Over the last 20 years, cancer has increased worldwide [1]. Reassuringly, the survival rates are also increasing [2,3]. However, a significant proportion of cancer survivors have overweight or obesity [4]. Oncologic treatment side effects, such as loss of muscular strength, fatigue, and physical inactivity, contribute to weight management issues during cancer and survivorship [5]. Overweight and obesity are associated with all-cause mortality, increased risk for cancer development, and increased risk of development of secondary cancer or subsequent primary cancer in cancer survivors [6-8]. Healthy lifestyle behaviors, such as regular exercise and healthy diet, have the potential to reduce treatment-associated morbidity and mortality in cancer survivors [9]. Therefore, interventions that support cancer survivors’ management of lifestyle behaviors are imperative.

Mobile health (mHealth) involves the use of mobile technologies, such as smartphones, tablets, apps, and wearable activity trackers, to improve health care, health outcomes, and public health. mHealth has been associated with significant reductions in weight and BMI [10] and positive health behavior changes [11,12]. Therefore, mHealth tools have the potential to meet the need for interventions that support weight and lifestyle management and, at the same time, are low resource and cost effective. At the same time, it is critical that the design of mHealth interventions is based on theory and evidence [13]. Evidence from systematic reviews suggests that using behavior change techniques (BCTs) significantly increases the success of weight management programs [14]. Interventions, including a greater number of BCTs, were associated with greater changes in health behaviors [15]. This was also the case for digital interventions [16]. Although there are a good number of lifestyle interventions for cancer survivors using BCTs [14], there are fewer mHealth interventions that have adopted this evidence-based approach to development, in the sense that very few have included relevant BCTs [17,18].

Interventions that are evidence based and effective may not be deemed acceptable to the prospective recipients. Acceptability is a key construct in intervention evaluation and is explicitly recommended in the UK Medical Research Council and National Institute of Health Research guidance on evaluating complex interventions [19]. Acceptability is often inferred from behavioral measures (eg, uptake, adherence, and retention), with fewer studies using explicit measures of acceptability or gathering feedback through postintervention interviews or focus groups [20]. Sekhon et al [21] proposed a theoretical framework to advance the systematic assessment of the acceptability of health care interventions. They defined acceptability as a “multi-faceted construct that reflects the extent to which people delivering or receiving a healthcare intervention consider it to be appropriate, based on anticipated or experienced cognitive and emotional responses to the intervention” [21]. The framework takes into account constructs, such as the participant’s attitude toward the intervention, the burden of participating, and the participant’s understanding of the intervention and how it is intended to work (a full description is given in Textbox 1). This method of assessment comprising multiple constructs offers a more fine-grained approach to detecting the source of acceptability issues that can be used to improve future interventions [20].

Qualitative methods add significant value to trial research [22,23], contributing feedback that would not otherwise be captured. For successful uptake and implementation, it is important that the content, design, and delivery of interventions are considered acceptable to the target group [24]. In the area of mHealth, many studies have focused on the usability of technologies rather than the acceptability of interventions that use mobile technology. Some studies have supported the acceptability of mHealth solutions for people living with and beyond cancer [25-28]. However, acceptability is often operationalized behaviorally as adherence and engagement with technology. Few studies in this area have moved beyond behavioral measures or adopted a broader qualitative approach to understanding acceptability [29,30]. Therefore, we currently lack a robust understanding of the acceptability of mHealth behavior change interventions for cancer survivors with obesity or overweight. This study is the first to apply the theoretical framework of acceptability [21] to systematically assess the acceptability of an mHealth intervention for this high-risk group.

This study aims to examine the acceptability of the Moving On intervention to its recipients and to obtain constructive feedback on the intervention and participants’ recommendations for improving the intervention content or delivery. The actual trial of the intervention was completed before this study. Full details of the trial protocol are reported elsewhere [31], and this paper presents the results of the mixed methods analysis of the retrospective acceptability of the intervention.

The Moving On study is a 2-arm parallel randomized controlled trial (RCT) of an intervention for cancer survivors with a BMI of 25 or more (International Standard Randomized Controlled Trial Number (ISRCTN): 18676721). Participants were assessed for clinical outcomes (ie, anthropometric measurements and functional exercise capacity), psychological outcomes (eg, quality of life), and behavioral outcomes (ie, dietary behavior and physical activity) at baseline (T0), at postintervention (12 weeks later: T1), and at the 24-week follow-up (T2).

The intervention uses mHealth in combination with BCTs to increase physical activity and improve diet, thereby improving health and well-being outcomes. This is a complex intervention targeting individuals’ self-management of lifestyle behavior, which is described in detail elsewhere [31]. In short, the intervention had 2 components.

The design of this intervention is well matched with the National Institute for Health and Care Excellence guidelines for weight management of people with obesity [32]. They identify multicomponent interventions as the treatment of choice, which should include effective behavior change strategies aimed at increasing physical activity or decreasing inactivity, improving dietary behavior and quality of diet, and reducing energy intake.

This study was a mixed methods study nested within an RCT. In the context of the RCT, postintervention semistructured interviews were conducted, and a brief quantitative survey was administered to assess the retrospective acceptability of the intervention from the perspective of the recipients. Response rates and retention rates for the full trial were also used as indicators of the acceptability of the intervention. The design of this study was approved by the Research Ethics Committee of the National University of Ireland, Galway, and by the Research Ethics Committee at Letterkenny University Hospital.

The Moving On study was a nonblinded RCT implemented in the oncology outpatient department at Letterkenny University Hospital in County Donegal, Ireland. Participants were cancer survivors with a BMI of 25 or more who had completed active cancer treatment. For this study examining the acceptability of the Moving On intervention, only participants randomized to receive the intervention were eligible for participation.

At the 24-week follow-up, all participants in the intervention condition were asked to complete a 5-item survey to provide feedback on the intervention. For the qualitative component, all participants in the intervention condition were included and invited to participate in the qualitative study. No exclusion criteria were applied. An invitation to participate was included with their appointment letter for the final 24-week assessment of the intervention. One week before the final assessment, the invitation letter was followed up with an SMS to the group. The participants were asked to contact the research team if they were interested in participating in the qualitative study. Interviews were conducted with all volunteers who were available to be interviewed at their 24-week assessment (ie, 12 weeks postintervention). Written consent was obtained from 13 participants. The sample size was primarily determined by the size of the participant pool (ie, the intervention group) and the availability of participants. Within the context of this study, the data collected were judged by the research team to be adequate in both amount and variety to answer the research question. The participants did not receive any compensation.

Thematic analysis was conducted using the essentialist or realist approach (ie, reporting experiences, meaning, and the reality of participants). This is a semantic-level analysis that looks for explicit meanings in data [33]. All interviews were individual and face-to-face and were conducted in the study setting. The semistructured interview topic guide was based on the theoretical framework of acceptability of health care interventions (Textbox 1) [21]. The following constructs were deemed the most relevant and used to inform the topic guide: affective attitude, burden, intervention coherence, perceived effectiveness, and self-efficacy. Participants were also asked about their motives for participating in the study and to provide feedback and recommendations for improving the Moving On intervention. The topic guide is included in Multimedia Appendix 1.

A 5-item self-report measure of acceptability was created for this study, corresponding to the 5 constructs of the theoretical framework of acceptability of health care interventions used to inform the qualitative topic guide. Using a 5-point Likert scale, participants in the intervention were asked to rate their satisfaction with the intervention (affective attitude), the amount of effort required to participate (burden), the extent to which they understand the intervention and how it is intended to work (intervention coherence), the perceived effectiveness of the intervention, and their confidence in performing the behaviors required to participate (self-efficacy). Response rates and retention rates for the trial were used as behavioral indicators of the acceptability of the intervention.

Descriptive statistics were used to analyze the quantitative data using IBM (International Business Machines corporation) SPSS Statistics 24. Interviews were audio recorded and transcribed professionally. Transcripts were imported into NVivo 12 for analysis. Following the procedure outlined by Braun and Clarke [33], the transcripts were initially read by the primary coder (JG) several times to become familiar with the data. JG and JM independently coded 2 transcripts and agreed upon initial codes. The remaining transcripts were coded by JG. The analysis followed a deductive approach guided by the theoretical framework of acceptability. Transcripts were read in full and coded line by line to identify units of meaning relevant or interesting to answering the research question. These initial codes were then grouped together into the 5 domains described in the theoretical framework of acceptability and into a theme reflecting recommendations to improve the intervention by a multidisciplinary team (JG, JM, JR, MK, JW, and AG). Finally, members of the research team with practice-based experience (JR and MK) reviewed all the transcripts in relation to the themes identified. After analyzing all the transcripts, the analytic team was satisfied that no further interviews were necessary as the codes and themes identified were sufficient to develop a thorough understanding of the acceptability of the intervention and, therefore, meet the aims of the study [34]. The rigor of this study was enhanced by the secondary coding carried out by a multidisciplinary team of 6 such that each transcript was reviewed by at least three team members.

A total of 13 participants (2 males) were interviewed, and 36 participants completed the quantitative survey. Participant characteristics are described in Tables 1 and 2.

In opening the semistructured interviews, participants were asked about their motives for participating in the trial. Participants were clear that interventions to improve the health and well-being of cancer survivors were both acceptable and necessary. One participant had felt “very traumatised” by the experience (010). Other participants reported living with and beyond cancer as “very difficult” (012) and felt isolated because people who have not had cancer “don’t...really understand” (007) what they have been through and what they continue to deal with. They emphasized the desire for continuing support during survivorship. They also described ongoing health and psychological issues and having intrusive thoughts and concerns about these issues:

Findings on the perceived acceptability of the Moving On intervention, developed through deductive analysis guided by the theoretical framework of acceptability, are presented under Theme 1: Acceptability, with 5 subthemes relating to the 5 aspects of the framework applied in this analysis. Theme 2 describes participants’ recommendations for improving the Moving On intervention.

The results are consistent with constructs from the theoretical framework of acceptability of health care interventions [21], suggesting high acceptability of the intervention. Overall, the mixed methods results indicate that the majority of participants were satisfied and had a positive attitude toward the intervention, with high perceived effectiveness and high coherence. Most participants were confident that they could perform the required behaviors for the intervention, and qualitative data showed that the design of the intervention helped build self-efficacy in performing health behaviors. In both quantitative and qualitative responses, there was greater individual variability in participants’ determinations of the intervention’s level of burden and participants’ sense of self-efficacy. The high response rate and low rate of attrition in the trial support these qualitative and quantitative findings. Therefore, we are confident that this intervention using mHealth and BCTs to improve health and well-being outcomes is acceptable to recipients. The findings of this study are consistent with research using quantitative and qualitative methods that have found that mobile and digital health solutions are acceptable to people living with and beyond cancer [25,27,28,30] and studies supporting the acceptability of health behavior change interventions more broadly [35,36].

Participants reported a high level of satisfaction with the intervention. Most participants held a positive attitude toward the intervention and enjoyed many of its elements. In particular, mHealth components were rated positively. In terms of intervention coherence, participants’ understanding of the aim of the intervention was less focused on weight management and more focused on moving on psychologically from cancer. The intervention was perceived to have high efficacy, and the participants described the effects in different areas, especially physical health and fitness, mental health and well-being, and self and self-image. Although weight loss was an important outcome of the intervention, it is noteworthy that weight loss did not feature strongly in participants’ responses in terms of the effects of the intervention. This may be linked to the finding that many participants judged the goal of the intervention to be much broader than weight loss. Discordance between patient and provider priorities with regard to health goals and treatment outcomes has been previously observed in the management of cancer and other chronic conditions [37-39]. This highlights the importance of patient-centered care [40], along with the person-centered design of health care interventions [41]. There was a variety of responses in terms of the level of burden posed by participating in the intervention and participants’ confidence that they could perform expected behaviors. The low level of attrition implies that participating in the intervention itself was not particularly onerous or burdensome; however, the qualitative responses reveal how challenging health behavior change is on a personal level. This is evident in the low success rate of behavioral weight loss interventions [42]. mHealth tools may support health behavior change [43,44]; however, these tools alone do not necessarily increase self-efficacy or reduce the burden of changing well-established dietary and physical activity behaviors.

Participants were generally very positive about the content and delivery of the intervention. Most of the recommendations for improving the Moving On intervention focused on increasing the frequency, duration, and scope of the existing intervention components. For example, the goal setting intervention focused on increasing physical activity, and participants recommended that we extend this goal setting to dietary behavior. This is not unexpected, given reports that dietary behavior change was deemed the most difficult aspect of the intervention, and perhaps, this was because of them receiving less behavioral support in this respect. Clinical guidelines for obesity management recommend that interventions target both physical activity and diet [32,45]. Interventions using goal setting have typically focused on either diet or physical activity, rather than targeting both simultaneously [46]. However, goal setting has been used in interventions focusing on dietary behavior change with positive effects [47]. Participants recommended longer and additional lifestyle information and education sessions, with more time to interact with health care professionals and more goal setting SMS over a longer duration in declining frequency. There are a limited number of mHealth interventions using BCTs with cancer survivors; however, one previous study found that behavioral prompts sent by email in declining frequency over 12 weeks significantly increased physical activity in survivors of breast cancer [18].

It is clear from the feedback that participants also wanted more opportunities for social interaction and support from other participants during the intervention. This is consistent with the findings of another study on breast cancer survivors and their preferences for wearable activity trackers. Their recommendations were to incorporate online and in-person peer support and doctor monitoring [29]. Social support is a common component of psychosocial interventions aimed at improving emotional adjustment in people with cancer [48,49]. The Moving On intervention aimed to improve health and well-being outcomes by improving health behavior. However, the findings presented here highlight the broader issues that cancer survivors face. Participants’ recommendations for increased social support are in keeping with participants’ descriptions of the cancer experience as lonely and isolating. Together, these findings suggest that this mHealth behavior change intervention, although acceptable for self-management of lifestyle behavior, may not be sufficient to meet the full range of physical and psychological health needs of people living beyond cancer. Future research will be needed to determine whether the addition of social support components to the Moving On intervention has an incremental benefit on the effectiveness of the intervention.

Qualitative research can inform the development of future trials and interventions. Participants’ responses raised important questions about the ideal timing of the Moving On intervention. Some felt the burden of participating would be higher in the earlier stages of cancer survivorship when fatigue and treatment effects were most pronounced [50]. At the same time, others described feeling lost and afraid when their acute cancer treatment was complete and their follow-up would be an outpatient visit every 6-12 months, supporting the idea that this may be a good time to intervene. Participants also identified times when behavior change was more difficult, opening up the potential for adapting the mHealth intervention as a just-in-time intervention [51]. These are complex, technologically advanced interventions that provide real-time behavioral support, for example, when the participant has an opportunity to engage in a health behavior or is likely to experience a trigger for unhealthy behavior. A recent review found that these interventions are acceptable to participants and are often used in combination with the same BCTs as in this study (ie, goal setting, prompts, feedback, and action planning) [52].

Qualitative findings can also inform the interpretation of the results of RCTs. The assessment of study outcomes at baseline, at 12 weeks, and at 24 weeks were not coded as BCTs; however, based on the results of this study, they may have been important motivators of behavior change. It is possible that future iterations of the intervention that do not include assessment of clinical, psychological, and behavioral outcomes would be less effective. Our intention was that the goal setting intervention would motivate participants to increase their activity level. However, these qualitative results suggest that participants’ motivation was influenced by multiple factors, including self-monitoring their activity via Fitbit, an awareness of being monitored by the research team, accumulating self-efficacy, encouraging interactions with staff (at the lifestyle information and education session and SMS contact), and attending assessments.

Although not an explicit focus of this study, a small number of issues relevant to fidelity were described by participants during the interviews. Some participants spoke of being enrolled in other programs (eg, mindfulness courses and sports clubs), and although this was not discouraged, it was not measured or controlled for. Another participant revealed that he had broken his Fitbit and started using a different service (MyFitnessPal) without informing the research team. Other participants used the Fitbit app in ways they had been explicitly asked not to (ie, for food logging and calorie counting). Fidelity is an important aspect of evaluating complex interventions [53] and may be relevant to the acceptability of an intervention, but it was not measured explicitly in the trial and would be useful to consider in future testing of similar interventions.

Acceptability is an aspect of feasibility. However, future studies are needed to assess the feasibility of delivering this intervention in clinical practice. Although the intervention was perceived as low burden by participants, the burden of delivering the intervention in routine care may be high for clinical staff. There is significant promise of and potential for mHealth to limit the burden on the health care system [13]. However, the findings of this study suggest that in addition to mHealth solutions, cancer survivors want additional and extended support from their health care providers in relation to self-management of lifestyle behavior. In many clinical contexts, this may be challenging; however, this program could be facilitated external to the hospital setting, and advancements in digital health technology make this more of a possibility than ever before.

This study used qualitative research to assess the retrospective acceptability of the Moving On intervention. The person-based approach recommends in-depth qualitative research at the planning and development stage to enhance the acceptability and feasibility of interventions [41]. Unfortunately, participants’ perspectives were not considered during the design phase of the intervention. Reassuringly, the results of this study suggest that the Moving On intervention was seen as acceptable by those who participated in it. Relative to other qualitative studies in health research [54], the sample size was limited. At the same time, a sizable portion of the intervention group was interviewed (13/53, 24%). Furthermore, there was a limited number of male participants; however, this is reflective of the composition of the sample in the full trial. With regard to acceptability, the qualitative responses were consistent with the quantitative data with a larger sample size of 36 (representing 36/53, 68% of the intervention group). This concordance provides further support for our judgment that the number of interviews was sufficient to meet the goals of the analysis. Finally, this study focused on the acceptability of the intervention to those receiving the intervention only. It would be equally interesting and important to assess the acceptability of the intervention to those responsible for delivering it.

This study provides insight into the experiences of participants enrolled in the Moving On intervention. This mixed methods analysis identified that the intervention was acceptable to the participants. Participants spoke of cancer survivorship as lonely and isolating, and they wanted more social support and peer interaction to be incorporated into the intervention. Overall, participants were very positive about the intervention in terms of both content and delivery. As it was acceptable and was perceived as having the broad goal of moving on psychologically, if found to be effective, the intervention could be offered to individuals following completion of their active cancer treatment as part of standard care to promote the health and well-being of survivors.