Exploring the views and experiences of people recovering from a stroke about a new text message intervention to promote physical activity after rehabilitation—Keeping Active with Texting After Stroke: A qualitative study

Abstract Background Participating in exercise following a stroke is essential for recovery. When community‐based rehabilitation services end, some people struggle to remain active. We codesigned Keeping Active with Texting After Stroke (KATS), a text message intervention to support home‐based, self‐directed plans to continue exercising. KATS delivers a series of automated text messages over a 12‐week period from the point of discharge from National Health Service‐funded therapy. The aim of this study was to explore the views and experiences of the first cohort of participants to complete the KATS intervention about the meaning, engagement, workability and worth of the intervention. Methods We undertook a qualitative study, theoretically informed by Normalisation Process Theory. We conducted semi‐structured telephone interviews with people with stroke from two Health Boards in Scotland. Data collection took place over two phases, with each participant being interviewed twice: first, halfway through intervention delivery (Week 6) and then again at the end of the intervention (Week 12). All interviews were audio‐recorded, transcribed and analysed thematically. Results A total of 24 interviews were conducted with 12 participants. Our findings were organised around four overarching analytical themes: (1) making sense of KATS: timing and complementarity in the rehabilitation journey; (2) engaging with KATS: connection and identification with others; (3) making KATS work: flexibility and tailorable guidance; (4) appraising the worth of KATS: encouragement and friendliness. Participants differentiated KATS from current rehabilitation practice, finding it relevant, fitting and worthwhile. Variations were reported in engagement with behaviour change techniques, but participants were able to tailor KATS use, making it work for them in different ways. Conclusions Perceived benefits went beyond promoting physical activity, including feeling supported and connected. Future research will test the effectiveness of KATS in promoting physical activity and explore any associations with relevant social and emotional secondary outcomes. Patient or Public Contribution A research funding proposal was developed in collaboration with five people with stroke and three spouses. After securing funding, six people with stroke were invited to join the project's Collaborative Working Group, alongside health professionals and stroke rehabilitation experts, to codevelop the intervention and support the feasibility study.


| INTRODUCTION
In the United Kingdom and worldwide, stroke is among the top leading causes of death and disability combined, with the number of people living with stroke globally having almost doubled over the last 30 years. 1 Participating in exercise and physical activity following stroke contributes to substantial health benefits. The direct physical benefits of exercise can support recovery through the improvement of walking ability, balance and fitness. 2 In addition to the physical benefits of exercise, regular physical activity can also improve healthrelated quality of life, reduce poststroke fatigue, enhance social participation and help to restore independence. 3 However, evidence suggests that physical activity levels after stroke are low and further decline over time, 4 with stroke survivors often experiencing physical deconditioning and leading sedentary lifestyles. 3 This may be due to a wide range of factors both directly and indirectly related to stroke (e.g., prestroke physical inactivity and sedentary lifestyles, direct neurological effects of stroke which can reduce the muscle mass available for activation, presence of comorbid conditions) resulting in few people with stroke meeting recommended levels of physical activity. Therefore, finding effective ways to support people to become and remain active after stroke is critical.
Many people with stroke in the United Kingdom receive physiotherapy and occupational therapy at home following discharge from intensive in-patient-based rehabilitation. 5 When this community rehabilitation ends, some people feel there is a gap in support provided and still struggle to remain active. 6 The structured exercise programmes, guided by therapists, must be replaced by self-directed plans to continue exercising and increase physical activity, 7 which can be challenging for many. 8,9 Text message-based interventions have the potential to support and improve home-based, self-directed plans to continue exercising when community rehabilitation ends, or when community rehabilitation services are not available (e.g., in countries or communities where access to basic rehabilitation services may be lacking).
Research to date has shown promising effects on increasing physical activity in general populations, 10 however, interventions for people with stroke have yet to be fully tested. Pilot studies have reported the potential use of text message interventions for people with stroke, but their use has been limited and for diverse purposes: The STROKEWALK study delivered instructional text messages to promote regular walking and functional leg exercises over 3 months 11 ; the iVERVE intervention used text messages as part of a self-management programme to support goal attainment for recovery after stroke and in secondary prevention after stroke 12 ; and a text message reminder based intervention, which was part of a family-centred intervention, sought to support participation in daily activities. 13,14 These studies demonstrated that people with stroke can use text messaging as an intervention, although none specifically focused on using behaviour change strategies to support continuity with rehabilitation whilst facilitating the transition to active living after rehabilitation.
To provide continuity beyond formal rehabilitation and to help people with stroke to be physically active at the end of rehabilitation, we codesigned a novel text message intervention, the 'Keeping Active with Texting After Stroke' (KATS) intervention. The intervention and its development is described in detail elsewhere, 15 but briefly, we used a multistage iterative process to codevelop a theoretically informed text message intervention in collaboration with people with stroke, health professionals and experts in the field of stroke rehabilitation who were invited to join the study's Collaborative Working Group. 16 Key contributions from people with stroke and the Collaborative Working Group throughout the codevelopment process included: assessment of mobile phone use following stroke; identification of current needs and gaps; design of intervention goals; design of message contents and message delivery patterns; acceptability assessment; revision and refinement of messages. The intervention was designed to dovetail with community rehabilitation services after a stroke, to provide support and continuity at a time when many people with a stroke feel vulnerable. It was intended to enhance motivation, combat feelings of abandonment postrehabilitation and support the uptake and maintenance of physical activity and recovery-specific exercises. The text messages were designed to provide support for goal setting, planning and self-monitoring of physical/recovery activities and exercises. The intervention was theoretically informed by the Health Action Process Approach 17 and used a range of established behaviour change techniques 18 to increase motivation and provide support for people with stroke to be physically active.
An ongoing feasibility study was designed to test and refine the KATS intervention ready for evaluation in a future randomised controlled trial. We undertook a qualitative study to explore the views and experiences of the first cohort of participants to complete the KATS intervention as part of their participation in the intervention's feasibility study. This qualitative study was informed by Normalisation Process Theory (NPT) 19 -a sociological theory which explains the processes involved in implementing and/or making a new intervention work in practice to allow for the intervention to become 'normalised' or embedded in individuals/groups everyday practices. Four core constructs describe generative mechanisms that facilitate normalisation: coherence (work to make sense of an intervention), cognitive participation (work to engage with an intervention), collective action (work to enable an intervention to happen) and reflexive monitoring (work to appraise an intervention).
The aim of this study was to explore the views and experiences of the first cohort of people with stroke to complete the KATS intervention about the meaning, engagement, workability and worth of the intervention.

| METHODS
We undertook a qualitative study, theoretically informed by NPT, using semi-structured telephone interviews with people with stroke from two Health Boards in Scotland. Ethical approval was granted by the North of Scotland Research Ethics Service (21/NS/0028).
The KATS intervention 15 comprised 95 text messages delivered to participants over a period of 12 weeks. Participants received at least one message every day. The first week was used to foster interest and engagement. Messages then followed a sequence to address and illustrate the process of behaviour change to increase physical activity. Messages used conversational, informal language to encourage engagement. Participants were advised that, while they were welcome to respond to any of the text messages, the KATS messaging system did not allow for the research team to reply to any of their responses. Some messages included pseudonymised quotes and examples from people with stroke who had participated in the intervention development process, and from participants in our previous studies. 20,21 These messages modelled behaviours and provided encouragement. Some messages were personalised to include participants' names. Text messages were delivered by an automated computer system which was programmed to send the messages to participants' mobile phones in a predetermined sequence. The software tool for delivery was developed by the Health Informatics Centre at the University of Dundee (C. J.).
Participants were provided with a calendar (to facilitate recording of daily activities and reflection on progress) and a handbook (to reinforce key components of the intervention, and to provide information and signposting to online resources offering exercises for people who have had a stroke). At the end of the 12-week intervention participants received a £20 gift voucher. Data collection took place over two phases between July and November 2021, using semi-structured telephone interviews, with each participant being interviewed twice: First, at 6 weeks postrecruitment (halfway through intervention delivery) and again at the end of the KATS intervention at Week 12. This was to enable the exploration of experiences of the intervention 'in use' alongside perceptions at the end of the intervention cycle. The choice of conducting telephone interviews, rather than using virtual platforms, was chosen to minimise the potential impact of digital literacy, or stroke-related problems which can affect the ability to use digital technologies, as barriers to participation. Interviews were conducted by a female nonclinical university-based researcher (L. I.) with extensive experience in intervention studies, including intervention development and feasibility testing studies using participatory and qualitative methods.
All participants from the first cohort to complete the KATS intervention as part of their participation in the intervention's feasibility study were invited and took part in this qualitative study. Recruitment for the feasibility study was undertaken in collaboration with staff from stroke rehabilitation services in two Health Boards in Scotland, who identified patients receiving rehabilitation and invited them to take part in the study. Interested patients signed an expression of interest form, which gave the research team permission to contact them when community rehabilitation was nearing completion. The research team did not have any prior relationship with the study participants. Potential participants were given general information about the research team and detailed information about the study and their potential participation at the point of being contacted by the research team. Informed consent was obtained using audio-recorded telephone conversations, as face-toface contact was not permitted due to COVID-19 pandemic restrictions.
A copy of the consent form signed and dated by the researcher was sent to the participant by email or post. Suitable times were arranged with all participants to collect data on participant characteristics before the start of the intervention. Participants were characterised by age, sex, time since stroke, whether they lived alone or not, sociodemographic category 22

(Scottish Index of Multiple Deprivation [SIMD]) and Modified
Rankin Scale 23 to provide an assessment of disability/dependence. Times for telephone interviews were agreed at 6 and 12 weeks after the start of the intervention. All telephone interviews were audio-recorded and transcribed. Interview topic guides were informed by NPT 19 in combination with topics/prompts suggested by the larger feasibility study's Collaborative Working Group. 16 Interview data were managed using NVivo software and analysed using thematic analysis 24 theoretically informed by NPT. 19 Data were

| RESULTS
A total of 24 interviews were conducted with 12 participants (Table 1). Each participant took part in two interviews, the first one halfway through intervention delivery (Week 6) and then again at the end of the KATS intervention (Week 12). Nine participants were male and three were female. Their ages ranged from 31 to 74 years (median 61 years). Three participants lived alone. There was representation across all sociodemographic categories of the SIMD. 22 The time since stroke ranged from 5 to 184 weeks (median: 57 weeks). Participants self-assessed their degree of disability/dependence using the Modified Rankin Scale, 23 with scores ranging between 1 and 4 (maximum 5) and the majority of participants scoring 3 on the scale.
Below we describe our main findings, organised around four overarching analytical themes, which were informed by the four core constructs of NPT (Table 2). Offering guidance that can be adapted to a range of activities.
Messages perceived as helpful and encouraging.

Reflexive monitoring
Appraising the worth of KATS: encouragement and friendliness.
Language and content perceived as accessible and friendly.
Abbreviations: KATS, Keeping Active with Texting After Stroke; NPT, Normalisation Process Theory. interview) The motivational aspects embedded in the KATS messages One key perceived benefit of the KATS intervention was its potential to address an unmet need experienced at the point of discharge from NHS-funded therapy, which some participants associated with feelings of abandonment. 25,26 This important finding has already informed some intervention refinements. To ensure that the initial sense of abandonment is not echoed at the end of the KATS intervention, two additional weeks will be added at the end of the original 12-week intervention cycle. During these two additional weeks, the frequency of message delivery will gradually decrease, and the focus will shift to preparing participants for the end of the intervention and maintaining engagement in activities.
The KATS intervention was perceived as both relevant and fitting in the context of participants' rehabilitation journeys. These perceptions were emphasised by the experiences during COVID-19 restrictions, which brought about an increased sense of isolation that the KATS intervention helped address. This finding also helped demonstrate another important finding, that is, that some participants' understandings and perceived importance of the intervention were primarily tied to the very idea of receiving a text message (feeling seen, counted and supported) and were less about the content and components of the KATS intervention, which did not seem to matter as much to them as not feeling forgotten. Future research will further examine this finding with a view to establish whether it was disproportionally salient in our data due to the increased sense of isolation brought about by COVID-19 restrictions.
Participants sense of feeling seen and supported is a valuable finding and a potentially important outcome for the KATS intervention to consider in a future trial. However, our analysis also showed that the more complex behavioural aims/components of the intervention were not relevant to all participants. Therefore, future research should investigate whether addressing participants' need for support following discharge might overshadow the full projected benefits of the intervention, that is, whether some participants' satisfaction with 'just receiving' text messages may also mean that they were not sufficiently receptive or motivated to invest thought and energy into changing their physical activity behaviour. If so, further intervention refinement may be warranted, and any future trials to evaluate the effectiveness of the intervention should reflect this in the set of primary and secondary outcomes to be measured.
These findings could help refine the intervention to support readiness for change in physical activity after stroke more comprehensively, although the KATS intervention was not designed to require a specific form or level of engagement from participants. The development of the intervention consciously adopted a nonprescriptive approach focusing on providing guidance, ideas and choices (rather than giving rigid instructions or directive messages) to allow for different forms and levels of engagement and tailoring of the intervention to individual circumstances, needs and preferences. This is an aspect that differentiates the KATS intervention from existing interventions in this area. [11][12][13][14] The range of ways to engage with the intervention were apparent in participants' experiences, from those more thoroughly committed to enact the full range of proposed intervention activities through to those making the intervention work for them by just engaging with general suggestions/signposting provided or linking their physical activity to meaningful everyday tasks. A nonprescriptive approach that acknowledges different forms of engagement and tailoring is in line with a person-centred rehabilitation model 27 and was an aspect of the intervention highly valued by participants, alongside other aspects of the messages' tone such as informality, humour and trivia.
One key aspect that shaped participants' positive experiences Our study has some limitations. The sampling strategy for this qualitative study had to rely on the feasibility study sampling and recruitment strategy, which meant that our sample was not as diverse and information rich as it would have been if we had been able to employ a purposeful sampling strategy. The feasibility study sample had limited variation on certain domains (e.g., only three participants were women; only three participants lived alone; and the Modified Rankin Score for nine participants was three). Future research should seek to recruit from a wider range of geographical areas and clinical settings to allow for a more robust and comprehensive qualitative sampling strategy, and a more in-depth exploration of any issues of relevance to those specific domains for which our sample was limited.