This qualitative study included in-depth semistructured interviews to examine the barriers and facilitators that cancer survivors with chronic painful CIPN experienced in using an online self-management intervention for cancer survivors with chronic painful CIPN, called Embrace Pain [21]. The interviews focused on individuals’ subjective experiences regarding their use of Embrace Pain.

The positionality of the research team is presented in Table 1.

The online self-management intervention Embrace Pain includes psychoeducation and all ACT processes. Patients learn to perform value-oriented activities in daily life with pain, to reduce pain avoidance and increase pain acceptance. ACT is a third generation of cognitive behavioral therapy (CBT) [23]. The main therapeutic goal of ACT is to focus attention on personally valuable activities by stimulating pain acceptance, as an alternative to (experiential) avoidance. ACT has been proven effective for several types of chronic pain [24,25] and is increasingly implemented in global cancer care [26,27]. However, there had been no previous research on ACT for CIPN, although a study into cognitive and behavioral intervention for CIPN had demonstrated positive results [1]. Due to similar aspects and the greater suitability of ACT compared with CBT for the current patient population, it was expected that the current ACT-based intervention would yield positive effects [22].

An overview of the intervention of the main study is illustrated in Table 2 [21,22]. The RCT’s method, including an overview of the intervention, was previously published in a protocol paper [22] along with a qualitative paper describing the patient-centered development [21]. Effects of the RCT were still pending at the time of writing. Furthermore, within this study, there were 2 arms, namely the guided ACT group and the WLC group. The only distinction between the guided and unguided groups was the provision of guidance through therapeutic email communication. Despite this, the content of the intervention administered to both groups was identical, which is why both groups were examined in this study. Participants could engage with the intervention autonomously by navigating through it individually. The guidance provided was minimal and had a low threshold, consisting solely of an asynchronous chat function.

Participants were selected from the group of participants in the RCT of this project, which was conducted between December 2021 and July 2024 [22]. Inclusion and exclusion criteria are described in the protocol paper previously published [22]. Inclusion criteria regarded being 18 years or older, being identified as having painful sensations by a clinician or self (ie, aching, burning, “pins-and-needles,” shock-like, painful tingling, numbness, and cramps), and experiencing sensations for at least 3 months. Furthermore, a score of 3 or higher on an 11-point pain intensity scale (numeric rating scale), the pain was not present before the administration of chemotherapy, chemotherapy concluded at least 6 months ago, and a score of 1 or higher on the European Organisation for Research and Treatment of Cancer QLQ-CIPN20 Questionnaire was required. The RCT included a total of 112 participants, of whom 25 were invited to participate in an interview. In a previously administered questionnaire within the RCT, patients indicated their willingness to participate in an interview at a later stage. Those who responded affirmatively were considered potential interview participants. Potential interview participants were approached via email within a month after completion of the intervention. Not all patients who expressed willingness in the questionnaire were invited to participate in an interview, because recruitment was stopped when data saturation was reached. During participant selection, efforts were made to ensure a balanced distribution between the guided and unguided groups, as well as gender distribution. It was decided to include both guided and unguided participants, as it was anticipated that the strengths and areas for improvement of the Embrace Pain intervention would exhibit significant overlap between both groups. Efforts were also made to include participants with different levels of engagement, attempting to include both patients with high and low engagement levels. This was estimated by one of the researchers (DvdG), who was aware of the patients’ usage by accessing user data in the intervention’s online system. Excluded were patients who had not activated their account of the intervention Embrace Pain and who were unable or unwilling to participate in an online video call via Microsoft Teams. Patients were able to participate regardless of the time elapsed since they had completed the intervention. Participants sometimes had previous contact via email or phone regarding their participation in the RCT. Some participants had participated in a previous interview with the same researcher (DvdG).

The interview scheme was based on 2 qualitative studies into subjective cancer patients’ experiences of online self-management interventions [16,17]. It is important to note that these studies followed the definition of adherence, while the operationalization of these studies fits the concept of engagement and of the subjective experience of use as defined in our study. Furthermore, the input from previous patient interviews regarding user needs in the context of this online intervention [21] was used as input for the interview scheme. The final semistructured interview scheme is presented in Table 3.

Interviews were conducted by 1 interviewer with each participant. Interviews were conducted by one of the authors (DvdG) between September 21, 2022, and December 19, 2023. Interviews took place via video calling through Teams. Recruitment for interviews stopped when data saturation was reached [28], indicating that additional interviews yielded no novel information, with recurring topics consistently emerging.

Thematic analysis was used for analyzing the data [29]. Audio of the interviews was recorded and transcribed verbatim by 1 author (DvdG) and by a Psychology postgraduate student from Tilburg University. These students contributed to the research study in the context of their data collection task as part of their master’s program. They were mentored by FM or HT in writing their thesis on a related topic and did not contribute to the development of this manuscript. Transcripts were not returned to participants for comment and correction because of the potential burden on patients. DvdG reviewed the transcriptions to confirm accuracy.

All transcripts were inductively coded, deriving themes directly from the data. Text fragments were coded when they involved barriers or facilitators in the use of the online training Embrace Pain from the patient’s perception. When patients discussed potential barriers and facilitators relevant to other patients but not applicable to their situation, these were not coded. The analysis began with open coding, where the primary themes of facilitators and barriers guided the coding process. Once all transcripts were coded, an initial codebook code was reviewed and adjusted if necessary. Axial coding was then applied to explore how the codes related to the themes’ facilitators and barriers. This process was repeated once for verification. Subsequently, the frequency with which each code was mentioned, as well as the number of participants who mentioned each code, was calculated. Ultimately, condensed meaning units were generated for every text fragment that was coded, capturing the core message of each quote. This approach facilitated a comprehensive understanding of the data, organized around the identified themes.

Interviews were coded in Atlas.ti 23 (Lumivero). One transcript was coded by DvdG and FM independently. Differences in coding were thoroughly discussed to ensure consistency and accuracy. Following these discussions, a consensus was reached regarding the appropriate coding approach. Subsequently, DvdG undertook the coding of the remaining transcript, which was performed inductively as well. In cases where uncertainties arose, these were addressed through further discussion to ensure clarity and agreement on the coding process. Due to the differences in background and expertise of DvdG and FM, these discussions were conducted in a manner that allowed for as objective an evaluation as possible.

This study was approved by the Medical Research Ethics Committee Brabant, the Netherlands (#NL78436.028.21) and the Ethical Review Board of Tilburg University (School of Social and Behavioral Sciences), the Netherlands (#TSB_RP70). Participants signed informed consent upon enrollment in the RCT that included the interview component. Participants did not receive financial compensation for participating in either the interview or the RCT. Only anonymized data were used in the study. Each participant was given a participant number. This number was kept in a secure password-protected document along with the necessary personal information (name, date of birth, email address, date, and time of interview). Only DvdG had access to this file. For all elements of subsequent procedures, including audio files, transcription, and coding, participant numbers were used. This means that no personal information was available in the transcription and analysis process.

The most common barrier identified was the program schedule. Many patients indicated having insufficient time for completion:

Furthermore, many patients indicated that the timing was unfavorable due to personal schedule or holidays.

Another barrier mentioned by several patients was the overall burden of the intervention. Some patients indicated that it was confronting to work on the training:

Also, some patients indicated physical limitations or cognitive limitations due to chemo brain:

Burden is also related to remembering to attend the training during the intervention period, as indicated by some patients:

Many patients indicated barriers related to guidance. Several patients who did not receive guidance expressed a need for some form of guidance to be able to ask questions:

Some other patients indicated that they would have preferred guidance for motivation or reminders:

Furthermore, 1 patient who did receive guidance indicated that the delivery of guidance (ie, by email) was not as required:

Furthermore, several patients indicated that irrelevant content was a barrier. This meant that some patients experienced no connection to their own situation:

In addition, some patients indicated that they were already familiar with the content:

Several patients reported experiencing barriers related to the included mindfulness exercises. This relates to the length (P7) or quantity (P9) of mindfulness.

Some patients reported experiencing barriers regarding usability. Some patients indicated difficulties navigating through the training:

Also, one patient indicated that the lack of face-to-face communication was a barrier:

Furthermore, one patient indicated problems logging in:

Finally, barriers were reported by a few patients due to missing content. One patient indicated that the exercises contained insufficient explanation and examples:

Another patient indicated that they missed the opportunity for peer contact:

A facilitator that many patients mentioned was usability. Several patients indicated benefiting from the flexibility of the training, allowing them to use it at their convenience, pausing when needed:

Several patients also indicated that the freedom in the training facilitated their use:

Also, a few patients indicated that the training was easy to use:

Furthermore, many patients indicated recognition as a facilitator. Several patients indicated that they found recognition in the fictitious personas:

Several patients also indicated that the examples and personas clarified the information and theory that was provided:

In addition, many patients indicated facilitators related to positive self-management. For several patients, the training was a way to find reflection:

Also, several patients indicated that they appreciated being able to further develop themselves through the training:

Furthermore, some patients indicated that they received confirmation of their coping through the training:

Many patients indicated the program schedule as a facilitator, which was mainly related to the flexibility of the program schedule. This is mainly related to being able to work on the training whenever and wherever the patient wants:

Some patients appreciated the option to pause or request extensions for the training, particularly during personal circumstances that temporarily hindered use:

There were also some patients who indicated that they valued being able to make their own choices about completing the exercises:

Several patients indicated symptom management as a facilitator. Several patients felt that participation in training might improve their symptoms:

Furthermore, some patients indicated that they liked getting tips:

There were also some patients who experienced psychoeducation as a facilitator:

Finally, certain elements in the training were identified as facilitators by several patients. Several patients indicated that the relevance of the training was a facilitator because this made it possible to apply the training to one’s own situation:

Furthermore, several patients liked the option for guidance for various reasons, such as motivation, clarification of information, and feedback:

In addition, several patients indicated that they experienced both experiential exercises (P5) and mindfulness exercises (P8) as facilitators.

Content related to value-based living was also perceived by several patients as a facilitator:

This study qualitatively examined barriers and facilitators to engagement of an online self-management intervention for cancer survivors with chronic painful CIPN. Participants generally reported more facilitators than barriers. Facilitators most often related to the perceived ease of use of the intervention, feelings of recognition and validation, the opportunity to actively engage in self- and symptom management, and a favorable program schedule. In contrast, commonly reported barriers included an unfavorable program schedule, perceived burden, lack of guidance, and content irrelevance. Notably, some factors, such as the program schedule, were described both as facilitators and barriers, underscoring the influence of individual preferences and circumstances. Factors seem to include a combination of mainly psychological, intervention-related, and technology-related characteristics. Furthermore, certain aspects are related to illness or individual circumstances. These findings align with previous research on barriers and facilitators [17,18], as well as theoretical frameworks like the conceptualization of engagement in digital behavior change interventions [7,12].

This study revealed that many patients felt a strong sense of recognition in the online intervention, which is likely related to the inclusion of quotes and personas added to make the program more personal, which were based on real patient experiences [21,30,31]. These patient experiences, depicted in a first-person narrative, can contribute to the feeling of social support that can help alleviate feelings of isolation for patients in coping with symptoms [32-34]. However, patient experiences shared by individual users may contain unbalanced information or misinformation [35]. In this online intervention, therefore, researcher-written narratives based on experiences shared by patients were incorporated. This can be seen as an accessible and controlled way of incorporating patient experiences in online interventions to increase the sense of social support.

Furthermore, this study has shown that many patients perceived participating in the online intervention as a pleasant way of engaging in positive self-management. Self-management empowers cancer survivors to enhance their health and well-being [36]. Previous research has shown that theoretically based self-management interventions for various types of chronic diseases, including cancer, can positively influence outcomes such as quality of life, coping, health goals, self-regulatory skills, and self-efficacy [36-39]. Self-efficacy for symptom management in patients with cancer is defined as “the ability to implement behaviors to prevent, recognize, and relieve symptoms” and is a vital component in enabling patients to effectively self-manage their cancer-related symptoms [40,41]. Self-efficacy empowers patients to change their self-management behaviors and helps patients set realistic goals in coping with the symptoms [40,42]. This closely aligns with the acceptance of symptoms and the pursuit of a value-driven life, which reflects the ACT-based approach of the online intervention used in this study. This alignment may help explain why active engagement in positive self-management and symptom management emerged as a frequently reported facilitator in this study.

While for many patients, the flexibility of the training program was a facilitator, many patients experienced the training program as a barrier. In this case, this is related to both unfavorable start time and insufficient time to complete the training due to personal situations, in the context of the study. This aligns with previous studies regarding the use of online psychological interventions, which have shown that time-related factors (eg, not having enough time or being too busy) deteriorate adherence and engagement [15,17,43]. It is important to mention that in this study, the start of the training was not aligned with the participants in advance. In the study, participants received an invitation at the start, and had 8 weeks for completion, which was only extended upon request. Based on this study’s results, it is recommended to initiate online interventions in coordination with participants. In addition, it is suggested that the intervention’s total duration is sufficiently long, incorporating several weeks of flexibility in scheduling (eg, 9 weeks for an online intervention with 6 components). It can also be considered to allow unlimited access to the intervention. We also recommend that participants be encouraged to contact the contact person or supervisor of the online intervention if they encounter problems, so that it can be assessed when an extension of access is needed to avoid dropout.

Furthermore, barriers identified in this study included the burden and chemo-brain that patients may suffer from. Chemo-brain is known to be a consequence of chemotherapy, which is characterized by self-perceived thinking and memory problems [44]. In this study, for example, patients reported that the training was too stressful or that they had concentration problems. However, this does not only apply to chemo-brain, but also to other physical and mental symptoms caused by cancer or cancer treatment, which causes people to have more difficulty performing daily activities. These may include fatigue, which increases the need for more frequent breaks, or disease-related fear, which may make individuals more vulnerable and prone to heightened feelings of fear in general. Thereby, having a higher age with additional physical and cognitive challenges, which is often the case in cancer survivors, can make this more difficult. Therefore, a low cognitive effort of the online intervention is necessary for proper understanding of the content and effective use [45]. This can be achieved by dividing the intervention into short, manageable sections to reduce the complexity on the pages and to provide simple navigation between pages and sections [45]. Patients also indicated that they valued the ability to schedule the use of the intervention within the provided time frame of the training themselves, as well as the ability to pause when needed, which has also been shown in previous qualitative research regarding computerized CBT for depression [46]. This is also relevant for personal situations of participants, for example, in case of receiving a new cycle of chemotherapy treatment during the intervention period. During the study, it became evident that not all participants were fully aware of this optional flexibility in the schedule. For instance, 1 participant did not report receiving a new cycle of chemotherapy, which ultimately limited her time to complete the intervention due to restricted access, as a modified schedule had not been discussed. This also applied to, for example, holidays or family issues, which aligns with earlier research [11]. Given the complex circumstances participants often encounter, it is essential, in addition to providing an intervention that considers factors such as breaks and division into manageable sections, to engage in dialogue with users to identify an approach that aligns with their individual needs, ensuring the intervention is as adaptable as possible.

In terms of guidance, patients who had access to it indicated finding it valuable, despite not all of them using it consequently. Among those without access to guidance, some expressed a clear need for it. For example, these patients would have appreciated the opportunity to ask questions for clarification on information or exercises, as well as to discuss personal circumstances that made engaging with the online intervention more challenging. Previous research has shown that online interventions with guidance compared with nonguided versions result in better outcomes in satisfaction, usage, and adherence [19,20,47]. Also, an earlier study examining a self-guided online CBT intervention for patients with CIPN suggests that counseling could be a useful addition [1]. However, a systematic review examining engagement with online psychological interventions found that preferences for guidance varied greatly, ranging from a need for intensive support to enhance engagement to a preference for no guidance to maintain anonymity and freedom [15]. Therefore, offering optional guidance rather than mandatory support appears to be recommended.

Some limitations of this study should be mentioned. First, no results were available from patients with very low engagement levels, since these people were reluctant to participate in interviews. Several attempts were made to include these patients, but despite these efforts, they remained unwilling to take part, and some potential respondents did not respond at all. It was not possible to engage in discussions with individuals who were dissatisfied or who had discontinued participation prematurely, suggesting that this group was unwilling to discuss their experiences with us. The sample in this study is therefore not representative of the entire user group. Barriers that might exist with low-engaged patients, such as lack of time and low levels of motivation [15,17], were therefore not completely reflected in this study. Furthermore, people with low levels of use might also comprise patients who realized they were already proficient in managing their symptoms and consequently discontinued the intervention. Unfortunately, obtaining insights into this matter was not possible because these patients did not participate in the interviews. Therefore, the results of this study must be interpreted with caution. Second, there was a lot of difference in the time between intervention completion and the interview. Possibly, patients with a longer period between the intervention and interview may have remembered fewer details of their experience with the intervention, which could have also affected the results of this study. Third, the results of this study may be less applicable to other self-management interventions. Each intervention has its characteristics in terms of tasks, technology, and users, all of which affect ultimate use [7,48-50]. Consequently, the experiences encountered by patients also vary. Fourth, the study sample included highly educated and highly motivated patients who experienced little or no difficulty with language and the use of technology. Therefore, the transferability of the results to groups with lower education levels and lower technology literacy levels might be limited. Fifth, no participant checks were performed because of the possible burden on patients. However, in terms of validating the results, it is desirable to perform participant checking in future studies.

Some strengths should also be emphasized. First, interviews were semistructured, enabling patients to freely express their individual experiences without being constrained by the interview’s format. The core questions were addressed in all interviews, leading to a consistent structure across all interviews. Second, interviews were conducted with patients from all conditions, including both those who received therapeutic email guidance and those who did not. This enabled the assessment of the experienced barriers and facilitators associated with the availability or unavailability of guidance. Third, interviews took place via video call, enabling vulnerable and disabled patients with cancer to participate without being hindered by, for example, physical limitations.

This study qualitatively identified facilitators and barriers to engagement in an online self-management intervention for cancer survivors with chronic painful CIPN. Facilitators and barriers varied greatly and were dependent on personal situations. Recommendations could be given in terms of recognition, program schedule, burden, and guidance. Consideration of such recommendations following qualitative evaluations is imperative to ensure careful implementation in practice.