How to Best Develop and Deliver Generic Long-Term Condition Rehabilitation Programmes in Rural Settings: An Integrative Review

People living rurally frequently experience health disparities especially if living with a long-term condition (LTC) or multi-morbidity. Self-management support is a key component of LTC management and commonly included in rehabilitation programmes to enhance ability to self-manage health and encourage physical activity. Such programmes are however often condition focussed and despite evidence for their effectiveness, are not always feasible to deliver in rural settings. Generic programmes are arguably more optimal in the rural context and delivery can be face to face or remotely (via telehealth). The aim of this explorative integrative review was to collate and present international evidence for development, delivery, integration, and support of community-based, generic LTC group rehabilitation programmes delivered rurally in person, or remotely using telehealth. Electronic databases were systematically searched using MeSH terms and keywords. For inclusion, articles were screened for relevance to the aim, and practical information pertaining to the aim were extracted, charted, and organized deductively into themes of Development, Delivery, Integration, and Support. Within each theme, data were synthesized inductively into categories (Theory, Context, Interpersonal aspects, and Technology and Programme aspects). Fifty-five studies were included. Five studies contributed information about community based programmes delivered via the internet. Development was the only theme populated by information from all categories. The theme of Support was only populated with information from one category. Our review has drawn together a large body of diverse work. It has focused on finding practical information pertaining to the best ways to develop, deliver, integrate, and support a community-based generic rehabilitation programme for people living with long-term health conditions, delivered rurally and/or potentially via the internet. Practical suggestions were thematically organized into categories of theory, context, interpersonal aspects, and technology and programme aspects. While the findings of this review might appear simple and self-evident, they are perhaps difficult to enact in practice.


BACKGROUND
Long-term health conditions (LTCs) are any ongoing, long term or recurring health conditions (>6 months) (1). LTCs impact significantly on a person, their family and their wider community (2). Self-management support is a key component in the health care of people living with LTCs. Rehabilitation programmes are important in the management of LTCs and usually comprise of components of exercise and education, with a focus on selfmanagement support so that the person can learn to live and manage their condition (3). A systematic review identified the key features of LTC rehabilitation programmes as being of 4-8 weeks in length, and include education on symptom management, exercise, time to develop and embed self-management skills and self-efficacy, and led by health professional/s together with lay or peer leaders (4). In previous work, we identified that the viability and sustainability of rehabilitation programmes may be contingent on a "closer to home" generic approach catering for people with more than one long-term condition (5). Further, building relationships, not just between the healthcare providers and people attending with LTCs, but between both these groups and the wider community are vital to enable and maintain participation (5,6). These factors may become even more crucial when working in rural or remote communities to enable health equity and promote supported self-management in a wider context.
Health care for people living with LTCs, particularly in secondary care, has largely been driven by models relating to one condition (7), whereas the increase in prevalence of multimorbidity demands more complex models of care (8)(9)(10). In terms of rehabilitation programmes, LTCs are mostly dealt with as single conditions, for example a current large undertaking by the World Health Organization and Cochrane Rehabilitation is developing a "WHO Package of Rehabilitation Interventions" (11). This project is developing rehabilitation guidelines for 20 separate health conditions as opposed to grouping conditions together by functional outcomes (12). For many LTCs, one of the mainstays of management is exercise or physical activity. Whilst specific conditions have specific exercise guidelines [for example, cardiac rehabilitation, (13) pulmonary rehabilitation (14)], in reality, the optimal exercise regimen (i.e., exercise type, intensity, duration, and frequency) is remarkably similar across LTCs (3). The benefits of condition-specific rehabilitation include high evidence in improving exercise capacity, symptoms, health related quality of life, and reducing hospitalization (15,16). The challenges to condition specific rehabilitation include having available healthcare professional specialists to run it, sufficient class attendees, and the nonsensical approach of people living with multiple morbidity attending specific rehabilitation programmes for each condition they are diagnosed with. Many LTC rehabilitation classes are delivered from a secondary care (hospital) setting, and thus become, and are perceived to be, "medicalised" in nature (16,17). Conducting generic, as opposed to single condition, LTC rehabilitation programmes is an emerging concept.
Despite robust evidence for benefit of LTC rehabilitation (15), in many countries in the world attendance at rehabilitation programmes is hindered by many factors, and particularly in urban or remote areas (18)(19)(20)(21). Inequities in healthcare provision are compounded by distance from health services, reduced access to primary and specialist care clinicians, and reduced socioeconomic status and low health literacy (21)(22)(23). In New Zealand, rural towns have the lowest socioeconomic status, highest proportion of Māori, and the highest avoidable and amenable mortality rates. Telehealth (delivery of healthcare when patients and healthcare professionals are in separate locations) (24) may be a possibility for delivering healthcare. Delivering generic community-based rehabilitation programmes in rural areas in person or by using telehealth may provide more equitable access to services beneficial to improving the health and wellbeing of those living with long-term conditions, and to a population in need of such a service (25). However challenges remain to using both approaches and in particular telehealth (such as equitable access to the internet, cost of technology, security breaches, technological and software limitations, changes in patient expectations and engagement, difficulty in maintaining therapeutic relationship and reading non-verbal cues) (26,27).
Nevertheless, informal consultation suggests that the potential benefits of offering a generic programme in person or by using telehealth include (i) healthcare delivered closer to home to remove some of the barriers for consumers through using community facilities (church and community halls, local gyms) set up for in person or telehealth delivery of a proactive programme enabling people to take control of their own health and make healthy choices; and (ii) potential reduced requirements for acute care by keeping people fitter and independently living at home. Whilst Mulligan,Wilkinson (4) identified the components of a generic community rehabilitation programme for people with LTCs, these were not specific to a rural setting or indeed one delivered using telehealth. This explorative integrative review thus sought information pertaining to international practice in developing, delivering, sustaining, and supporting a community-based, generic LTC group rehabilitation programme delivered rurally in person or remotely using telehealth.

METHODS
As our review was exploratory, we employed an integrative review method as it has the broadest type of search remit, allowing for multiple study types and methodologies to be included in the review (28,29). The inclusion of such diverse literature provides the opportunity to gather a greater scope of articles and gain a deeper understanding of the topic to answer the research question more effectively.
An initial search of Google Scholar was undertaken to explore potential search terms relating to the research question. After discussion with a subject librarian and individual research team members, and exploration of OVID Medline, a table of potential search terms and their associated MeSH terms was developed (see Table 1). A methodical search (30) of Google Scholar, SCOPUS, TRIP, Cochrane, EBSCO (CINAHL), JBI, OVID (Medline, Embase, Emcare, Psychinfo) and SCiello was then undertaken (November-December 2020) using combinations of MeSH terms and keywords (as appropriate for each database). All searches used Boolean operators "AND" and "OR". Discussion with experts in the field and searching of relevant journals (such as Journal of Rural Health) were also undertaken to generate further potential articles. Reference lists of potential articles were not searched for further potential articles.
Potential articles were title and abstract screened for relevance to the research question and had to include key terms of "Chronic illness/disease/long-term conditions", "Adult", "Community, rural" and "Self-management" combined with terms of "Exercise", "Education", "Program, viability, acceptability", and "Telemedicine" as appropriate to the individual databases (see Table 1). Articles were not included if they discussed home-based interventions delivered to one person, were delivered in a hospital or outpatient setting, included children/young people, or were not written in English. Extracted data required relevance to the research question with a focus on practical information pertaining to themes of development, delivery, integration, and ongoing support of a community-based, generic rehabilitation programme for people living with long-term health conditions (irrespective of the type of condition) delivered rurally and/or potentially via the internet (see Table 2 for definitions of themes). The full article was read if it were unclear in the abstract if it were relevant to the research aim. One author (AW) was responsible for decisions around suitability of articles for inclusion. Given the nature of this explorative integrative review and expected capture of publications with diverse study designs, included studies were not quality appraised. Data analysis, undertaken by one author (AW), involved extracting information pertaining to author, year, country or paper methodology, aim, and "demographics" of the study, review or report. This information was tabulated into an overall summary of included studies (see Table 3). Data pertaining to practical information about how to develop, deliver, integrate, and support a remotely delivered programme was then extracted from included studies. Through discussion and consensus by two authors (AW, LH) this information was deductively organized into "themes" (development, delivery, integration and support). These themes were derived from the research question, which was informed by multiple collaborative conversations with community stakeholders and modified from the RE-AIM framework (54) Data within each "theme" was then inductively (84) synthesized, again by the two authors, into five categories.

RESULTS
The search resulted in 24,485 potential articles from which 55 studies were included in the review (Figure 1). A mixture of primary studies (n = 27), reviews (n = 18), theoretical papers (n = 6) and reports (n = 4) were included. Five studies related to community-based programmes delivered to a group remotely via the internet [ Table 3 also provides an indication of which studies contributed information to the themes.
From the inductive analysis, five categories were derived, theory, context, interpersonal aspects, technology, and programme aspects. Table 4 details the contribution of these categories to the themes. Development was the only theme populated by all categories. Only one category contributed to the theme of Support. The categories are summarized below.

Category 1: Theory
This category contributed information to two themes, "Development" and "Integration". A framework or theory should be used to both guide development of a rehabilitation programme and its implementation and maintenance (49, 50, 53-55, 64, 70). Use of a theoretical framework makes explicit what the health professional is addressing (66,70) and thus may also facilitate personal growth for participants (65). Recommended is to develop, with the end-users, goals and a well-defined, efficient (procedures and process, cost), inclusive and adaptable implementation plan (underpinned by an implementation theory) that includes a sustainability plan for the programme/initiative (32,34,37,39,42,43,47,49,67,73). These goals, plans and definitions of success can be identified upfront and need constant reviewing (49,73). In rural and remote settings, flexibility and creativity are important and need to be utilized in programme design and delivery (42,71). Focus on outcomes rather than outputs and identify and address barriers (16,32,33,44,73). Be cognisant of the fact that "one size does not fit all" (67) and that the community needs to want and own the programme or initiative (55,63,(67)(68)(69). Note however, that a community development approach is more time intensive (34,37,67,69).

Category 2: Context
The category of "Context" contributed information to the themes of "Development" and "Support". Context is important and collated local knowledge should drive selection of intervention and assessment (47). This necessitates local consultations to find out what people want, need, and prefer (37,43,50,69,80). Also, of importance is a readiness assessment, for example identification of attitudes to the programme components and intention or readiness to attend (42,73,83). A continuing process for identifying and addressing barriers needs to be developed (42,47,51,53,57,73,74,79,81). It is important to create an environment whereby attendees become active in managing their own requirements (43). Further, any data collected must be securely stored and privacy is maintained (41,43,82). Development of a plan for ongoing infrastructure investment (26,43,60) and staff training was emphasized.

Category 3: Interpersonal Aspects
Three themes, "Development, Delivery and Integration" had contributing information from the "Interpersonal aspects" category. Working together on "the project" is essential (33,34,39,69). Create an interactive environment (33) that facilitates development of relationship/social cohesiveness between the participants, spouses, family, and friends (33,40,41,55,56,67,71,72,76,77,80,82). Focus on grass roots engagement, identifying shared goals and outcomes, building local resources and networks (67). Ensure projects are community owned and driven, that leadership is representative and inclusive (67). This builds relationships and a collaborative environment that values the contribution of everyone (34). Be aware though that it takes time to learn to work together (53). Build capacity in individuals, groups, and other stakeholders (67). Attract influential members (67). People may need training and require payment for their time (35,68). Ensure programme is well supported by highly trained staff and volunteers (32,41,42,68,73). Encourage peer support by using male and female role models/lay leaders (4,59) and via discussion, sharing of stories within the group (33,81).

Category 4: Technology
The category of "Technology" contributed information to themes of "Development", "Delivery", and "Integration". Synthesized findings suggest programmes should use technology that is simple, easy to use, adaptable, compatible with existing systems and cost effective (58,73). Be cognisant of and action regulatory standards, ethics, privacy, security, and storage issues for any data collected (43,55,73,82). Consider use of tools, such as the Universal Design Survey, to assess IT needs/requirements of programme leaders and participants, and train people to use the technology (43). Use creative ways to assist attendees to remember session dates and times (74) and develop telehealth etiquette with them (59). Use innovative ways such as slides and videos to enhance group discussion (33). Plan for interruptions and disconnections to the video feed (35,81) and hearing issues for attendees (62,82). Consider where equipment (conferencing and exercise equipment) will be stored (41) and ensure room set up is easy for telehealth and exercise (41,81). Train the trainers in telehealth etiquette and equipment use, conduct practice teaching sessions (35,59), and prior to sessions provide a reminder session to review procedures (59) Embed regular monitoring and evaluation (M&E) into all aspects of the programme (49,66,73,83). Involve the team in evaluation and communicate M&E information in multiple ways to stakeholders (49). Link any data collection with existing activities and processes (49,73).

Category 5: Programme Aspects
Two Themes, "Development" and "Delivery" had information derived from Category 5: "Programme aspects". Include/invite/involve people (end-users) in development (33,34,39,69) and provide/create a manual for participants and leaders (46). Address health literacy requirements (4,55) through use of an ehealth literacy framework (38,78). Consider use of clinically meaningful assessment and evaluation measures (4), and collection of attendance rates, cost effectiveness (61) and other pertinent data. Include exercise with clear guidance. Advertise the programme in a variety of ways (31,46), understand and address barriers to attending the programme (42, 47, 51-53, 57, 73, 74, 79, 81), and provide flexibility in programme delivery (e.g., times and places)  (35,42,51,55,64,79,85). Need to consider the class size and instructor-to-participant ratio (75) and who will attend, including the minimum level of walking ability, if including physical activity (75). Programme length is recommended to be 4-8 weeks, and use lay and peer led (4) "buddy coaches" with teaching skills to work with the attendees (46).

DISCUSSION
This integrative review explored literature for international evidence for developing, delivering, sustaining, and supporting a rural or internet delivered, community-wide, generic longterm conditions rehabilitation programme. While the review has several potential limitations (its explorative nature and broad approach, lack of quality appraisal of included studies, and an inherent risk of bias through one author working on inclusion of studies and data extraction), the review nevertheless provides a practical, important and timely contribution to the wider literature. Information gleaned and synthesized from the included studies suggest practical, fundamental points for consideration and were organized into categories of theory, context, interpersonal aspects, technology and programme aspects. The practical implications arising from our findings are summarized in Box 1.
When creating, delivering, sustaining, and supporting a generic rehabilitation programme, the findings from this review suggest the programme needs to be underpinned by "theory." Such theory is often derived from the field of implementation science (86). Davidoff, Dixon-Woods (87) suggest that while the word "theory" might be an abstract or irrelevant academic term to some, they contend that all people "find and use reasons-and thus theorize" (p. 229) daily. They propose the challenge is to "make explicit the informal and formal theories" (p. 230) people use because this may highlight assumptions, weaknesses, or contradictions in the proposed intervention programme's hypothesis, and expose any lack of consensus among the team (87). Use of what is termed a "small theory" or "programme theory" provides a framework for outlining programme components, expected outcomes and their assessment methods (87). Additionally the theory assists to make explicit and clear the assumptions and rationale linking "processes and inputs to outcomes . . . and conditions (or context) necessary for effectiveness" (p. 230) (87). Many people skip working out the programme theory and rush to implementation, thus limiting "learning that can inform planning of future interventions" (p. 232) (87). Choosing a theory may not be that straight forward. Lynch, Mudge (88) and Nilsen (86) in their debate papers provide useful summary for understanding available theories (current at time of publication of their papers), and a starting point and pragmatic guide for selection of "theory/ies" to underpin programmes/interventions. This review highlighted the importance of interpersonal factors for developing, delivering, sustaining, and supporting a programme. Working together with the people to whom it matters on programme development requires time to build relationships, talk, acknowledge and share power, reflect, and return repeatedly to these processes as the programme is developed, delivered, and evaluated (89,90). Time that is often not always available in the research arena because of constraints applied by funders and commissioners, or even because of a difference in world views between team members and community members (a biomedical v a biopsycho-social viewpoint). While time may not be "available, " relationships are integral to care and healing processes (91). Development and maintenance of meaningful relationships with other people is acknowledged to lead to improvement in wellbeing and health (92)(93)(94)(95). The concept of relationshipcentered care, argued to be the founding principle for all healthcare provision (91), may provide a framework for understanding the interrelated relationships necessary when working on programme development together with people to whom it matters.
Linked with the importance of developing meaningful relationships and working together with stakeholders (individuals, groups, communities, policy makers) are issues of pertaining to the context, particularly of valuing local knowledge about what is wanted, needed, and preferred. For developers, there are many ways to approach this depending on the philosophical and methodological viewpoint. For example, in included studies where the programme developers have already defined the topic of interest, to a study where the developers join with a community of stakeholders, and the community discuss what needs to be explored (96) (using a Participatory Action Research or co-design approach). Such stakeholder involvement can range from defining the issue/s, developing the programme, through to contributing as a participant, or interest only in the outcomes of the programme development project (97). Boaz,Hanney (97) suggest the literature assessing the impact of stakeholder engagement is limited but an increasing area of interest. They put forward three design principles for stakeholder engagement of organizational, values, and practices (with supporting literature) for developers to consider when thinking about stakeholder engagement and promoting impact of project development (97).
The idea of assessing readiness for change/engagement by people, communities, and organizations would also seem BOX 1 | Summary of practical implications arising from ndings of the review.
• Co-development with community end-users should drive intervention and assessment choices and thereby facilitate local ownership of the programme. • Building local resources, networks, capacity and leadership that is representative and inclusive is important. • Ensure flexible programme design and delivery. • Place importance on relationships, social cohesiveness and peer support between attendees, partners, family, and friends, and on highly trained staff and volunteers. • Adopt simple, cost effective technology that is easy to use, adaptable and compatible with existing systems. • Assess information technology needs of programme leaders and participants, and train people to use the technology. • Address health literacy requirements. • Be cognizant of and action regulatory and ethical standards for data collected, plan for interruptions to the video feed, and for hearing issues for attendees. • Advertise the programme widely and work to understand and address barriers to attendance.
useful. Yet terminology used in the area is confused, and there is no gold standard assessment available as instruments available are tailored to specific contexts and/or interventions (98). Miake-Lye, Delevan (98), in their systematic review of organizational readiness assessments mapped to the Consolidated Framework for Implementation Research (CFIR), suggest the seven most frequent CFIR constructs identified (readiness for implementation, implementation climate, other personal attributes, structural characteristics, networks and communications, self-efficacy, and culture) could provide something to consider when developing or tailoring a readiness assessment. Miake-Lye, Delevan (98) made only minor amendments to classify items, suggesting readiness for change is captured in the CFIR, with one addition relating to teams. It seems using a broad framework such as the CFIR may be another useful tool for programme developers.
The main findings from the review regarding "Technology" highlight attention to the principles of "KISS" (keep it simple stupid) (99) and Universal Design (100,101), and integration of health and eHealth literacy concepts (38) across all phases of implementation of the programme to facilitate access to the programme for a wider range of people. Additionally, the importance of embedding monitoring and evaluation processes within all phases of programme "development" seems pertinent to assess effectiveness of an intervention.
Many of the practical tips embedded in "Programme aspects" reiterate the importance of attending to the interpersonal factors discussed above. Further findings highlighted the importance of addressing health literacy requirements of participants. Health literacy is important and much has been written about it (78,(102)(103)(104)(105). However, "health literacy" is greater than individual competencies alone, it also includes community, services, and health system literacy capacities (106) and all these areas need to be considered and addressed when developing a programme or intervention.

CONCLUSION
Our review has drawn together a large body of diverse work. It has focused on finding practical information pertaining to the best ways to develop, deliver, integrate, and support a community-based group generic rehabilitation programme for people living with long-term health conditions, delivered rurally in person and/or potentially via the internet. Practical suggestions were thematically organized into categories of "theory", "context", "interpersonal aspects", and "technology" and "programme aspects". Box 1 provides a summary of the practical implications derived from the review. While the findings of this review might appear simple and self-evident, they may be difficult to enact in practice.