted cancer patients. A randomized controlled trial (RCT; n=62) investigated the quantitative feasibility of Internet-based MBSR for cancer patients (mindfulness-based cancer recovery [MBCR]) [13]. The Internet-based MBCR (eMBCR) consisted of synchronous videoconferencing sessions. Feasibility targets for recruitment and adherence (5% response rate, 30% eligible, and 85% consented) were achieved, and patients considerably improved on mood disturbance, stress symptoms, spirituality, and acting with awareness in the Web-based group relative to waitlist controls. Results suggested that eMBCR led to improved energy while also inducing relaxation [18]. In addition, an uncontrolled cohort study (n=257) of severely fatigued cancer patients evaluated an Internet-based mindfulness-based cognitive therapy (eMBCT) [12]. In total, 34.6% (89/257) of the patients showed clinically relevant improvement in fatigue severity and 61.8% (159/257) of the patients adhered to treatment. In sum, evidence for eMBIs in cancer is scarce, but the first results seem promising.
However, how to optimally deliver eMBIs remains unknown [17]. It is unclear whether either synchronous (real time, eg, instant messaging or videoconferencing) or asynchronous (delayed, eg, email or message boards) is to be preferred. Patients are supposed to engage in an experiential inquiry-based learning process together with the therapist in eMBIs [6], but it is unknown whether such an experiential inquiry-based learning process is at all possible in an asynchronous format. In addition, it is unclear whether either facilitated (guided) or self-directed eMBIs are to be preferred. It is argued that the therapists' capacity to embody qualities and attitudes of mindfulness in the process of teaching is vital for effective delivery of MBIs [6]. Guidance seems to be a beneficial feature of Internet-based interventions in general [19], and exploratory subgroup analyses of a systematic review indicated higher effect sizes of stress and mindfulness skills for guided than unguided eMBIs [15]. However, a previous review also provided some initial support for unguided eMBIs [20]. In short, the question of which eMBI delivery format is preferable in terms of program adherence, mindfulness skills, and outcome improvement needs further investigation [17].
Previously, a qualitative study has provided important perspectives for examining the user experience in an MBI. In a qualitative study of an eMBI for recurrent depression, patients identified aspects such as flexibility and reduced cost, as well as the need for support in time management [21]. Qualitative information on how patients and their therapists experience eMBCT could identify barriers and facilitators, and inform us whether it is possible to design useful, user-friendly, and effective eMBIs for cancer patients and where to improve delivery mode and design if necessary and possible. Therefore, the aim of this study was to gain understanding of the experienced facilitators and barriers of asynchronically delivered eMBCT in a sample of heterogeneous cancer patients and their therapists.

Study Population and Procedure
The patients of this study took part in a 3-armed trial on the (cost-) effectiveness of MBCT for distressed cancer patients (Clinicaltrials.gov no. NCT02138513) [22]. Patients were randomized to either eMBCT, group-based MBCT, or treatment as usual. The RCT is described in more detail in a protocol paper [22]. Patients for this trial were mainly recruited via online media (26.9%, 66/245), patient associations (17.6%, 43/245), and participating mental health care centers (16.7%, 41/245).
In total, 245 cancer patients with any tumor type and any stage of disease scoring 11 or higher on the Hospital Anxiety and Depression Scale were randomized. The local ethics committee approved this study (CMO Arnhem Nijmegen 2013/542).

Qualitative Assessments: Semistructured Interviews (Patients) and Focus Group (Therapists)
Both patients randomized to eMBCT and their therapists were invited by the researcher to talk about the following questions: • How did you experience the eMBCT?
• What facilitated and what impeded your participation in eMBCT?
• How did you experience the relationship with the therapist or patient?
• How would you improve the eMBCT?
The abovementioned questions were followed by specific probes . Questions were asked in an open nondirective manner, allowing participants to speak freely about their experiences.
Patients were interviewed via telephone or in person within 3 months after eMBCT treatment completion or dropout. Patients were purposefully sampled to gather an even distribution of completers versus noncompleters and breast cancer versus other tumor types. Patient interviews were conducted by FC and EJ. FC is a PhD student with an MSc degree in behavioral science with no prior experience in qualitative research. He was the trial coordinator for the larger RCT [22], and had conducted baseline and posttreatment research interviews conducting the posttreatment interviews. EJ is a psychologist and mindfulness teacher with an MSc degree in psychology with prior experience with qualitative research.
Therapists were invited for a focus group interview during the last plenary therapist supervision session approximately 3 months after completing the last MBCT. Both the patient interviews and therapist focus group started by explaining confidentiality and the explorative nature of the interview. AS and ML conducted the focus group interviews. AS is a professor of psychiatry in the role of principle investigator of the larger RCT [22], with experience in several qualitative research projects in MBIs. ML is a senior researcher and clinical psychologist in the role of principle investigator of the larger RCT [22], with experience in several qualitative research projects on MBIs.

Data Analysis
We used conventional qualitative content analysis to analyze the data in which coding categories are derived directly from the text data [23]. We used ATLAS.ti version 7.1 software (Scientific Software Development GmbH, Berlin, Germany). Analysis started as soon as the first interview was conducted and continued with each additional interview. Interviews were transcribed verbatim, and each transcript was coded by 2 independent researchers (FC and EJ) to identify facilitators and barriers. Coding was performed as closely related to the patient's words as possible to minimize subjectivity. After 5 interviews, FC and EJ compared codes with each other, and a common coding scheme was developed. Remaining transcripts were coded using this common coding scheme and earlier transcripts were recoded. New codes were added when data were encountered that did not fit in the existing coding scheme. After 12 interviews, a larger group of researchers (FC, EJ, MS, ML, and AS) discussed all data within the coding scheme. Some codes were combined during this process, whereas others were split in subcategories. After 31 interviews, no new codes were added and it was concluded that saturation had been reached. All codes referring to the same phenomenon were grouped in a hierarchical structure in subcategories, and subcategories in themes by FC and EJ. The group of researchers (FC, EJ, MS, ML, and AS) subsequently discussed this classification until reaching consensus.

Intervention and Therapists
The eMBCT was based on the MBCT protocol for recurrent depression published by Segal et al [4]. The content was adapted to cancer patients by particularly tailoring the psychoeducation (eg, managing cancer-related fatigue, pain, fear of cancer recurrence, and effects of cancer on partner communication) and movement exercises to their needs [24].
The eMBCT was mainly text-based and included asynchronous interaction with a therapist similar to the study of Bruggeman-Everts et al [12]. Patients were granted access to a website divided into a workspace containing 9 sessions (8 weeks + 1 full-day silent retreat) and a secured inbox. The therapist initiated the eMBCT by sending a welcome message to the patient. When patients logged-in, they were presented with the overview of all information and assignments due for that week ( Figure 1). Each session contained an introductory text, and daily formal (eg, sitting meditation) and informal exercises (eg, awareness of everyday activities) with guided audiotaped files and accompanying diaries. Sessions also contained other home practice such as the pleasant or unpleasant events' diary. The eMBCT was performed individually. To demonstrate the rationale and possible obstacles of each exercise, patients were provided with experiences of other (fictional) patients ( Figure  2). Patients were asked to practice and complete the diaries on a daily basis ( Figure 3). The therapist provided written feedback on their progress via the secured inbox on a prearranged day of the week ( Figure 4). Next week's session only became available after completing the previous session. Patients always had access to their therapist via the inbox. Patients and therapists were notified via their regular email when they received a message via the secured inbox.
We defined adherence as having attended ≥4 sessions. Therapists without prior eMBCT experience were provided with guidelines and were supervised by more experienced eMBCT therapists. See Table 1 for therapists' characteristics. All therapists fulfilled the advanced criteria of the Association of Mindfulness Based Teachers in the Netherlands and Flanders, which are in concordance with the UK Mindfulness-Based Teacher Trainer Network Good Practice Guidelines for teaching mindfulness-based courses [25].

Sample
Out of the 125 patients randomized to eMBCT, 45 were invited for a posttreatment interview. In total, 12 patients declined and 2 recordings failed. As a result, 31 interviews were used in the qualitative analysis. Interviews lasted from 5 to 25 min. Out of the patients interviewed, 14 had participated in 4 or more sessions of eMBCT, 10 had attended less than 4 sessions of eMBCT, and 7 had not started at all. See Table 2 for patient characteristics.
A total of 11 out of 12 eMBCT therapists were invited for a focus group interview after completion of all eMBCTs. Out of these 12 therapists, 7 therapists agreed to participate and 1 therapist agreed to provide an individual interview with FC for scheduling reasons. Therapists declined either because of having provided too few individual online treatments to share experiences (n=1) or because of scheduling reasons (n=2). The focus group interview lasted for 90 min. The single individual therapist interview lasted for 25 min. The final sample of therapists included both therapists who had experience with online mindfulness before this project (n=4) and therapists who had no prior experience with online mindfulness before this project (n=4). See Table 2 for therapist characteristics.
All patient facilitators (Textbox 1), patient barriers (Textbox 2), therapist facilitators (Textbox 3), and therapist barriers (Textbox 4) could be divided into the following four themes: treatment setting, treatment format, role of the therapist, and patient characteristics. First, patients' facilitators and barriers are presented per each theme. Then, therapists' facilitators and barriers are presented per each theme.

Patient Theme 1: Treatment Setting-Facilitators
Treatment setting concerned subthemes on the external conditions of the eMBCT: flexibility of timing, the individual nature, and the home practice environment of the training.

Time Management
It was considered convenient to be able to manage your own time schedule, which increased treatment receptivity. One patient stated the following:

Illness Barriers
One patient indicated that her cancer type caused her to have trouble reading. As the eMBCT was mainly text-based, this was a problem to her. She stated the following:

It was mainly physical, I didn't have the energy and my vision is in such a bad state. Even with medication, my vision is bad. And my eyes itch and burn and hurt.
[Female bone marrow cancer patient (palliative), 55 years, no start]

Lack of Information
Moreover, patients indicated that they would have wished more information on the way the platform and course were organized before the start of the training. One patient stated:

Patient Theme 2: Treatment Format-Facilitators
The treatment format theme included codes on the facilitators and barriers of the means by which the eMBCT was internally organized and delivered.

Website
The website was accessible and navigating throughout the website was easy. One patient stated:

Patient Theme 3: Role of the Therapist-Facilitators
This theme included codes on the role of the therapist and the way the therapist facilitated or hindered participating in eMBCT.

Practical Guidance
Patients indicated that the therapist was often able to clarify practical aspects that were unclear. One patient stated:

Asynchronicity
The asynchronous nature of the feedback proved to be a barrier.
According to the patient, the written feedback of therapist did not seem to encourage a dialogue but rather seemed limited to giving responses to questions. One patient stated:

Patient Theme 4: Patient Characteristics-Facilitators
Individual characteristics seemed to influence the fit between patient and eMBCT.

Writing Fluency
The ability to express themselves in writing was very helpful for some to give words to their subjective experiences and to ask for clarification to the therapist if it was necessary. One patient stated:

Writing Fluency
The heavy reliance on writing skills was a barrier to some patients. One patient stated:

Anonymity
The fact that patients were able to write about their experiences rather anonymously was helpful in opening up to experiences, which meant that in general, they shared their experiences in rather great detail. Moreover, it rendered the therapist to use patients' own quotes. One therapist stated:

Patients think I don't see them and they don't see me.
They tend to confide more to a diary.

Technical Issues
Therapists indicated that technical issues also proved to be a barrier to treatment continuity. One therapist stated:

Self-Discipline
Therapists indicated that for some patients, the eMBCT was partly a training in self-discipline, which supported patients' self-sufficiency after the training. One therapist stated:

Self-Efficacy
In the eMBCT, patients need to be resolute and determined. This was mentioned as a barrier to complete eMBCT. One therapist stated:

Principal Findings
The aim of this study was to gain qualitative understanding of the facilitators and barriers of eMBCT in a sample of heterogeneous cancer patients. Both eMBCT completers and dropouts participated in posttreatment interviews. Moreover, we conducted a focus group interview with eMBCT therapists. In all, this study adds to the existing quantitative evidence for eMBIs in cancer [7,8] by providing a qualitative perspective. Four overarching themes emerged, which were largely convergent between patients and therapists: treatment setting, treatment format, role of the therapist, and patient characteristics. Patients and therapists are much more flexible in when, where, and how they engage in eMBCT compared with group-based MBCT. However, most eMBCT advantages seemed to come at a price. When patients and therapists mentioned a certain aspect as facilitating (eg, the individual setting: not having to cope with other patients' stories), they also mentioned it as barrier (no peer support).
Patients and therapists reported similar advantages and disadvantages of the timing, the individual nature, the asynchronous nature (for patients, this was detrimental to the relevance of therapist feedback, and for therapists, this was a threat to treatment continuity), the diaries, and the importance of self-discipline. The fact that so many aspects of the eMBCT were mentioned both as facilitator and barrier emphasizes the importance of offering flexibility in eMBIs [21].
There were also differences between patients and therapists. As known from a previous qualitative study on eMBCT [21], bearing the responsibility for time management was a barrier for patients. For therapists, the eMBCT seemed to require a larger time investment compared with group-based MBCT. Moreover, therapists were more concerned with the (vulnerability of) continuity of the training. They also mentioned that missing out on nonverbal information rendered them unable to spot patient withdrawal at an early stage, and to determine the reason for empty diaries. Furthermore, therapists seemed more bothered by communicating mindfulness values in eMBCT than patients. Patients specifically mentioned asynchronicity as barrier to the role of the therapist because the asynchronous communication hindered emergence of a dialogue.

Clinical Implications
Although studies to date do not suggest that differences between how therapists handle the contact with their clients explain much variance in treatment outcome [16], the necessity of training and support for Internet-based therapists should be acknowledged. New eMBCT therapists should understand the importance of flexible availability and the dynamics of asynchronous interaction to pick up early signs of patient withdrawal.
The current eMBCT was individual, asynchronous, and therapist-assisted. One important adaptation may be to offer a peer support group [26]. The group context in MBIs supports perspective taking and the transition from personal story into investigation of common patterns of distress [6], and may foster skills relevant to valuing self and feeling close to others, which may help participants feel less isolated [21]. As a stand-alone intervention, formal online peer support group interventions for cancer patients have demonstrated preliminary feasibility and effectiveness [27].
Another consideration may be to employ a synchronous videoconferencing format [13]. This takes away the barriers associated with asynchronous delivery and may facilitate dialogue with the therapist and peer support. A possible caveat may be that videoconferencing does not alleviate the scheduling issues inherent in group eMBIs [13] and that synchronous videoconferencing solutions are technically much more demanding. An alternative to videoconferencing may be to include synchronous written conversations (or "chats") with therapists or trained volunteers. Chats are becoming increasingly popular as Web-based mental health interventions by themselves and show inconclusive but promising evidence [28].
Eventually, one could employ a blended format, combining the advantages of Web and group-based therapy [29]. Blended eMBIs could have group-based group meetings at the start, midst, and end of the programme. The meeting at the start of the intervention could be used to address practical and technical matters, a midtreatment meeting to address common barriers experienced by patients during practice, and meeting at the end to say goodbye to each other and support patients to take responsibility for the continuation of their mindfulness practice in the future. In between group-based sessions, patients could be offered online sessions. In our view, these practical arrangements could greatly improve the acceptability and effectiveness of eMBCT.

Research Implications
Previous studies have provided encouraging quantitative evidence, for example, eMBIs in cancer patients [13]. Together with this study, these results provide support for a larger, quantitative trial directly comparing eMBCT with group-based MBCT for cancer patients. Moreover, it would be interesting to directly compare individual eMBCT with individual group-based MBCT. Future trials should test for differences in treatment accessibility, program adherence, and treatment outcome between eMBCT with and without peer support, with and without synchronous communication modalities, and with and without therapist assistance [17]. This would allow us to further elucidate the predictors and mediators of treatment effect in Internet-based interventions [30] to help us determine which patient to offer group-based versus Internet-based treatment. Moreover, all of the abovementioned design alterations likely impact cost-effectiveness of the interventions, which should be considered [31]. Thus, future studies should preferably assess how different eMBCT delivery formats influence program adherence, mindfulness skills, and treatment outcome, and how different versions of eMBCT delivery formats compare both qualitatively and quantitatively with group-based MBCT.

Strengths and Limitations
This is the first study to qualitatively explore facilitators and barriers of eMBCT for cancer patients. The relatively large sample size enabled us to reach data saturation and report a broad view of experiences. Moreover, we interviewed both completers and dropouts. Furthermore, we had the opportunity to gather data in the therapists. Nevertheless, our results should be interpreted within the limitations of our findings. We did not perform member checks to ensure validity of the verbatim transcripts. Moreover, the sample of the larger RCT consisted of cancer patients who self-selected themselves for a trial on an MBI. This implies that our findings cannot be extrapolated to cancer patients in general. In addition, some patients or therapists who participated in the training and were invited for focus groups or individual interviews declined participation, which may further limit the generalizability of our findings to all participating patients.

Conclusions
We aimed to gain understanding of the facilitators and barriers of individual, asynchronous, and therapist-assisted eMBCT for cancer patients. Patients and therapists reported similar advantages and disadvantages of the timing, the individual nature, the asynchronous nature, the diaries, and the importance of self-discipline. Future studies should assess how different eMBCT delivery formats could further improve treatment accessibility, program adherence, and treatment outcome.