Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Social prescribing for people with complex needs: a realist evaluation

Social prescribing for people with complex needs: a realist evaluation Background: Social Prescribing is increasingly popular, and several evaluations have shown positive results. How‑ ever, Social Prescribing is an umbrella term that covers many different interventions. We aimed to test, develop and refine a programme theory explaining the underlying mechanisms operating in Social Prescribing to better enhance its effectiveness by allowing it to be targeted to those who will benefit most, when they will benefit most. Methods: We conducted a realist evaluation of a large Social Prescribing organisation in the North of England. Thirty‑five interviews were conducted with stakeholders (clients attending Social Prescribing, Social Prescribing staff and general practice staff ). Through an iterative process of analysis, a series of context ‑mechanism‑ outcome con‑ figurations were developed, refined and retested at a workshop of 15 stakeholders. The initial programme theory was refined, retested and ‘applied’ to wider theory. Results: Social Prescribing in this organisation was found to be only superficially similar to collaborative care. A com‑ plex web of contexts, mechanisms and outcomes for its clients are described. Key elements influencing outcomes described by stakeholders included social isolation and wider determinants of health; poor interagency communi‑ cation for people with multiple needs. Successful Social Prescribing requires a non‑stigmatising environment and person‑ centred care, and shares many features described by the asset‑based theory of Salutogenesis. Conclusions: The Social Prescribing model studied is holistic and person‑ centred and as such enables those with a weak sense of coherence to strengthen this, access resistance resources, and move in a health promoting or saluto‑ genic direction. Keywords: Social Prescribing, Salutogenesis, Realist evaluation, Co‑morbidity, Multi‑morbidity, Depression Background typically involve a named professional who supports the The popularity and prominence of social prescribing (SP) person and helps link services and agencies involved in is growing. In October 2018, the UK Government com- their care [4, 5]. Asset-based approaches seek to posi- mitted to investment in SP as part of its Loneliness Strat- tively to mobilise the assets, capacities or resources egy [1]. The National Health Service Long Term Plan [2] available to individuals and communities which enable also highlights a key role for SP in the provision of health them to gain control over their lives and circumstances and social care in the UK. SP is not one intervention, it [6]. NHS England define personalised care (or person- is a pathway [3]; it is an umbrella term that encompasses centred care) as people having choice and control over interventions that are asset-based, person-centred and the way their care is planned and delivered based on what matters to them and their individual strengths and needs [7]. Most SP interventions aim to support indi- *Correspondence: e.f.wood@sheffield.ac.uk 1 viduals to have greater control over their own health School of Health and Related Research, The University of Sheffield, Regent Court, 30 Regent Street, Sheffield S1 4DA, UK [8], for instance through exercise or benefit advice pro - Full list of author information is available at the end of the article grammes [9]. While evidence for social prescribing is © The Author(s) 2021. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creat iveco mmons .org/licen ses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creat iveco mmons .org/publi cdoma in/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Wood et al. BMC Fam Pract (2021) 22:53 Page 2 of 12 broadly supportive [10], rigorous studies of effective - and ‘facilitators’ [16] to uncover underlying mechanisms ness and cost-effectiveness remains scarce [9, 11]. Previ- and understand how, when and why they are activated. ous reviews found indications that SP may be effective in RE requires researchers to have an initial theory that can improving health and well-being and reducing healthcare be challenged and refined during the study. Mid-range usage but included studies were small and many had low theories (which may or may not be novel) can be applied methodological quality, reducing confidence in the out - to the findings to improve generalisability of case study comes [9, 11] meaning the existing evidence base for SP findings [17, 18]. as a means of improving health outcomes is poor. By con- RE recognises that wider context influences outcomes trast, there is considerable support for SP at policy and and specifically how individual actors respond to dif - commissioning levels. ferent parts of the intervention at different times. To The link worker role in SP is considered key for peo - describe this complexity, statements describing path- ple with multiple long-term health conditions where care ways between intervention or individual Context (C), and can often be fragmented [7]. This resonates with collab - underlying Mechanisms (M) that subsequently shape pat- orative care models [12] which have four elements [13]: terns of outcomes (O) are created (CMO configurations) multi-professional approach (with one acting as a case [15]. CMO configurations aim to describe why a person coordinator); enhanced communication between profes- (or case) responds to an intervention in a certain way and sionals; structured management plan; and scheduled fol- how this can change depending on circumstances. CMOs low up. We therefore hypothesised that the underlying can be linear or more complex [15, 19]. mechanism for SP might be consistent with collaborative We focused on one large voluntary sector, community care [14]. anchor organisation providing SP to an inner-city area of This research set out to elucidate the mechanisms that high socioeconomic deprivation, predominantly deprived facilitate engagement and positive outcomes with SP white working class council estates. Clients often have intervention among people with multiple health condi- multiple health conditions including co-existing physi- tions and social needs. We sought to develop, test and cal and mental health concerns. Most have social needs refine the initial programme theory (that the benefits associated with housing, benefits and lack of support associated with SP derive from enacting collaborative networks. Many are isolated. Unlike some areas, there care) using stakeholder experiences. was no central referral point; general practices and other referrers (such as housing officers) referred to one or Methods more of many community anchor organisations that pro- Realist evaluation vide services in their local area (Table  1 provides details Realist evaluation (RE) looks at generative causation [15], of the setting). In some parts of the country, general prac- a key strength is modelling complexity. Adopting a realist tices have begun employing link workers but in this area, approach enabled us to go beyond identifying ‘barriers’ all social prescribing is provided by these third sector Table 1 Details of the study setting Organisation Type: Community Anchor Organisation ( Voluntary sector) Location: Inner‑ city area of high socioeconomic deprivation in a city in the north of England Referral type: General Practitioners and other services refer direct to the organisation. Self‑referral is also possible. Triage worker signposts to the most appropriate service, based on the client’s personal goals rather than the doctor’s determination of the problem. In 2018, there was a total of 1372 referrals; 813 from General Practitioners, 207 self‑referrals, 315 from other sources (e.g. housing, community mental health teams) and 37 whose refer ‑ ral status was not recorded. The number of clients with a mental health condition is not recorded; however, in 2018, 56 clients enrolled on the coping and self‑management programme and 59 enrolled on the emotional well‑being programme Services provided: Advice and services around health, employment and training. For the purposes of this study, we only considered the health sec‑ tion: this includes health training (e.g. weight loss or health eating advice, alcohol or cigarette reduction and exercise advice), social café’s, benefits and housing advice, and volunteer work. There is no set pathway through the service. Clients can access different services at different times in what ‑ ever order meets their needs. The service has no time limit Staff team: Paid health trainers, health activity workers and advocacy workers (primarily giving benefits and housing advice) and unpaid volunteers. Any of the paid workers could be a link worker, this would be decided based on client goals. Clients with predominantly health goals would have a health trainer as a link worker. Once these goals were met, the client may be referred to other colleagues if needed, for example for benefits advice. The person acting as link worker would change Care pathway: New clients are screened by the in‑house triage service to ensure the client is seen by the right part of the service to meet their goals. It is possible to move from one service to another or see multiple workers at one time depending on the nature of the client’s personal goals. The service also includes social café’s, which can take referrals or clients can drop in. You can attend the social café concurrently with other one to one services W ood et al. BMC Fam Pract (2021) 22:53 Page 3 of 12 organisations, link workers are not provided by primary Phase one tested the IPT through data collected in inter- care networks. views with people working for the organisation, those The initial programme theory (IPT) was informed by receiving SP and external referring organisations, to test the wider research team and relevant literature reviews and refine this theory. In Phase two, we developed CMO of collaborative care [20], realist methods [18] and social statements which refuted the IPT, leading to modification prescribing [9, 10, 21]. It was formed over multiple meet- of the theory. Phase three focused on applying and devel- ings between researchers and stakeholders from the host oping wider theory to synthesise the CMOs into a mid- site as well as another SP organisation in the same city, range theory. A mid-range theory is a theory that lies to further understand why and how SP works for certain between working hypotheses, contains testable predic- populations. The IPT to be tested was that collaborative tions and evolves as efforts are made to develop a unified care and SP are overlapping concepts (Fig. 1). We hypoth- theory. It was developed and related, where appropriate, esised that SP organisations are potentially effective in to existing theories [17, 18]. The mid-range theory pro - improving self-management for adults with co-existing vides a level of abstraction to the analysis and therefore a physical and mental health conditions because they work generalisability beyond the immediate context [17]. on a collaborative care model [12]. For the purpose of this study, intervention type was disaggregated from con- Data collection text in order to differentiate between interventions in one Phase 1: interviews contextual setting. We aimed to interview stakeholders with different points The study was iterative in nature and conducted over of view and experiences of SP. Clients were approached three phases. Stakeholders from the host community if they met inclusion criteria of having both physical and anchor organisation were included in development and mental health issues. Recruitment was via convenience design meetings from the start. This helped ensure that sampling and took place at different services within the the research was appropriate for the local model of SP organisation: health training, advocacy, volunteer devel- delivery and that it would provide insights of value to opment and social cafes. At the social cafes one of the both community anchor organisation s and researchers. researchers gave a short talk to attendees to introduce Fig. 1 The initial programme theory Wood et al. BMC Fam Pract (2021) 22:53 Page 4 of 12 the research and who we were looking for as partici- interviewed to get a wider range of opinions. As before, pants. Attendees were then asked to approach either clients self-identified as having co-morbid physical and the researcher or café staff if they wished to participate. mental health issues. Referrers came from different sec - For all other services, the staff member suggested the tors including primary care, housing, and community research to their client on a one to one basis. Everyone mental health teams and had referred at least one client was told participation was optional. to the service. As we had interviewed almost all the host All staff from the host SP organisation were invited to SP staff members, most staff members at the workshop interview via emails detailing the research. Doctors and had participated in interviews. practice nurses at local general practices were invited to Initial results from the interviews were reported at the interview as SP referrers. All practices associated with workshop, to gauge attendees’ views on our interpreta- the host organisation were approached. tions and areas where we considered there may be miss- The interview schedule evolved via a constant compar - ing data. Participants were asked to provide comments, ative method [22] such that each interview was informed criticisms and feedback on those interpretations. These by the ones that had taken place before it. The focus of were then incorporated into final analyses. the interviews was to present components of the IPT to The workshop lasted for three hours and clients the interviewee, on flash cards (face-to-face interviews received a store voucher for attending. for staff and clients) or verbal hypothesis (telephone interviews for referrers) for them to comment on with a Analysis view to providing theory refinement [23]. The interview Realist analysis takes an iterative approach, moving structure and questions were adapted from the RAME- between different sources of data and using deductive SES-II project [24] and flash cards were adapted accord - and inductive reasoning [25] (see Table 2). ing to who was being interviewed. The IPT was interrogated using data collected at inter - Interviews were audio recorded and lasted between view, by testing our hypothesis that SP is consistent with 15 and 60  min. Participants signed a consent form collaborative care [26]. This was done by analysing expe - before the interview commenced. Most interviews riences of three stakeholder subgroups (clients, staff and took place face-to-face in the host organisation’s prem- referrers) to see if their views corroborated or differed ises, though referrer interviews were conducted by from the IPT. A deductive approach was used. Analysis telephone as this was more convenient for clinicians. consisted of applying data to the four key elements of Interviews were conducted by EW and SO between collaborative care using a Framework approach. Themes February and April 2018. about what worked for whom, when and why that did and did not fit with the IPT were used to continually Phase 2: stakeholder workshop refine the model. Coding was undertaken by SO; EW and A workshop with key stakeholders [18] supported refine - JC independently coded 14 interviews to cross check the ment of the coding frame and analysis of emerging coding frame. NVivo software was used to assist in data themes. The results contributed to the emerging mid- management. range theory and CMO configurations. The developing themes became the premise for dis - Twenty participants were invited to the workshop. puting the IPT as the data were related to wellbeing not They were referrers into the service (e.g. National Health ‘health’ and ‘disease management’. These were developed Service staff), SP staff, or clients. Referrers and cli - into a series of CMO configurations by the wider research ents were only invited if they had not been previously team until agreement was reached. New configurations Table 2 Key elements of the Realist analysis process The realist methodology uses the following approaches judiciously and in combination: • Organizing and collating primary data and producing preliminary thematic summaries of these • Repeated writing and rewriting of fragments of the case study • Presenting, defending, and negotiating particular interpretations of actions and events both within the research team and also to the stakeholders themselves • Testing these interpretations by explicitly seeking disconfirming or contradictory data • Considering other interpretations that might account for the same findings • Using cross‑ case comparisons to determine how the same or submechanism plays out in different contexts and produces different outcomes, thereby allowing inferences about the generative causality of different contexts From Greenhalgh, T., Humphrey, C., MacFarlane, F., Bulter, C., & Pawson, R. (2009) W ood et al. BMC Fam Pract (2021) 22:53 Page 5 of 12 were then tested with data collected from the workshops of their experiences that would have aligned with col- in phase two. laborative care had it been a key underlying mechanism. Finally, we looked to extend generalisability beyond the They were then asked about what they felt was help - immediate context [17] by relating, where appropriate, ful about attending the SP service. When researching the newly created theory (mid-range theory) to existing whether an intervention ‘works’ it is important to define theories [17, 18]. Client-led interventions and asset-based its aims. One of the aims of SP is to improve health and approaches emerged as key to the mid-range theory. This well-being. However, other outcomes necessarily exist, is closely linked with salutogenesis [27, 28]. the service is client-led so clients will have their own Salutogenesis focuses on promoting health and well-being desired outcomes, which may or may not link directly as opposed to managing symptoms of disease. It further to improved health as defined by health services. The posits that life experiences shape one’s sense of coherence answers allowed us to refine the IPT (Fig. 2). (ability to comprehend a situation, find meaning and be The interventions in SP (Fig.  2), although superficially able to act). A strong sense of coherence aids in mobilising similar to collaborative care, were different in prac - resources (internal and external) for dealing with stress and tice and activated a wider range of mechanisms and helping a person move towards health rather than disease outcomes. [27, 29]. Salutogenic interventions are those which seek to The CMO configurations derived from the data strengthen a person’s ‘sense of coherence’ [29]. resulted in multiple interconnections, especially between mechanisms and outcomes, which did not Results fit [19] simple linear progressions such as ‘C + M = O’ Participants [15]. Many configurations reported multiple contexts The realist analysis was based on 35 interviews with cli - resulting in multiple mechanisms, which lead to an ents (n = 15), staff (n = 13), and referrers (n = 7) who outcome attained which then became a mechanism were all involved the health section of the SP community to achieve another outcome. For example, many cli- anchor organisation. All but one member of host SP staff ents were referred or self-referred to the organisation (who was on leave during recruitment) from the health because they were isolated (context). The social cafes section were interviewed in phase one. Staff backgrounds (intervention) facilitated making new social connec- varied and included health trainers, benefits/housing tions (mechanism) which reduced isolation (outcome). advisors and people who supported the organisation’s vol- Although this follows the CMO framework it does not unteers. The health section manager was also interviewed. represent the complexity found in the data, which were The stakeholder workshop had 15 delegates including more multifaceted than the above statement implies. seven staff, five referrers and three clients. Five staff at The familiar location of the cafés, which are set in local the workshop had been involved in the interviews. Two communities, was also a mechanism for increased senior managers attended who had not been invited to engagement, and therefore also reduced isolation. As interview. None of the referrers or clients at the work- did the increased confidence (Mechanism (M)) clients shop had been interviewed in a deliberate attempt to gained from trusting staff (M) who could work flexibly ensure we were getting a range of viewpoints. See Table 3 (M) and were not sited within the NHS (Context (C)) for participant information. but were still knowledgeable (M). Intermediate out- comes can, in turn become mechanisms for longer term Modification of the IPT outcomes. For example, increased engagement could The IPT was that SP works in a similar way to collabora - be an outcome in its own right but it is also a mecha- tive care. We asked participants about specific elements nism for improved mental and physical health. Figure 3 Table 3 Demographic details of the participants Interviews Referrers N = 6 (general practitioners, practice nurse, 2 male, 4 female) Staff N = 13 (1 manager, 4 health trainers, 3 advisors, 3 volunteer coordinators 2 triage, 6 males, 6 females, one preferred not to say) Clients N = 15 (12 clients, 3 clients who also volunteer, 5 males, 9 females, 1 preferred not to say) Focus group Referrers N = 5 (community mental health team, housing, social work) Staff N = 7 (2 managers, health trainers, advisors) Clients N = 3 (2 male and 1 female) Wood et al. BMC Fam Pract (2021) 22:53 Page 6 of 12 Fig. 2 The mid‑range theory developed after interviews with SP stakeholders shows an overview of the interweaving CMOs that we of the staff roles (M) and staff having time to listen* discovered. (M); resulting in improved Mental Health (O) and more Some of the configurations demonstrate the overlaps appropriate health service use* (O). and differences between the original and modified IPTs. Staff felt clients valued this approach as they would Points marked with a * are from the modified IPT but receive word of mouth self-referrals from friend and fam- not the original, unmarked points stem from both, sup- ily of existing clients. porting quotes are illustrated below. Person‑centred care When thinking about person centred interventions one Non‑stigmatising environments staff member spontaneously described their own idea of Adults with depression but no diagnosis/treatment* (C) the CMO configuration. are accessing SP services in community settings* (C), When someone is new to the service (C), it’s crucial this is perceived as less stigmatising (M) and intimidat- that we do it in an approached manner (M) that ing than clinical services* (M), resulting in increased we can do it at their speed, (M) feeling comfortable engagement (Outcome (O)) and improved mental about it (O), giving confidence that they can do it health (O). (M), and allowing that to flourish (O) and say ‘come SP [organisations] are there for people with low on, we can move forward’ (M). So, it’s empowered confidence (C) so they’re not going to look down on them (O), they’ve got to make that choice (M) and you (M). Client 1 they’ve got to make those decisions but it’s about Chatting to people (M), you know you’re not on being supportive isn’t it (M), to doing it. And that’s your own (M). You know you’re not the only person what I see my role, is supporting people and moving who’s had problems (C/M). Client 3 them on to next… every individual has structured management plan (O), speaks for itself. Every one’s This SP organisation offers a welcoming setting that different. It’s not my plan. It’s their plan (M). Staff 11 helps the client’s mental health but does not focus on it directly. This suggests that staff valued the flexibility of their Adults who are isolated* (C) possibly as a result of role working at the client’s pace, seeing it as integral to a bereavement* (C) with mental health issues (C) are offering person-centred care. ‘It’s not my plan, it’s their receiving personalised client centred management plans plan’ implies that their role is to facilitate and guide the in SP* (Intervention). This is can be provided on a flexible individual to choose rather than ‘intervening’ in the basis to meet their need* (M), partly due to the flexibility W ood et al. BMC Fam Pract (2021) 22:53 Page 7 of 12 Fig. 3 Diagram detailing some of the main CMOs and their interlinking nature classic model. Whilst many health care workers do have Clients value the peer support that the social cafes’ a degree of autonomy and flexibility, they are usually con - provide; reducing not only physical isolation but also strained in this. The SP workers here described a level of emotional isolation by introducing clients to people with working—supported by management—to meet the needs similar issues, who are able to support each other. This of clients however unusual, for example, aiding in house reduction in isolation was also felt to have an effect on clearance for a hoarder. the physical health of clients and ultimately their attend- ance at General practice clinics. For elderly widowed males who had been dependent on their partner for Social isolation essential life skills (cooking) (C) personal help with shop- Clients with the similar contexts attend social cafes: Have ping and guidance may increase confidence and motiva - a space/location to engage in participative social function tion (M) to eat better and lose weight (O). (meet people/peers/friends/shared experiences)* (M) are By regularly attending SP activities(C), clients often listened to (by peers)* (M), which improves/increases meet others who have been through similar situations social functioning (O) and reduces isolation* (O) and (C) it creates a social network(O/M), shared experience improves mental health (O). gives peer support (O), Reducing sense of isolation and It’s building my confidence up great (C/O). I’m mak - reliance on health services being the only place you can ing loads of friends (O). I mean, I’m in a craft group discuss your health(M/O). Leading to reduced primary but I don’t really do much crafting when I’m It’s health care use. (O). more chatting (M) and helping the others (M), so it’s lovely, and they’re just so friendly(M). Client 14 Wider determinants of health Instead of once a fortnight, they’re going to some- Wider determinants of health are a range of economic, where twice a week now (O) so there’s, there’s always social and environmental factors that directly and indi- something for them to do (M) and it brings them rectly affect people’s health [30]. Clients present with together (M),  I mean they say to me, things like oh, social issues* (C) contributing to mental health issues (C), if I didn’t come here I’d have nowhere else to go (O), staff in SP have skills to support social issues* (C). Clients I’d be sat, four walls (C), I don’t know what I’d do if I receive personalised management plans, the interventions didn’t have this group, and that type of thing. Staff 7 Wood et al. BMC Fam Pract (2021) 22:53 Page 8 of 12 are adaptable to this individual need* (M) with staff being Many SP organisations are in the voluntary sector flexible in how this support is offered/delivered* (M) and are isolated from statutory services. Data shar- Resulting in improvements in housing/finances/employ - ing is therefore problematic. Information is not passed ment* (O) and positive impacts on mental health (O). between organisations in a timely manner and even when it is, there is no interoperability in IT systems. The lack of We can even do a home visit (Intervention), because professional status of SP staff leads NHS and other staff even asking someone to come and see us here for the to be unsure about what information they can or cannot rfi st time is daunting (C). so I think with us, slot - share with them. ting in with them (M), I see it as like a jigsaw, so it just all slots in because they see us and we look at the barriers to health and put them steps in first Mid‑range theory and work through them with them (M), and then it’s Salutogenesis just giving them that bit of self-belief that they can Central to the model of salutogenesis is the concept of do something and show them how they can make a sense of coherence which is a ‘generalised, long last- small changes that that’s leads to bigger things so by ing way of viewing the world and one’s place in it’ [27]. us being there, they can then move on to volunteer- Although considered to be stable in adults, it can be ing (O) and then move on to employment advocacy altered particularly by radical changes. Additionally, (O), if they are on [employment and support allow- it influences whether or not an individual attempts to ance] they can help them sort out the benefits and change their situation [27]. People with strong sense of what have you but then they can refer back into us coherence, view the world as predictable and therefore again, to say well actually they are on ESA but they comprehensible. Antonovsky links this theory to those are looking at wanting to return to work but they of Bowlby (attachment) [31] and Seligman (learned help- have got no confidence (C), you know, so we sort of lessness) [32] while highlighting differences. The sense of can keep seeing them (M). Staff 9 coherence is considered to be a continuum from strong to weak, characterised by a normal distribution [27]. SP in this organisation contributed to health improve- Salutogenesis is an individual level explanation of ments by concerning themselves with the wider determi- health behaviours. Previous studies have shown a rela- nants of health, not just presenting health concerns. tionship between a strong sense of coherence and good perceived health, particularly mental health [33]. It also Poor interagency communication seeks to explain why some individuals do not respond to Clients present with complex social issues* (C). Clients health information from professionals. From a saluto- receive personalised management plans (Intervention). genic perspective, this should not be seen as an individual But poor shared systems with external organisations (M) failure but a failure of the service to provide understand- and external organisations perceiving SP staff as non- able information [33]. Salutogenesis also refers to the professional (M) resulted in difficulty with systematic ability to utilise resources (both external and internal) to sharing of information (O) and longer waits with more manage stressful situations. The ability to recognise and chance for clients to disengage (O). use these resources is important for sense of coherence. Clients reported that staff helped their understanding I think that would make a huge difference, because of their issues. They found health trainers to be motivat - if a [General Practitioner] was to log in and see that ing and knowing they had someone to turn to helped they’re working with social prescribers and they’re them to feel that their situation was more manageable. going to groups and this has happened and that has The location of SP away from statutory health services happened, then we can work and keep encouraging and the ability of staff to work differently to health staff them to go, you know those sort of things. General (with difference in time and role flexibility) aided accept - Practitioner 2 ability of the service. However, the fact that it was a Very often there’ll be interruptions in claims, bene- non-statutory service did cause problems for informa- fits will get suspended. If I could talk to [the Council] tion sharing. In this way the client’s comprehension and at that time when the client’s here I could stop that understanding was often improved but this is not always happening, whereas now… the letters’ll be god knows the case for SP and primary care staff. SP can empower where… and in the meantime you know the benefit people to utilise their resources and develop new ones. might get suspended… If I could talk to them I could Resources can be internal, such as confidence or self- solve a lot of problems because I can put in a nut- esteem, or external, such as friends or community who shell what the client might struggle to sort of want to provide advice, support, or bolster internal resources. put across. Staff 3 W ood et al. BMC Fam Pract (2021) 22:53 Page 9 of 12 SP can be considered an external resource but there are invalidate the efforts made in providing these services. many facets to this, due to the different models of SP For example, people in financial hardship may struggle that exist. However, examples include, the link worker as to access a service that requires bus travel to attend. One source of support in a crisis or as a way to access other of the things participants liked about the organisation sources of support, and the community groups provided studied in this research was that it provided services in as part of SP that offer support, companionship and several local community locations, meaning most people advice. could walk to services, knew the area and felt comfort- able there. Although sense of coherence is stable, it can be Discussion changed, but to do so in a positive way is slow and takes Summary ‘hard work’, such as undertaking psychotherapy [27]. SP in this setting was a not collaborative healthcare inter- More recent research shows that sense of coherence can vention but rather a client led, person-centred, asset- change across the life course and that many prerequisites based service addressing the wider social determinants for strengthening sense of coherence may be provided of health including co-morbid conditions. In doing so, it by or mediated by the community [36]. The version of worked on salutogenic principles, providing new resist- SP that was delivered by the provider organisation, at ance resources, helping people access existing ones and the time of our research, and the ways that it was expe- even strengthening low sense of coherence. rienced by clients, was consistent with the theory of salu- Personalised care is at the heart of social prescribing, togenesis. This research and that of Payne and colleagues staff refer to goal planning in SP as ‘ their plan not mine’; [5] found that SP facilitated change in perceptions of per- SP goals are set by the clients not staff. Even when clients sonal assets through personal and social development. are referred by health care staff for specific purposes, if Meaning that clients became more aware of what assets that is not what the client wishes to address first (or at all) were available to them and more able to access them. then this is not a condition of entring the service. Whilst This is consistent with theory of change for the sense of this personalised approach clearly has its merits, and the coherence and therefore supports Health Education Eng- repeated return to SP organisations and word of mouth land’s suggestion that the theoretical base for SP is Salu- recommendations show it is a popular policy, it may dis- togenesis [4]. advantage those whose sense of coherence is so low that When considering what aspects of SP works for whom they cannot recognise or articulate the need to change. and in what circumstances, context is important. Access However, given the broad range of activities that is avail- to a supportive community can strengthen a person’s able via SP this should be less of a disadvantage than in sense of coherence through the life course [36]. People traditional health care settings. It also suggests a signifi - with a low sense of coherence may struggle to access cant departure from the collaborative framework of col- these resources without help and these people can par- laborative care. The client is not being consulted by the ticularly benefit from SP interventions, although the spe - healthcare professional; they are leading the direction of cific intervention needs to be determined on a case by the intervention. This has the potential to have an effect case basis and remain person centred. on the level of dependancy people have on the service. A recent realist review [3] reported that there are three However, although many of the interviews with clients stages which contribute to pathway success in social pre- mentioned dependecy it was in relation to others. This scribing. Enrolment, engagement and adherence and suggests participants were aware of it but that it might be the link workers are key to avoiding disruption of the difficult to recongise or admit to, it is however, an issue process. Our findings are consistent with these points. that is both live and sensitive. Link workers have a key role in ensuring people are sup- ported to attend and understand what social prescribing is. Similarly, another recent realist review of social pre- Comparison with previous studies scribing had two main concepts, creating and sustaining Pelikan concluded that salutogenic thinking has good buy-in, and establishing and maintaining connections potential to be applied in health settings [34]. Specifically [37]. The first of these was not a major concern in our in health promoting interventions, structures and cul- study as referrers were enthusiastic about social prescrib- tures and improving a person’s sense of coherance could ing and keen to work with the organisation. Clients too, be an explicit goal of chronic disease management [34]. would regularly report recommending the organisation The wider determinants of health must not be underes - to friends and neighbours. The caveat here is that both timated when planning SP delivery [35], they are part of were self-selecting samples and we did not (and could the context to our realist theory and may constrain the not) gauge how representative they were. The second, effects of the SP intervention. However, this does not Wood et al. BMC Fam Pract (2021) 22:53 Page 10 of 12 maintaining connections, was more of an issue, both in social prescribing referrals. It is important to note that terms of logistics around data sharing and governance, those with the weakest sense of coherence, who might but also regarding remit and scope of the social prescrib- be expected to benefit most from social prescribing, ing provider and its services. These changed over time may lack sufficient resources to access the service and and some family doctors reported that they were unsure may need significant help (from referrers and social pre - about which service to refer too. This was the reason for scribing providers). A key barrier to this is difficulty in developing the triage role, so that referrers had a sin- information across organisational boundaries, and those gle point of contact to improve relations. Our findings who commission SP services are perhaps best placed to complement Tierney et  al.’s conclusions that social pre- address this. scribing can, through developing wider social networks, We found that flexibility on the part of link workers was reduce isolation, increase meaning and activity and give critical, and they require sufficient time and resources to people the confidence to manage their own health. Our work this way. Link workers in our study recognised the theory development diverged from that of Tierney et al., need to support engagement and that regular attend- whose primary focus was on the setting up of services ance was key for reducing isolation. Processes or policies whereas we studied a mature service that had existed regarding discharge based on time spent in the service for many years. This allowed us to look beyond logistical were therefore felt by many to be counter-productive, considerations into deeper theory behind social prescrib- although the consequences for service capacity was also ing in action. recognised. Commissioners and service providers need to find the correct balance for their chosen aims as differ - ent organisations may need to work in different ways to Strengths and limitations satisfy budgets and local need. SP providers are very varied. The service we studied was provided by a large organisation was large and comprised multiple interventions. As the local authority did not Implications for research have a central SP referral point the organisation had to Theory development moved from a large scale IPT of liaise with referrers directly, attempting to find local solu - SP, then modified to the SP organisation in the study tions to a city-wide problem. This situation is not the then widened out again for the mid-range theory. The same in other cities or for small single intervention SP mid-range theory therefore should be tested in addi- groups. However, the central finding of the importance of tional SP organisations to ensure it is not context being client centred and of acknowledging wider deter- specific. minants of health remain relevant to a wide variety of dif- ferent SP models of implementation. This study focussed on an organisation that predomi - Conclusions nantly served deprived white working-class council Although superficially similar, social prescribing does not estates. There were limited opportunities to capture the appear to operate as a type of collaborative care. Collabo- voices of people from minority ethnic and non-white rative care does represent a move towards more holistic British backgrounds. thinking within healthcare settings, but social prescrib- ing takes this concept even further. It is holistic and per- Implications for practitioners and commissioners son-centred and as such may enable those with a weak We sought a better understanding of the mechanisms of sense of coherence to strengthen this, access resistance action for social prescribing, to provide greater clarity resources and move in a health promoting or salutogenic about who can be expected to benefit and why, and there - direction. fore who to prioritise in referral practices. Although in- depth, our study was based on a single social prescribing Abbreviations provider in one city and our findings need to be viewed SP: Social Prescribing; IPT: Initial Programme Theory; CMO: Context‑mecha‑ cautiously. nism‑ outcome; C: Context; M: Mechanism; O: Outcome. We found that people with long-term social difficul - Acknowledgements ties who struggle with chronic health conditions because The authors would like to thank all of the social prescribing staff, clients and they have limited support networks may benefit the most health services staff who participated in our research and offered their valu‑ able insights. from social prescribing. Our results support the view that this is because building support networks and develop- Authors’ contributions ing coping mechanisms enables better engagement with All people entitled to be authors have been included. EW drafted the paper, collected and analysed the data and contributed to the survey and protocol sources of treatment. We therefore recommend that development. SO and JC assisted with the analysis. SJF gave significant these individuals should be considered a priority for W ood et al. BMC Fam Pract (2021) 22:53 Page 11 of 12 guidance on the Realist method. SO, SJF and SW contributed to protocol com/uploa ds/docum ents/17101 ‑asset Based Appro aches toHea lthIm development. All authors read and approved the final manuscript.prove mentB riefi ng.pdf 7. NHS England. Social prescribing and community‑based support Sum‑ Funding mary guide. London; 2019 [cited 2019 May 16]. Available from: https :// This article presents independent research funded by Sheffield Hospitals www.engla nd.nhs.uk/wp‑conte nt/uploa ds/2019/01/socia l‑presc ribin Charity and the National Institute for Health Research Collaboration for g‑commu nity‑based ‑suppo rt‑summa ry‑guide .pdf Leadership in Applied Health Research and Care Yorkshire and Humber 8. The King’s Fund. What is social prescribing? 2017 [cited 2017 Aug (NIHR CLAHRC YH). The views and opinions expressed are those of the 25]. Available from: https ://www.kings fund.org.uk/publi catio ns/socia authors, and not necessarily those of Sheffield Hospitals Charity or the NIHR l‑presc ribin g?gclid =EAIaI QobCh MI052 5kffx 1QIVi BXTCh 3_OAHhE or the Department of Health. The funding bodies had no role in the design AAYAS AAEgL ecPD_BwE of the study, collection, analysis or interpretation of the data, or in writing 9. Bickerdike L, Booth A, Wilson PM, Farley K, Wright K. Social prescribing: the manuscript. less rhetoric and more reality. A systematic review of the evidence. BMJ Open. 2017 Apr 1 [cited 2018 Dec 20];7(4):e013384. Available from: Availability of data and materialshttp://www.ncbi.nlm.nih.gov/pubme d/28389 486 Data from this study may be available upon reasonable request to the authors. 10. Polley MJ, Pilkington K. A review of the evidence assessing impact of Individual participant data that underlie the results reported in this article will social prescribing on healthcare demand and cost implications. Univer‑ only be available after deidentification (text, tables, figures and appendices) sity of Westminster; 2017 [cited 2019 Jan 11]. Available from: https :// if the requestor has Research Ethics Committee approval to work with data westm inste rrese arch.westm inste r.ac.uk/item/q1455 /a‑revie w‑of‑the‑ without gaining additional consent from the original participants. The study evide nce‑asses sing‑impac t‑of‑socia l‑presc ribin g‑on‑healt hcare ‑deman protocol will be available. The data will be accessible beginning 3 months d‑and‑cost‑impli catio ns and ending 5 years following article publication to researchers who provide 11. Health England P. Eec ff tiveness of social prescribing ‑ An evidence a methodologically sound proposal and only to achieve the aims of that pro‑ synthesis. 2019 [cited 2020 Sep 11]. Available from: www.faceb ook. posal. Proposals should be directed to e.f.wood@sheffield.ac.uk. To gain access com/Publi cHeal thEng land requestors will need to sign a data access agreement. 12. Coventry PA, Hudson JL, Kontopantelis E, Archer J, Richards DA, Gilbody S, et al. Characteristics of Eec ff tive Collaborative Care for Treat ‑ ment of Depression: A Systematic Review and Meta‑Regression of 74 Declarations Randomised Controlled Trials. PLoS ONE. 2014;9(9):e108114. 13. Gunn J, Diggens J, Hegarty K, Blashki G. A systematic review of com‑ Ethics approval and consent to participate plex system interventions designed to increase recovery from depres‑ This research received ethical approval from the University of Sheffield sion in primary care. BMC Health Serv Res. 2006 Dec 16 [cited 2017 Apr Research Ethics Committee (reference number 016802) on 22 December 13];6(1):88. Available from: http://bmche alths ervre s.biome dcent ral. 2017 and performed in accordance with the ethical recommendations of the com/artic les/10.1186/1472‑6963‑6‑88 Helsinki Declaration. All participants provided informed written consent to 14. Wood E, Ohlsen S, Ricketts T. What are the barriers and facilitators to participate. implementing Collaborative Care for depression? A systematic review. J Affect Disord. 2017 May [cited 2017 Mar 7];214:26–43. Available from: Consent for publication http://linki nghub .elsev ier.com/retri eve/pii/S0165 03271 63155 06 The consent form contained elements related to publication. 15. Pawson R, Tilley N. Realist Evaluation. London: Sage Publications; 1997. 16. Checkland K, Harrison S, Marshall M. Is the metaphor of “barriers to Competing interests change” useful in understanding implementation? Evidence from None to declare. general medical practice. J Health Serv Res Policy. 2007;12(2):95–100. 17. Astbury B. Making claims using realist methods. In: Emmel N, Green‑ Author details haugh J, Manzano A, Monaghan M, Dalkin S, editors. Doing Realist School of Health and Related Research, The University of Sheffield, Regent Research. London: Sage Publications; 2018. Court, 30 Regent Street, Sheffield S1 4DA, UK. Institute for Mental Health, 18. Emmel N, Greenhaugh J, Manzano A, Monaghan M, Dalkin S. Doing School of Social Policy, University of Birmingham, Edgbaston, Birmingham B15 Realist Research. London: SAGE Publications Inc.; 2018. 2TT, UK. 19. Byng R, Norman I, Redfern S. Using Realistic Evaluation to Evaluate a Practice‑level Intervention to Improve Primary Healthcare for Patients Received: 21 May 2020 Accepted: 8 March 2021 with Long‑term Mental Illness. Evaluation. 2005 Jan 24 [cited 2019 Mar 14];11(1):69–93. Available from: http://journ als.sagep ub.com/ doi/10.1177/13563 89005 05319 8 20. Ricketts T, Wood E, Soady J, Saxon D, Hulin J, Ohlsen S, et al. The effect of comorbid depression on the use of unscheduled hospital care by References people with a long term condition: A retrospective observational 1. DDCMS. A connected society: A strategy for tackling loneliness ‑ laying study. J Aec ff t Disord. 2017;227. the foundations for change. London: HM Government, Department for 21. Naylor C, Taggart H, Charles A. Mental Health and new models of care. Digital, Culture, Media and Sport; 2018. Lessons from the vanguards. London: The King’s Fund; 2017. 2. NHS. The NHS Long Term Plan. London; 2019 [cited 2020 Mar 13]. Avail‑ 22. Fram SM. The Constant Comparative Analysis Method Outside of able from: www.longt ermpl an.nhs.uk Grounded Theory. Qual Rep. 2013 [cited 2018 Dec 20];18(1):1–25. 3. Husk K, Blockley K, Lovell R, Bethel A, Lang I, Byng R, et al. What Available from: https ://nsuwo rks.nova.edu/cgi/viewc onten t.cgi?artic approaches to social prescribing work, for whom, and in what circum‑ le=1569&conte xt=tqr stances? A realist review. Health Soc Care Community. 2020 Mar 9 23. Manzano A. The craft of interviewing in realist evaluation. Evaluation. [cited 2020 Mar 27];28(2):309–24. Available from: https ://onlin elibr ary. 2016 Jul 5 [cited 2017 Oct 18];22(3):342–60. Available from: http://journ wiley .com/doi/abs/10.1111/hsc.12839 als.sagep ub.com/doi/10.1177/13563 89016 63861 5 4. HEE. Social prescribing at a glance. A scoping report of activity for the 24. Westhorp G, Manzano A. REALIST EVALUATION INTERVIEWING • The North West. Manchester: Health Education England; 2016. RAMESES II Project (www.rames espro ject.org) Realist Evaluation 5. Payne K, Walton E, Burton C. Steps to benefit from social pre ‑ Interviewing‑A “Starter Set” of Questions The RAMESES II Project. 2017 scription: a qualitative interview study. Br J Gen Pract. 2019 Nov [cited 2019 Jan 18]. Available from: www.socio logy.leeds .ac.uk/ 18;bjgp19X706865. 25. Greenhalgh T, Humphrey C, MacFarlane F, Bulter C, Pawson R. How Do 6. Sigerson D, Gruer L. Asset‑based approaches to health improvement. You Modernize a Health Service? A Realist Evaluation of Whole‑Scale 2011 [cited 2019 Jul 18]. Available from: http://www.healt hscot land. Transformation in London. The Milbank Quaterly. 2009;87(2):391–416. Wood et al. BMC Fam Pract (2021) 22:53 Page 12 of 12 26. Wong G, Westhorp G, Manzano A, Greenhalgh J, Jagosh J, Greenhalgh T. RAMESES II reporting standards for realist evaluations. BMC Med. 2016 Dec 24 [cited 2019 Jan 11];14(1):96. Available from: http://bmcme dicin e.biome dcent ral.com/artic les/10.1186/s1291 6‑016‑0643‑1 27. Antonovsky A. Health, stress and coping. San Francisco: Jossey‑Bass Publishers; 1979. 28. Antonovsky A. Unravelling the Mystery of Health. How people manage stress and stay well. San Francisco: Jossey‑Bass Publishers; 1987. 29. Mittelmark M, Sagy S, Eriksson M, Bauer G, Pelikan J, Lindstom B, et al. The Handbook of Salutogenesis. AG Switzerland: Springer Nature; 2017. 30. Public Health England. Wider Determinants of Health. 2021 [cited 2021 Feb 5]. Available from: https ://finge rtips .phe.org.uk/profi le/wider ‑deter minan ts 31. Bowlby J. Attachment and loss, vol. 1. London: Hogarth; 1969. 32. Seligman MEP. LEARNED HELPLESSNESS. Annu Rev Med. 1972 [cited 2019 Jun 7];23:407–12. Available from: www.annua lrevi ews.org 33. Eriksson M, Lindström B. Antonovsky’s sense of coherence scale and the relation with health: a systematic review. J Epidemiol Community Heal. 2006 [cited 2019 Jun 6];60:376–81. Available from: http://jech.bmj.com/ 34. Pelikan J. The Application of Salutogenesis in Healthcare Settings. In: Mit‑ telmark M, Sagy S, Eriksson M, Bauer G, Pelikan J, Lindstom B, et al., editors. The Handbook of Salutogenesis. AG Switzerland: Springer Nature; 2017. 35. Mackenzie M, Skivington K, Fergie G. “The state They’re in”: Unpicking fantasy paradigms of health improvement interventions as tools for addressing health inequalities. Soc Sci Med. 2020;1(256):113047. 36. Koelen M, Eriksson M, Cattan M. Older People, Sense of Coherence and Community. In: Mittelmark M, Sagy S, Eriksson M, Bauer G, Pelikan J, Lindstom B, et al., editors. The Handbook of Salutogenesis. Springer; 2017. 37. Tierney S, Wong G, Roberts N, Boylan AM, Park S, Abrams R, et al. Support‑ ing social prescribing in primary care by linking people to local assets: A realist review. BMC Med. 2020 Mar 13 [cited 2020 Sep 11];18(1):1–15. Available from: https ://link.sprin ger.com/artic les/10.1186/s1291 6‑020‑1510‑7 Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in pub‑ lished maps and institutional affiliations. Re Read ady y to to submit y submit your our re researc search h ? Choose BMC and benefit fr ? Choose BMC and benefit from om: : fast, convenient online submission thorough peer review by experienced researchers in your field rapid publication on acceptance support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png BMC Family Practice Springer Journals

Social prescribing for people with complex needs: a realist evaluation

Loading next page...
 
/lp/springer-journals/social-prescribing-for-people-with-complex-needs-a-realist-evaluation-6UEb4x2E0p
Publisher
Springer Journals
Copyright
Copyright © The Author(s) 2021
eISSN
1471-2296
DOI
10.1186/s12875-021-01407-x
Publisher site
See Article on Publisher Site

Abstract

Background: Social Prescribing is increasingly popular, and several evaluations have shown positive results. How‑ ever, Social Prescribing is an umbrella term that covers many different interventions. We aimed to test, develop and refine a programme theory explaining the underlying mechanisms operating in Social Prescribing to better enhance its effectiveness by allowing it to be targeted to those who will benefit most, when they will benefit most. Methods: We conducted a realist evaluation of a large Social Prescribing organisation in the North of England. Thirty‑five interviews were conducted with stakeholders (clients attending Social Prescribing, Social Prescribing staff and general practice staff ). Through an iterative process of analysis, a series of context ‑mechanism‑ outcome con‑ figurations were developed, refined and retested at a workshop of 15 stakeholders. The initial programme theory was refined, retested and ‘applied’ to wider theory. Results: Social Prescribing in this organisation was found to be only superficially similar to collaborative care. A com‑ plex web of contexts, mechanisms and outcomes for its clients are described. Key elements influencing outcomes described by stakeholders included social isolation and wider determinants of health; poor interagency communi‑ cation for people with multiple needs. Successful Social Prescribing requires a non‑stigmatising environment and person‑ centred care, and shares many features described by the asset‑based theory of Salutogenesis. Conclusions: The Social Prescribing model studied is holistic and person‑ centred and as such enables those with a weak sense of coherence to strengthen this, access resistance resources, and move in a health promoting or saluto‑ genic direction. Keywords: Social Prescribing, Salutogenesis, Realist evaluation, Co‑morbidity, Multi‑morbidity, Depression Background typically involve a named professional who supports the The popularity and prominence of social prescribing (SP) person and helps link services and agencies involved in is growing. In October 2018, the UK Government com- their care [4, 5]. Asset-based approaches seek to posi- mitted to investment in SP as part of its Loneliness Strat- tively to mobilise the assets, capacities or resources egy [1]. The National Health Service Long Term Plan [2] available to individuals and communities which enable also highlights a key role for SP in the provision of health them to gain control over their lives and circumstances and social care in the UK. SP is not one intervention, it [6]. NHS England define personalised care (or person- is a pathway [3]; it is an umbrella term that encompasses centred care) as people having choice and control over interventions that are asset-based, person-centred and the way their care is planned and delivered based on what matters to them and their individual strengths and needs [7]. Most SP interventions aim to support indi- *Correspondence: e.f.wood@sheffield.ac.uk 1 viduals to have greater control over their own health School of Health and Related Research, The University of Sheffield, Regent Court, 30 Regent Street, Sheffield S1 4DA, UK [8], for instance through exercise or benefit advice pro - Full list of author information is available at the end of the article grammes [9]. While evidence for social prescribing is © The Author(s) 2021. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creat iveco mmons .org/licen ses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creat iveco mmons .org/publi cdoma in/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Wood et al. BMC Fam Pract (2021) 22:53 Page 2 of 12 broadly supportive [10], rigorous studies of effective - and ‘facilitators’ [16] to uncover underlying mechanisms ness and cost-effectiveness remains scarce [9, 11]. Previ- and understand how, when and why they are activated. ous reviews found indications that SP may be effective in RE requires researchers to have an initial theory that can improving health and well-being and reducing healthcare be challenged and refined during the study. Mid-range usage but included studies were small and many had low theories (which may or may not be novel) can be applied methodological quality, reducing confidence in the out - to the findings to improve generalisability of case study comes [9, 11] meaning the existing evidence base for SP findings [17, 18]. as a means of improving health outcomes is poor. By con- RE recognises that wider context influences outcomes trast, there is considerable support for SP at policy and and specifically how individual actors respond to dif - commissioning levels. ferent parts of the intervention at different times. To The link worker role in SP is considered key for peo - describe this complexity, statements describing path- ple with multiple long-term health conditions where care ways between intervention or individual Context (C), and can often be fragmented [7]. This resonates with collab - underlying Mechanisms (M) that subsequently shape pat- orative care models [12] which have four elements [13]: terns of outcomes (O) are created (CMO configurations) multi-professional approach (with one acting as a case [15]. CMO configurations aim to describe why a person coordinator); enhanced communication between profes- (or case) responds to an intervention in a certain way and sionals; structured management plan; and scheduled fol- how this can change depending on circumstances. CMOs low up. We therefore hypothesised that the underlying can be linear or more complex [15, 19]. mechanism for SP might be consistent with collaborative We focused on one large voluntary sector, community care [14]. anchor organisation providing SP to an inner-city area of This research set out to elucidate the mechanisms that high socioeconomic deprivation, predominantly deprived facilitate engagement and positive outcomes with SP white working class council estates. Clients often have intervention among people with multiple health condi- multiple health conditions including co-existing physi- tions and social needs. We sought to develop, test and cal and mental health concerns. Most have social needs refine the initial programme theory (that the benefits associated with housing, benefits and lack of support associated with SP derive from enacting collaborative networks. Many are isolated. Unlike some areas, there care) using stakeholder experiences. was no central referral point; general practices and other referrers (such as housing officers) referred to one or Methods more of many community anchor organisations that pro- Realist evaluation vide services in their local area (Table  1 provides details Realist evaluation (RE) looks at generative causation [15], of the setting). In some parts of the country, general prac- a key strength is modelling complexity. Adopting a realist tices have begun employing link workers but in this area, approach enabled us to go beyond identifying ‘barriers’ all social prescribing is provided by these third sector Table 1 Details of the study setting Organisation Type: Community Anchor Organisation ( Voluntary sector) Location: Inner‑ city area of high socioeconomic deprivation in a city in the north of England Referral type: General Practitioners and other services refer direct to the organisation. Self‑referral is also possible. Triage worker signposts to the most appropriate service, based on the client’s personal goals rather than the doctor’s determination of the problem. In 2018, there was a total of 1372 referrals; 813 from General Practitioners, 207 self‑referrals, 315 from other sources (e.g. housing, community mental health teams) and 37 whose refer ‑ ral status was not recorded. The number of clients with a mental health condition is not recorded; however, in 2018, 56 clients enrolled on the coping and self‑management programme and 59 enrolled on the emotional well‑being programme Services provided: Advice and services around health, employment and training. For the purposes of this study, we only considered the health sec‑ tion: this includes health training (e.g. weight loss or health eating advice, alcohol or cigarette reduction and exercise advice), social café’s, benefits and housing advice, and volunteer work. There is no set pathway through the service. Clients can access different services at different times in what ‑ ever order meets their needs. The service has no time limit Staff team: Paid health trainers, health activity workers and advocacy workers (primarily giving benefits and housing advice) and unpaid volunteers. Any of the paid workers could be a link worker, this would be decided based on client goals. Clients with predominantly health goals would have a health trainer as a link worker. Once these goals were met, the client may be referred to other colleagues if needed, for example for benefits advice. The person acting as link worker would change Care pathway: New clients are screened by the in‑house triage service to ensure the client is seen by the right part of the service to meet their goals. It is possible to move from one service to another or see multiple workers at one time depending on the nature of the client’s personal goals. The service also includes social café’s, which can take referrals or clients can drop in. You can attend the social café concurrently with other one to one services W ood et al. BMC Fam Pract (2021) 22:53 Page 3 of 12 organisations, link workers are not provided by primary Phase one tested the IPT through data collected in inter- care networks. views with people working for the organisation, those The initial programme theory (IPT) was informed by receiving SP and external referring organisations, to test the wider research team and relevant literature reviews and refine this theory. In Phase two, we developed CMO of collaborative care [20], realist methods [18] and social statements which refuted the IPT, leading to modification prescribing [9, 10, 21]. It was formed over multiple meet- of the theory. Phase three focused on applying and devel- ings between researchers and stakeholders from the host oping wider theory to synthesise the CMOs into a mid- site as well as another SP organisation in the same city, range theory. A mid-range theory is a theory that lies to further understand why and how SP works for certain between working hypotheses, contains testable predic- populations. The IPT to be tested was that collaborative tions and evolves as efforts are made to develop a unified care and SP are overlapping concepts (Fig. 1). We hypoth- theory. It was developed and related, where appropriate, esised that SP organisations are potentially effective in to existing theories [17, 18]. The mid-range theory pro - improving self-management for adults with co-existing vides a level of abstraction to the analysis and therefore a physical and mental health conditions because they work generalisability beyond the immediate context [17]. on a collaborative care model [12]. For the purpose of this study, intervention type was disaggregated from con- Data collection text in order to differentiate between interventions in one Phase 1: interviews contextual setting. We aimed to interview stakeholders with different points The study was iterative in nature and conducted over of view and experiences of SP. Clients were approached three phases. Stakeholders from the host community if they met inclusion criteria of having both physical and anchor organisation were included in development and mental health issues. Recruitment was via convenience design meetings from the start. This helped ensure that sampling and took place at different services within the the research was appropriate for the local model of SP organisation: health training, advocacy, volunteer devel- delivery and that it would provide insights of value to opment and social cafes. At the social cafes one of the both community anchor organisation s and researchers. researchers gave a short talk to attendees to introduce Fig. 1 The initial programme theory Wood et al. BMC Fam Pract (2021) 22:53 Page 4 of 12 the research and who we were looking for as partici- interviewed to get a wider range of opinions. As before, pants. Attendees were then asked to approach either clients self-identified as having co-morbid physical and the researcher or café staff if they wished to participate. mental health issues. Referrers came from different sec - For all other services, the staff member suggested the tors including primary care, housing, and community research to their client on a one to one basis. Everyone mental health teams and had referred at least one client was told participation was optional. to the service. As we had interviewed almost all the host All staff from the host SP organisation were invited to SP staff members, most staff members at the workshop interview via emails detailing the research. Doctors and had participated in interviews. practice nurses at local general practices were invited to Initial results from the interviews were reported at the interview as SP referrers. All practices associated with workshop, to gauge attendees’ views on our interpreta- the host organisation were approached. tions and areas where we considered there may be miss- The interview schedule evolved via a constant compar - ing data. Participants were asked to provide comments, ative method [22] such that each interview was informed criticisms and feedback on those interpretations. These by the ones that had taken place before it. The focus of were then incorporated into final analyses. the interviews was to present components of the IPT to The workshop lasted for three hours and clients the interviewee, on flash cards (face-to-face interviews received a store voucher for attending. for staff and clients) or verbal hypothesis (telephone interviews for referrers) for them to comment on with a Analysis view to providing theory refinement [23]. The interview Realist analysis takes an iterative approach, moving structure and questions were adapted from the RAME- between different sources of data and using deductive SES-II project [24] and flash cards were adapted accord - and inductive reasoning [25] (see Table 2). ing to who was being interviewed. The IPT was interrogated using data collected at inter - Interviews were audio recorded and lasted between view, by testing our hypothesis that SP is consistent with 15 and 60  min. Participants signed a consent form collaborative care [26]. This was done by analysing expe - before the interview commenced. Most interviews riences of three stakeholder subgroups (clients, staff and took place face-to-face in the host organisation’s prem- referrers) to see if their views corroborated or differed ises, though referrer interviews were conducted by from the IPT. A deductive approach was used. Analysis telephone as this was more convenient for clinicians. consisted of applying data to the four key elements of Interviews were conducted by EW and SO between collaborative care using a Framework approach. Themes February and April 2018. about what worked for whom, when and why that did and did not fit with the IPT were used to continually Phase 2: stakeholder workshop refine the model. Coding was undertaken by SO; EW and A workshop with key stakeholders [18] supported refine - JC independently coded 14 interviews to cross check the ment of the coding frame and analysis of emerging coding frame. NVivo software was used to assist in data themes. The results contributed to the emerging mid- management. range theory and CMO configurations. The developing themes became the premise for dis - Twenty participants were invited to the workshop. puting the IPT as the data were related to wellbeing not They were referrers into the service (e.g. National Health ‘health’ and ‘disease management’. These were developed Service staff), SP staff, or clients. Referrers and cli - into a series of CMO configurations by the wider research ents were only invited if they had not been previously team until agreement was reached. New configurations Table 2 Key elements of the Realist analysis process The realist methodology uses the following approaches judiciously and in combination: • Organizing and collating primary data and producing preliminary thematic summaries of these • Repeated writing and rewriting of fragments of the case study • Presenting, defending, and negotiating particular interpretations of actions and events both within the research team and also to the stakeholders themselves • Testing these interpretations by explicitly seeking disconfirming or contradictory data • Considering other interpretations that might account for the same findings • Using cross‑ case comparisons to determine how the same or submechanism plays out in different contexts and produces different outcomes, thereby allowing inferences about the generative causality of different contexts From Greenhalgh, T., Humphrey, C., MacFarlane, F., Bulter, C., & Pawson, R. (2009) W ood et al. BMC Fam Pract (2021) 22:53 Page 5 of 12 were then tested with data collected from the workshops of their experiences that would have aligned with col- in phase two. laborative care had it been a key underlying mechanism. Finally, we looked to extend generalisability beyond the They were then asked about what they felt was help - immediate context [17] by relating, where appropriate, ful about attending the SP service. When researching the newly created theory (mid-range theory) to existing whether an intervention ‘works’ it is important to define theories [17, 18]. Client-led interventions and asset-based its aims. One of the aims of SP is to improve health and approaches emerged as key to the mid-range theory. This well-being. However, other outcomes necessarily exist, is closely linked with salutogenesis [27, 28]. the service is client-led so clients will have their own Salutogenesis focuses on promoting health and well-being desired outcomes, which may or may not link directly as opposed to managing symptoms of disease. It further to improved health as defined by health services. The posits that life experiences shape one’s sense of coherence answers allowed us to refine the IPT (Fig. 2). (ability to comprehend a situation, find meaning and be The interventions in SP (Fig.  2), although superficially able to act). A strong sense of coherence aids in mobilising similar to collaborative care, were different in prac - resources (internal and external) for dealing with stress and tice and activated a wider range of mechanisms and helping a person move towards health rather than disease outcomes. [27, 29]. Salutogenic interventions are those which seek to The CMO configurations derived from the data strengthen a person’s ‘sense of coherence’ [29]. resulted in multiple interconnections, especially between mechanisms and outcomes, which did not Results fit [19] simple linear progressions such as ‘C + M = O’ Participants [15]. Many configurations reported multiple contexts The realist analysis was based on 35 interviews with cli - resulting in multiple mechanisms, which lead to an ents (n = 15), staff (n = 13), and referrers (n = 7) who outcome attained which then became a mechanism were all involved the health section of the SP community to achieve another outcome. For example, many cli- anchor organisation. All but one member of host SP staff ents were referred or self-referred to the organisation (who was on leave during recruitment) from the health because they were isolated (context). The social cafes section were interviewed in phase one. Staff backgrounds (intervention) facilitated making new social connec- varied and included health trainers, benefits/housing tions (mechanism) which reduced isolation (outcome). advisors and people who supported the organisation’s vol- Although this follows the CMO framework it does not unteers. The health section manager was also interviewed. represent the complexity found in the data, which were The stakeholder workshop had 15 delegates including more multifaceted than the above statement implies. seven staff, five referrers and three clients. Five staff at The familiar location of the cafés, which are set in local the workshop had been involved in the interviews. Two communities, was also a mechanism for increased senior managers attended who had not been invited to engagement, and therefore also reduced isolation. As interview. None of the referrers or clients at the work- did the increased confidence (Mechanism (M)) clients shop had been interviewed in a deliberate attempt to gained from trusting staff (M) who could work flexibly ensure we were getting a range of viewpoints. See Table 3 (M) and were not sited within the NHS (Context (C)) for participant information. but were still knowledgeable (M). Intermediate out- comes can, in turn become mechanisms for longer term Modification of the IPT outcomes. For example, increased engagement could The IPT was that SP works in a similar way to collabora - be an outcome in its own right but it is also a mecha- tive care. We asked participants about specific elements nism for improved mental and physical health. Figure 3 Table 3 Demographic details of the participants Interviews Referrers N = 6 (general practitioners, practice nurse, 2 male, 4 female) Staff N = 13 (1 manager, 4 health trainers, 3 advisors, 3 volunteer coordinators 2 triage, 6 males, 6 females, one preferred not to say) Clients N = 15 (12 clients, 3 clients who also volunteer, 5 males, 9 females, 1 preferred not to say) Focus group Referrers N = 5 (community mental health team, housing, social work) Staff N = 7 (2 managers, health trainers, advisors) Clients N = 3 (2 male and 1 female) Wood et al. BMC Fam Pract (2021) 22:53 Page 6 of 12 Fig. 2 The mid‑range theory developed after interviews with SP stakeholders shows an overview of the interweaving CMOs that we of the staff roles (M) and staff having time to listen* discovered. (M); resulting in improved Mental Health (O) and more Some of the configurations demonstrate the overlaps appropriate health service use* (O). and differences between the original and modified IPTs. Staff felt clients valued this approach as they would Points marked with a * are from the modified IPT but receive word of mouth self-referrals from friend and fam- not the original, unmarked points stem from both, sup- ily of existing clients. porting quotes are illustrated below. Person‑centred care When thinking about person centred interventions one Non‑stigmatising environments staff member spontaneously described their own idea of Adults with depression but no diagnosis/treatment* (C) the CMO configuration. are accessing SP services in community settings* (C), When someone is new to the service (C), it’s crucial this is perceived as less stigmatising (M) and intimidat- that we do it in an approached manner (M) that ing than clinical services* (M), resulting in increased we can do it at their speed, (M) feeling comfortable engagement (Outcome (O)) and improved mental about it (O), giving confidence that they can do it health (O). (M), and allowing that to flourish (O) and say ‘come SP [organisations] are there for people with low on, we can move forward’ (M). So, it’s empowered confidence (C) so they’re not going to look down on them (O), they’ve got to make that choice (M) and you (M). Client 1 they’ve got to make those decisions but it’s about Chatting to people (M), you know you’re not on being supportive isn’t it (M), to doing it. And that’s your own (M). You know you’re not the only person what I see my role, is supporting people and moving who’s had problems (C/M). Client 3 them on to next… every individual has structured management plan (O), speaks for itself. Every one’s This SP organisation offers a welcoming setting that different. It’s not my plan. It’s their plan (M). Staff 11 helps the client’s mental health but does not focus on it directly. This suggests that staff valued the flexibility of their Adults who are isolated* (C) possibly as a result of role working at the client’s pace, seeing it as integral to a bereavement* (C) with mental health issues (C) are offering person-centred care. ‘It’s not my plan, it’s their receiving personalised client centred management plans plan’ implies that their role is to facilitate and guide the in SP* (Intervention). This is can be provided on a flexible individual to choose rather than ‘intervening’ in the basis to meet their need* (M), partly due to the flexibility W ood et al. BMC Fam Pract (2021) 22:53 Page 7 of 12 Fig. 3 Diagram detailing some of the main CMOs and their interlinking nature classic model. Whilst many health care workers do have Clients value the peer support that the social cafes’ a degree of autonomy and flexibility, they are usually con - provide; reducing not only physical isolation but also strained in this. The SP workers here described a level of emotional isolation by introducing clients to people with working—supported by management—to meet the needs similar issues, who are able to support each other. This of clients however unusual, for example, aiding in house reduction in isolation was also felt to have an effect on clearance for a hoarder. the physical health of clients and ultimately their attend- ance at General practice clinics. For elderly widowed males who had been dependent on their partner for Social isolation essential life skills (cooking) (C) personal help with shop- Clients with the similar contexts attend social cafes: Have ping and guidance may increase confidence and motiva - a space/location to engage in participative social function tion (M) to eat better and lose weight (O). (meet people/peers/friends/shared experiences)* (M) are By regularly attending SP activities(C), clients often listened to (by peers)* (M), which improves/increases meet others who have been through similar situations social functioning (O) and reduces isolation* (O) and (C) it creates a social network(O/M), shared experience improves mental health (O). gives peer support (O), Reducing sense of isolation and It’s building my confidence up great (C/O). I’m mak - reliance on health services being the only place you can ing loads of friends (O). I mean, I’m in a craft group discuss your health(M/O). Leading to reduced primary but I don’t really do much crafting when I’m It’s health care use. (O). more chatting (M) and helping the others (M), so it’s lovely, and they’re just so friendly(M). Client 14 Wider determinants of health Instead of once a fortnight, they’re going to some- Wider determinants of health are a range of economic, where twice a week now (O) so there’s, there’s always social and environmental factors that directly and indi- something for them to do (M) and it brings them rectly affect people’s health [30]. Clients present with together (M),  I mean they say to me, things like oh, social issues* (C) contributing to mental health issues (C), if I didn’t come here I’d have nowhere else to go (O), staff in SP have skills to support social issues* (C). Clients I’d be sat, four walls (C), I don’t know what I’d do if I receive personalised management plans, the interventions didn’t have this group, and that type of thing. Staff 7 Wood et al. BMC Fam Pract (2021) 22:53 Page 8 of 12 are adaptable to this individual need* (M) with staff being Many SP organisations are in the voluntary sector flexible in how this support is offered/delivered* (M) and are isolated from statutory services. Data shar- Resulting in improvements in housing/finances/employ - ing is therefore problematic. Information is not passed ment* (O) and positive impacts on mental health (O). between organisations in a timely manner and even when it is, there is no interoperability in IT systems. The lack of We can even do a home visit (Intervention), because professional status of SP staff leads NHS and other staff even asking someone to come and see us here for the to be unsure about what information they can or cannot rfi st time is daunting (C). so I think with us, slot - share with them. ting in with them (M), I see it as like a jigsaw, so it just all slots in because they see us and we look at the barriers to health and put them steps in first Mid‑range theory and work through them with them (M), and then it’s Salutogenesis just giving them that bit of self-belief that they can Central to the model of salutogenesis is the concept of do something and show them how they can make a sense of coherence which is a ‘generalised, long last- small changes that that’s leads to bigger things so by ing way of viewing the world and one’s place in it’ [27]. us being there, they can then move on to volunteer- Although considered to be stable in adults, it can be ing (O) and then move on to employment advocacy altered particularly by radical changes. Additionally, (O), if they are on [employment and support allow- it influences whether or not an individual attempts to ance] they can help them sort out the benefits and change their situation [27]. People with strong sense of what have you but then they can refer back into us coherence, view the world as predictable and therefore again, to say well actually they are on ESA but they comprehensible. Antonovsky links this theory to those are looking at wanting to return to work but they of Bowlby (attachment) [31] and Seligman (learned help- have got no confidence (C), you know, so we sort of lessness) [32] while highlighting differences. The sense of can keep seeing them (M). Staff 9 coherence is considered to be a continuum from strong to weak, characterised by a normal distribution [27]. SP in this organisation contributed to health improve- Salutogenesis is an individual level explanation of ments by concerning themselves with the wider determi- health behaviours. Previous studies have shown a rela- nants of health, not just presenting health concerns. tionship between a strong sense of coherence and good perceived health, particularly mental health [33]. It also Poor interagency communication seeks to explain why some individuals do not respond to Clients present with complex social issues* (C). Clients health information from professionals. From a saluto- receive personalised management plans (Intervention). genic perspective, this should not be seen as an individual But poor shared systems with external organisations (M) failure but a failure of the service to provide understand- and external organisations perceiving SP staff as non- able information [33]. Salutogenesis also refers to the professional (M) resulted in difficulty with systematic ability to utilise resources (both external and internal) to sharing of information (O) and longer waits with more manage stressful situations. The ability to recognise and chance for clients to disengage (O). use these resources is important for sense of coherence. Clients reported that staff helped their understanding I think that would make a huge difference, because of their issues. They found health trainers to be motivat - if a [General Practitioner] was to log in and see that ing and knowing they had someone to turn to helped they’re working with social prescribers and they’re them to feel that their situation was more manageable. going to groups and this has happened and that has The location of SP away from statutory health services happened, then we can work and keep encouraging and the ability of staff to work differently to health staff them to go, you know those sort of things. General (with difference in time and role flexibility) aided accept - Practitioner 2 ability of the service. However, the fact that it was a Very often there’ll be interruptions in claims, bene- non-statutory service did cause problems for informa- fits will get suspended. If I could talk to [the Council] tion sharing. In this way the client’s comprehension and at that time when the client’s here I could stop that understanding was often improved but this is not always happening, whereas now… the letters’ll be god knows the case for SP and primary care staff. SP can empower where… and in the meantime you know the benefit people to utilise their resources and develop new ones. might get suspended… If I could talk to them I could Resources can be internal, such as confidence or self- solve a lot of problems because I can put in a nut- esteem, or external, such as friends or community who shell what the client might struggle to sort of want to provide advice, support, or bolster internal resources. put across. Staff 3 W ood et al. BMC Fam Pract (2021) 22:53 Page 9 of 12 SP can be considered an external resource but there are invalidate the efforts made in providing these services. many facets to this, due to the different models of SP For example, people in financial hardship may struggle that exist. However, examples include, the link worker as to access a service that requires bus travel to attend. One source of support in a crisis or as a way to access other of the things participants liked about the organisation sources of support, and the community groups provided studied in this research was that it provided services in as part of SP that offer support, companionship and several local community locations, meaning most people advice. could walk to services, knew the area and felt comfort- able there. Although sense of coherence is stable, it can be Discussion changed, but to do so in a positive way is slow and takes Summary ‘hard work’, such as undertaking psychotherapy [27]. SP in this setting was a not collaborative healthcare inter- More recent research shows that sense of coherence can vention but rather a client led, person-centred, asset- change across the life course and that many prerequisites based service addressing the wider social determinants for strengthening sense of coherence may be provided of health including co-morbid conditions. In doing so, it by or mediated by the community [36]. The version of worked on salutogenic principles, providing new resist- SP that was delivered by the provider organisation, at ance resources, helping people access existing ones and the time of our research, and the ways that it was expe- even strengthening low sense of coherence. rienced by clients, was consistent with the theory of salu- Personalised care is at the heart of social prescribing, togenesis. This research and that of Payne and colleagues staff refer to goal planning in SP as ‘ their plan not mine’; [5] found that SP facilitated change in perceptions of per- SP goals are set by the clients not staff. Even when clients sonal assets through personal and social development. are referred by health care staff for specific purposes, if Meaning that clients became more aware of what assets that is not what the client wishes to address first (or at all) were available to them and more able to access them. then this is not a condition of entring the service. Whilst This is consistent with theory of change for the sense of this personalised approach clearly has its merits, and the coherence and therefore supports Health Education Eng- repeated return to SP organisations and word of mouth land’s suggestion that the theoretical base for SP is Salu- recommendations show it is a popular policy, it may dis- togenesis [4]. advantage those whose sense of coherence is so low that When considering what aspects of SP works for whom they cannot recognise or articulate the need to change. and in what circumstances, context is important. Access However, given the broad range of activities that is avail- to a supportive community can strengthen a person’s able via SP this should be less of a disadvantage than in sense of coherence through the life course [36]. People traditional health care settings. It also suggests a signifi - with a low sense of coherence may struggle to access cant departure from the collaborative framework of col- these resources without help and these people can par- laborative care. The client is not being consulted by the ticularly benefit from SP interventions, although the spe - healthcare professional; they are leading the direction of cific intervention needs to be determined on a case by the intervention. This has the potential to have an effect case basis and remain person centred. on the level of dependancy people have on the service. A recent realist review [3] reported that there are three However, although many of the interviews with clients stages which contribute to pathway success in social pre- mentioned dependecy it was in relation to others. This scribing. Enrolment, engagement and adherence and suggests participants were aware of it but that it might be the link workers are key to avoiding disruption of the difficult to recongise or admit to, it is however, an issue process. Our findings are consistent with these points. that is both live and sensitive. Link workers have a key role in ensuring people are sup- ported to attend and understand what social prescribing is. Similarly, another recent realist review of social pre- Comparison with previous studies scribing had two main concepts, creating and sustaining Pelikan concluded that salutogenic thinking has good buy-in, and establishing and maintaining connections potential to be applied in health settings [34]. Specifically [37]. The first of these was not a major concern in our in health promoting interventions, structures and cul- study as referrers were enthusiastic about social prescrib- tures and improving a person’s sense of coherance could ing and keen to work with the organisation. Clients too, be an explicit goal of chronic disease management [34]. would regularly report recommending the organisation The wider determinants of health must not be underes - to friends and neighbours. The caveat here is that both timated when planning SP delivery [35], they are part of were self-selecting samples and we did not (and could the context to our realist theory and may constrain the not) gauge how representative they were. The second, effects of the SP intervention. However, this does not Wood et al. BMC Fam Pract (2021) 22:53 Page 10 of 12 maintaining connections, was more of an issue, both in social prescribing referrals. It is important to note that terms of logistics around data sharing and governance, those with the weakest sense of coherence, who might but also regarding remit and scope of the social prescrib- be expected to benefit most from social prescribing, ing provider and its services. These changed over time may lack sufficient resources to access the service and and some family doctors reported that they were unsure may need significant help (from referrers and social pre - about which service to refer too. This was the reason for scribing providers). A key barrier to this is difficulty in developing the triage role, so that referrers had a sin- information across organisational boundaries, and those gle point of contact to improve relations. Our findings who commission SP services are perhaps best placed to complement Tierney et  al.’s conclusions that social pre- address this. scribing can, through developing wider social networks, We found that flexibility on the part of link workers was reduce isolation, increase meaning and activity and give critical, and they require sufficient time and resources to people the confidence to manage their own health. Our work this way. Link workers in our study recognised the theory development diverged from that of Tierney et al., need to support engagement and that regular attend- whose primary focus was on the setting up of services ance was key for reducing isolation. Processes or policies whereas we studied a mature service that had existed regarding discharge based on time spent in the service for many years. This allowed us to look beyond logistical were therefore felt by many to be counter-productive, considerations into deeper theory behind social prescrib- although the consequences for service capacity was also ing in action. recognised. Commissioners and service providers need to find the correct balance for their chosen aims as differ - ent organisations may need to work in different ways to Strengths and limitations satisfy budgets and local need. SP providers are very varied. The service we studied was provided by a large organisation was large and comprised multiple interventions. As the local authority did not Implications for research have a central SP referral point the organisation had to Theory development moved from a large scale IPT of liaise with referrers directly, attempting to find local solu - SP, then modified to the SP organisation in the study tions to a city-wide problem. This situation is not the then widened out again for the mid-range theory. The same in other cities or for small single intervention SP mid-range theory therefore should be tested in addi- groups. However, the central finding of the importance of tional SP organisations to ensure it is not context being client centred and of acknowledging wider deter- specific. minants of health remain relevant to a wide variety of dif- ferent SP models of implementation. This study focussed on an organisation that predomi - Conclusions nantly served deprived white working-class council Although superficially similar, social prescribing does not estates. There were limited opportunities to capture the appear to operate as a type of collaborative care. Collabo- voices of people from minority ethnic and non-white rative care does represent a move towards more holistic British backgrounds. thinking within healthcare settings, but social prescrib- ing takes this concept even further. It is holistic and per- Implications for practitioners and commissioners son-centred and as such may enable those with a weak We sought a better understanding of the mechanisms of sense of coherence to strengthen this, access resistance action for social prescribing, to provide greater clarity resources and move in a health promoting or salutogenic about who can be expected to benefit and why, and there - direction. fore who to prioritise in referral practices. Although in- depth, our study was based on a single social prescribing Abbreviations provider in one city and our findings need to be viewed SP: Social Prescribing; IPT: Initial Programme Theory; CMO: Context‑mecha‑ cautiously. nism‑ outcome; C: Context; M: Mechanism; O: Outcome. We found that people with long-term social difficul - Acknowledgements ties who struggle with chronic health conditions because The authors would like to thank all of the social prescribing staff, clients and they have limited support networks may benefit the most health services staff who participated in our research and offered their valu‑ able insights. from social prescribing. Our results support the view that this is because building support networks and develop- Authors’ contributions ing coping mechanisms enables better engagement with All people entitled to be authors have been included. EW drafted the paper, collected and analysed the data and contributed to the survey and protocol sources of treatment. We therefore recommend that development. SO and JC assisted with the analysis. SJF gave significant these individuals should be considered a priority for W ood et al. BMC Fam Pract (2021) 22:53 Page 11 of 12 guidance on the Realist method. SO, SJF and SW contributed to protocol com/uploa ds/docum ents/17101 ‑asset Based Appro aches toHea lthIm development. All authors read and approved the final manuscript.prove mentB riefi ng.pdf 7. NHS England. Social prescribing and community‑based support Sum‑ Funding mary guide. London; 2019 [cited 2019 May 16]. Available from: https :// This article presents independent research funded by Sheffield Hospitals www.engla nd.nhs.uk/wp‑conte nt/uploa ds/2019/01/socia l‑presc ribin Charity and the National Institute for Health Research Collaboration for g‑commu nity‑based ‑suppo rt‑summa ry‑guide .pdf Leadership in Applied Health Research and Care Yorkshire and Humber 8. The King’s Fund. What is social prescribing? 2017 [cited 2017 Aug (NIHR CLAHRC YH). The views and opinions expressed are those of the 25]. Available from: https ://www.kings fund.org.uk/publi catio ns/socia authors, and not necessarily those of Sheffield Hospitals Charity or the NIHR l‑presc ribin g?gclid =EAIaI QobCh MI052 5kffx 1QIVi BXTCh 3_OAHhE or the Department of Health. The funding bodies had no role in the design AAYAS AAEgL ecPD_BwE of the study, collection, analysis or interpretation of the data, or in writing 9. Bickerdike L, Booth A, Wilson PM, Farley K, Wright K. Social prescribing: the manuscript. less rhetoric and more reality. A systematic review of the evidence. BMJ Open. 2017 Apr 1 [cited 2018 Dec 20];7(4):e013384. Available from: Availability of data and materialshttp://www.ncbi.nlm.nih.gov/pubme d/28389 486 Data from this study may be available upon reasonable request to the authors. 10. Polley MJ, Pilkington K. A review of the evidence assessing impact of Individual participant data that underlie the results reported in this article will social prescribing on healthcare demand and cost implications. Univer‑ only be available after deidentification (text, tables, figures and appendices) sity of Westminster; 2017 [cited 2019 Jan 11]. Available from: https :// if the requestor has Research Ethics Committee approval to work with data westm inste rrese arch.westm inste r.ac.uk/item/q1455 /a‑revie w‑of‑the‑ without gaining additional consent from the original participants. The study evide nce‑asses sing‑impac t‑of‑socia l‑presc ribin g‑on‑healt hcare ‑deman protocol will be available. The data will be accessible beginning 3 months d‑and‑cost‑impli catio ns and ending 5 years following article publication to researchers who provide 11. Health England P. Eec ff tiveness of social prescribing ‑ An evidence a methodologically sound proposal and only to achieve the aims of that pro‑ synthesis. 2019 [cited 2020 Sep 11]. Available from: www.faceb ook. posal. Proposals should be directed to e.f.wood@sheffield.ac.uk. To gain access com/Publi cHeal thEng land requestors will need to sign a data access agreement. 12. Coventry PA, Hudson JL, Kontopantelis E, Archer J, Richards DA, Gilbody S, et al. Characteristics of Eec ff tive Collaborative Care for Treat ‑ ment of Depression: A Systematic Review and Meta‑Regression of 74 Declarations Randomised Controlled Trials. PLoS ONE. 2014;9(9):e108114. 13. Gunn J, Diggens J, Hegarty K, Blashki G. A systematic review of com‑ Ethics approval and consent to participate plex system interventions designed to increase recovery from depres‑ This research received ethical approval from the University of Sheffield sion in primary care. BMC Health Serv Res. 2006 Dec 16 [cited 2017 Apr Research Ethics Committee (reference number 016802) on 22 December 13];6(1):88. Available from: http://bmche alths ervre s.biome dcent ral. 2017 and performed in accordance with the ethical recommendations of the com/artic les/10.1186/1472‑6963‑6‑88 Helsinki Declaration. All participants provided informed written consent to 14. Wood E, Ohlsen S, Ricketts T. What are the barriers and facilitators to participate. implementing Collaborative Care for depression? A systematic review. J Affect Disord. 2017 May [cited 2017 Mar 7];214:26–43. Available from: Consent for publication http://linki nghub .elsev ier.com/retri eve/pii/S0165 03271 63155 06 The consent form contained elements related to publication. 15. Pawson R, Tilley N. Realist Evaluation. London: Sage Publications; 1997. 16. Checkland K, Harrison S, Marshall M. Is the metaphor of “barriers to Competing interests change” useful in understanding implementation? Evidence from None to declare. general medical practice. J Health Serv Res Policy. 2007;12(2):95–100. 17. Astbury B. Making claims using realist methods. In: Emmel N, Green‑ Author details haugh J, Manzano A, Monaghan M, Dalkin S, editors. Doing Realist School of Health and Related Research, The University of Sheffield, Regent Research. London: Sage Publications; 2018. Court, 30 Regent Street, Sheffield S1 4DA, UK. Institute for Mental Health, 18. Emmel N, Greenhaugh J, Manzano A, Monaghan M, Dalkin S. Doing School of Social Policy, University of Birmingham, Edgbaston, Birmingham B15 Realist Research. London: SAGE Publications Inc.; 2018. 2TT, UK. 19. Byng R, Norman I, Redfern S. Using Realistic Evaluation to Evaluate a Practice‑level Intervention to Improve Primary Healthcare for Patients Received: 21 May 2020 Accepted: 8 March 2021 with Long‑term Mental Illness. Evaluation. 2005 Jan 24 [cited 2019 Mar 14];11(1):69–93. Available from: http://journ als.sagep ub.com/ doi/10.1177/13563 89005 05319 8 20. Ricketts T, Wood E, Soady J, Saxon D, Hulin J, Ohlsen S, et al. The effect of comorbid depression on the use of unscheduled hospital care by References people with a long term condition: A retrospective observational 1. DDCMS. A connected society: A strategy for tackling loneliness ‑ laying study. J Aec ff t Disord. 2017;227. the foundations for change. London: HM Government, Department for 21. Naylor C, Taggart H, Charles A. Mental Health and new models of care. Digital, Culture, Media and Sport; 2018. Lessons from the vanguards. London: The King’s Fund; 2017. 2. NHS. The NHS Long Term Plan. London; 2019 [cited 2020 Mar 13]. Avail‑ 22. Fram SM. The Constant Comparative Analysis Method Outside of able from: www.longt ermpl an.nhs.uk Grounded Theory. Qual Rep. 2013 [cited 2018 Dec 20];18(1):1–25. 3. Husk K, Blockley K, Lovell R, Bethel A, Lang I, Byng R, et al. What Available from: https ://nsuwo rks.nova.edu/cgi/viewc onten t.cgi?artic approaches to social prescribing work, for whom, and in what circum‑ le=1569&conte xt=tqr stances? A realist review. Health Soc Care Community. 2020 Mar 9 23. Manzano A. The craft of interviewing in realist evaluation. Evaluation. [cited 2020 Mar 27];28(2):309–24. Available from: https ://onlin elibr ary. 2016 Jul 5 [cited 2017 Oct 18];22(3):342–60. Available from: http://journ wiley .com/doi/abs/10.1111/hsc.12839 als.sagep ub.com/doi/10.1177/13563 89016 63861 5 4. HEE. Social prescribing at a glance. A scoping report of activity for the 24. Westhorp G, Manzano A. REALIST EVALUATION INTERVIEWING • The North West. Manchester: Health Education England; 2016. RAMESES II Project (www.rames espro ject.org) Realist Evaluation 5. Payne K, Walton E, Burton C. Steps to benefit from social pre ‑ Interviewing‑A “Starter Set” of Questions The RAMESES II Project. 2017 scription: a qualitative interview study. Br J Gen Pract. 2019 Nov [cited 2019 Jan 18]. Available from: www.socio logy.leeds .ac.uk/ 18;bjgp19X706865. 25. Greenhalgh T, Humphrey C, MacFarlane F, Bulter C, Pawson R. How Do 6. Sigerson D, Gruer L. Asset‑based approaches to health improvement. You Modernize a Health Service? A Realist Evaluation of Whole‑Scale 2011 [cited 2019 Jul 18]. Available from: http://www.healt hscot land. Transformation in London. The Milbank Quaterly. 2009;87(2):391–416. Wood et al. BMC Fam Pract (2021) 22:53 Page 12 of 12 26. Wong G, Westhorp G, Manzano A, Greenhalgh J, Jagosh J, Greenhalgh T. RAMESES II reporting standards for realist evaluations. BMC Med. 2016 Dec 24 [cited 2019 Jan 11];14(1):96. Available from: http://bmcme dicin e.biome dcent ral.com/artic les/10.1186/s1291 6‑016‑0643‑1 27. Antonovsky A. Health, stress and coping. San Francisco: Jossey‑Bass Publishers; 1979. 28. Antonovsky A. Unravelling the Mystery of Health. How people manage stress and stay well. San Francisco: Jossey‑Bass Publishers; 1987. 29. Mittelmark M, Sagy S, Eriksson M, Bauer G, Pelikan J, Lindstom B, et al. The Handbook of Salutogenesis. AG Switzerland: Springer Nature; 2017. 30. Public Health England. Wider Determinants of Health. 2021 [cited 2021 Feb 5]. Available from: https ://finge rtips .phe.org.uk/profi le/wider ‑deter minan ts 31. Bowlby J. Attachment and loss, vol. 1. London: Hogarth; 1969. 32. Seligman MEP. LEARNED HELPLESSNESS. Annu Rev Med. 1972 [cited 2019 Jun 7];23:407–12. Available from: www.annua lrevi ews.org 33. Eriksson M, Lindström B. Antonovsky’s sense of coherence scale and the relation with health: a systematic review. J Epidemiol Community Heal. 2006 [cited 2019 Jun 6];60:376–81. Available from: http://jech.bmj.com/ 34. Pelikan J. The Application of Salutogenesis in Healthcare Settings. In: Mit‑ telmark M, Sagy S, Eriksson M, Bauer G, Pelikan J, Lindstom B, et al., editors. The Handbook of Salutogenesis. AG Switzerland: Springer Nature; 2017. 35. Mackenzie M, Skivington K, Fergie G. “The state They’re in”: Unpicking fantasy paradigms of health improvement interventions as tools for addressing health inequalities. Soc Sci Med. 2020;1(256):113047. 36. Koelen M, Eriksson M, Cattan M. Older People, Sense of Coherence and Community. In: Mittelmark M, Sagy S, Eriksson M, Bauer G, Pelikan J, Lindstom B, et al., editors. The Handbook of Salutogenesis. Springer; 2017. 37. Tierney S, Wong G, Roberts N, Boylan AM, Park S, Abrams R, et al. Support‑ ing social prescribing in primary care by linking people to local assets: A realist review. BMC Med. 2020 Mar 13 [cited 2020 Sep 11];18(1):1–15. Available from: https ://link.sprin ger.com/artic les/10.1186/s1291 6‑020‑1510‑7 Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in pub‑ lished maps and institutional affiliations. Re Read ady y to to submit y submit your our re researc search h ? Choose BMC and benefit fr ? Choose BMC and benefit from om: : fast, convenient online submission thorough peer review by experienced researchers in your field rapid publication on acceptance support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions

Journal

BMC Family PracticeSpringer Journals

Published: Mar 18, 2021

There are no references for this article.