Social prescribing and classed inequality: A journey of upward health mobility?

Social prescribing, characterised by a link worker connecting patients with local groups and services, is currently being widely implemented in the UK. Taking clients ’ experiences of a social prescribing intervention in the North of England between November 2019 and July 2020 as its focus, this paper employs ethnographic methods to explore the complex social contexts in which social prescribing is delivered. Building on Bourdieusian approaches to class, we concentrate on four case studies to offer a theoretically-grounded analysis which attends to the relationship between everyday contexts and the classed processes by which health capital may be accrued. By following clients ’ experiences and trajectories through shifting positions across time - often entailing moments of tension and disjuncture - we explore how processes of classed inequality relate to engagement in the social prescribing intervention. Our results show how structural contexts, and relatedly the possession of capital, shape clients ’ priorities to invest in the cultural health capital offered by the intervention. Importantly, while inequalities shaped participants ’ capacity to engage with the intervention, all participants recognised the value of the health capital on offer. We conclude by arguing that inequalities cannot be tackled through focusing on the individual in the delivery of personalised care and therefore offer a counter narrative to socio-political as- sumptions that social prescribing reduces health inequalities. Crucially, we argue that such assumptions wrongly presuppose that people are homogenously disposed to engaging in their future health.


Introduction
Social prescribing is a topic of growing policy, practice, academic and political interest. Promoted as an effective way to address social issues affecting health and well-being (South et al., 2008), social prescribing is particularly targeted at people with long-term health conditions (LTCs) and/or mental health issues and attends to the non-medical needs of individuals (Pescheny et al., 2018;Tierney et al., 2020). Ranging from active signposting through to more intensive approaches, social prescribing typically involves a non-medical link worker/facilitator who addresses patients' personalised support needs (Howarth and Donovan, 2019). This is often through 'co-producing' a personalised plan followed by referrals into relevant voluntary and community sector activities, local authority or health services (Wildman et al., 2019;Frostick and Bertolli, 2019). Examples include gym referrals, bene ts and housing advice, and community classes. Despite there being a paucity of robust evidence regarding its effectiveness (Bickerdike et al.,201 ), social prescribing is currently being implemented on a large scale in the UK and is gaining international traction (Rowlands, 2020). Identi ed as playing a key role in the delivery of the NHS Model of Personalised Care, the aim is that by 202 /2 over 900,000 people in England will be referred into social prescribing (NHS England, 2019c). It is increasingly framed as part of a broader commitment to reducing health inequalities (NHS England, 2019a, 2019b, 2019c. This is despite a lack of evidence that individualised interventions such as social prescribing can reduce health inequalities 201 ). Recently, a 10-year review of The Marmot Report (Marmot et al., 2020) observed that health inequalities are increasing in the UK and called for more research into how social prescribing might affect health inequalities.
Our paper illuminates the impact of social prescribing on health inequalities by exploring the classed contexts shaping clients' experiences of a social prescribing intervention in the North of England. We use Bourdieu's concepts of habitus, eld and capital as a lens through which to analyse how practices of client engagement are connected to class. We pay particular attention to the spatio-temporal nature of everyday practice to explore how class enables and constrains participation in social prescribing interventions. We illuminate how the possession of capital relates to the accrual of the socially-valued health dispositions presented by the intervention and the implications of this process for the impact of social prescribing on health inequalities. We focus on how participants' habitus aligns with the approach of this intervention, which employs behaviour change methods such as motivational interviewing to assist clients with identifying aspects of their lives which are considered to require changing. The intervention's expectation is that connecting clients from an area of high deprivation to local resources and activities will achieve its aims that clients will gain better access to specialised services, improve health-related 'behaviours', and develop an improved 'attitude' towards challenges, thereby improving health and wellbeing. In doing so, the intervention explicitly aims to tackle health inequalities.
It is well recognised that socio-economic status is a 'fundamental cause' of health inequalities (Phelan et al., 2010). Whilst this is useful for emphasising the relationship between economic resources and health, following Bourdieu we seek to explore how the material and symbolic con gurations of class shape access to health. An emerging body of work has employed Bourdieu's capital-based approach as a frame of reference to understand health inequalities (Nettleton and Green,201 ;Dumas et al.,201 ), and a limited number of studies have extended this approach to evaluate health interventions (Warin et al., 201 ;Hanckel et al., 2020;Luca et al., 2019;Wiltshire et al., 2019). For Bourdieu, class exists twice: through unequal distributions of capital in social structures and in the individual dispositions they produce (Bourdieu and Wacquant,201 ). Class is expressed and negotiated by individuals through everyday practice in relation to their position and journey across social elds. Crosscut with social divisions such as class, elds are structured according to the exchange and conversion of capital. There are three main forms of capital: economic capital ( nancial resources), social capital (resources accrued from belonging to particular social networks), and cultural capital (forms of knowledge) (Bourdieu,198 ). An individual's social position, as well as their future trajectory of accruing further capital, depends on the legitimacy of different types of capital they inherit (Bourdieu,198 ). Class therefore, shapes the biographical temporalities of practice: 'for in habitus the past, the present and the future intersect and interpenetrate one another' (Bourdieu and Wacquant, 1992: 22). Warin et al. (201 ) build on this understanding in their examination of the disjuncture between the future-orientated nature of health interventions and the temporalities of everyday practice. They argue that individual futures cannot be separated from present or past contexts and call for public health initiatives to be located in 'dynamics of a living present, tailored to the particular, localised spatio-temporal perspectives and material circumstances in which people live' (ibid.: 09).
Returning to Bourdieu, the concept of habitus captures the relational ways that classed social relations are embodied in continuous dialogue with an individual's position and journey through social space. Referring to the imprint of history which lies within an individual, habitus is durable. It is 'embodied history, internalised as second nature and so forgotten as history' (Bourdieu, 1990: ). It thus encapsulates the ways that the social world is in the body as well as on the body (Reay,200 ). But habitus is not passively inscribed in the Bourdieusian body. It is the generative result of a practical and mediating relationship between social structures and everyday action, where the social world is re exively understood through categories constructed by previous experience: it 'is an open system of dispositions that is constantly subjected to experiences, and therefore constantly affected by them' (Bourdieu and Wacquant, 1992: 1 ; italics original). It thus refers to an active, not passive, set of dispositions which is continually restructured through interactions with the eld (McNay, 2008a).
Particularly important to the argument we present here is that the tacit and taken-for-granted 'logic of practice' (Bourdieu, 1990) relates to the extent to which habitus and eld are congruent or not. Such practical knowledge shapes how change in practice is 'thinkable or unthinkable' (Nettleton and Green,201 : 2 9). When habitus encounters a eld with which it is familiar 'it is like a " sh in water": it does not feel the weight of the water, and it takes the world about itself for granted' (Bourdieu and Wacquant, 1992: 12 ). Conversely, when habitus and eld are incongruent a 'hysteresis effect' (ibid: 1 0) disrupts the taken-for-granted nature of practice. Since habitus can improvise, readjust itself and override its primary dispositions (McNay, 2008a), the heightened re exivity manifested by this temporal lag between embodied dispositions and structural norms can generate change in practice.
With the exception of Hanckel et al. (2020) and Luca et al. (2019) who employ 'hysteresis effect' to understand how interventional logics clash with the logic of the individuals affected by them, to date, hysteresis is largely unused in health research. Following Hanckel et al. (2020) we argue that attending to such occasions allows habitus to be captured 'in motion', not least because the tacit nature of practice becomes noticeable when it is out of sync with the eld. To this, we add that while the destabilising effects of such moments of disjuncture can generate change and transformation, the heightened re exivity results in 'disquiet, ambivalence, insecurity and uncertainty' (Reay et al., 2009: 110 ). Such moments are therefore a vital way to understand class processes because they provide a platform from which to consider how the experience of capital accrual via a social prescribing journey differs according to social position. In the analysis that follows we attend to individual trajectories of cultural health capital accrual via a social prescribing intervention to capture the extent to which learnt dispositions change from place-to-place and from time-to-time, but are nonetheless related to a 'feel for the game' (Bourdieu, 1990: ).
Finally, echoing the sentiment of Mackenzie et al. (2020) we do not seek to criticise social prescribing per se, which is especially valuable insofar as it recognises that we access health as socially-located individuals. We rather wish to broaden our critique to account for the eld within which it is implemented and accessed in order to question the extent to which social prescribing alleviates the material and social factors shaping health inequalities. In what follows, we shine a light on that relational complexity to consider how class shapes client engagement with a social prescribing intervention. As social prescribing continues to be advocated both in the UK and internationally (Roland et al., 2020), exploring how it 'comes to seep into or saturate its context' (Hawe et al., 2009: 2 0) is timely and necessary. We now introduce the intervention and research methods in turn.

Method
This study is part of a larger evaluation combining quantitative and qualitative methods to evaluate the impact of social prescribing on people with Type 2 Diabetes (T2D) (Moffatt et al., 2018). The intervention under study is delivered in an ethnically and socially mixed urban area of the North of England. Clients are referred into the intervention from primary care, are aged between 0 and , and have at least one of eight qualifying LTCs. Once assigned a link worker, clients are guided through a pathway which is organised around an interventional logic which focuses on 'activating' and 'motivating' them to follow an action plan and work towards goals. Journeys with the intervention varied considerably and could last for up to four years (a lengthy period compared to other social prescribing interventions which are often much shorter (e.g. Pescheny et al., 2018)).
This paper draws on data generated from the ethnographic component of the research which explores how those referred in engage with the intervention and how it impacts on their lives. Our ethnographic approach allowed for multiple angles from which to 'see' the intervention. This is because interventions and contexts co-exist, interact and adapt over time (Hawe et al., 201 ;Orton et al., 2019;Shiell et al., 2020). Thus, by situating context as part of an 'open system' within which an intervention operates -rather than conceptualising context as an external factor (Barnes et al.,200 : 2 9) -we were able to more fully capture the ways the intervention 'couples and embeds' (Hawe et al.,201 : 10) with the clients' lives. Most importantly, by actually 'being there' for over 20 months, the eldwork generated an important temporal understanding of how the intervention unfolded in peoples' lives over an extended period of time (Reynolds and Lewis, 2019). Reynolds and Lewis suggest that evaluation research should prioritise the temporal over the spatial 'to 'stretch' the eld that is available to us, in order to explore the pasts, presents and futures that comprise the system of which the intervention is now part' (2019: 10). Such a perspective is particularly relevant here because it allows us to capture how clients' engagement with the social prescribing intervention is temporally connected to often competing priorities, the experience of which is intersected by personal pasts and presents.
In order to present a rich picture of the contrasting ways that social prescribing embeds in different contexts, here we work in detail with the temporal stories of four participants who were part of a larger sample referred into the intervention (n = 19). Our analysis of these four cases draws also from the whole dataset, and we use them to demonstrate how class shapes engagement and dispositions to invest in health. This focused approach has been used effectively elsewhere to delineate how variations in experience can be connected to social contexts (Morris et al., 2019). Class was certainly not the only social division shaping our participants' experiences. For instance, some participants experienced the double (dis)advantage of class and ethnicity or gender-based inequality.
The 19 key participants were purposively sampled for the ethnography in order to recruit a diverse group across age, gender, ethnicity, employment status, service provider, and duration with the intervention. Fieldwork was conducted by KG and entailed an initial interview (n = 19), photo-elicitation interviews (n = 9), interviews with family members (n = ), exit interviews (n = 1 ), and extensive participant observation. Participant observation included visiting participants' homes, meeting in coffee shops, joining participants in activities such as gardening, the gym, and social groups, and accompanying participants to meetings with LWs and visits to the foodbank. In total, this equated to over 200 h spent with participants and/or family members over a period of over 20 months (December 2018-July 2020). Additionally, with the participant's consent, the intervention provided data about clients recorded by LWs (e.g. notes made following meetings or telephone calls) (n = 1 ). The rst sixteen months of eldwork were conducted in person and face-to-face. From March 2020, eldwork, including exit interviews, was conducted over the phone or by video-call due to Covid-19 restrictions. Interviews were audio-recorded and transcribed. All participants and other people who are mentioned or observed are assigned pseudonyms. Durham University Research Ethics and Data Protection Committee provided ethical approval for the research.
Ethnographic knowledge is constructed in the eld; it is 'interpretive, emerging from social interaction and negotiation' (Prentice, 2010: 1 ). Typical of a grounded theory approach (Charmaz,201 ), data collection and analysis was an iterative process and systematically documented through re exive eld notes written directly after each research encounter. This was an integral tool for connecting existing theoretical knowledge with interpretations of contexts and identifying emerging empirical themes. On completion of eldwork, a further process of data immersion was undertaken. The research team, TP (an anthropologist), KG and SM (both sociologists), met regularly to iteratively review the coding framework, discuss emerging themes and develop the analysis. Analysis was carried out with the assistance of the data management software, N ivo 10, which was used to code, connect, store and retrieve the data. Data were analysed vertically and horizontally. ertical analysis allowed for the consideration of each participant's trajectory through the intervention, and horizontal analysis facilitated comparison of experiences across the sample. This analytical process has allowed for an understanding of differences and similarities as being related to contexts and structures (Dale, 201 ), resulting in the empirical abstraction of data into something of theoretical relevance. The participant stories we tell here are by no means representative of all clients' experiences of social prescribing. However, using habitus as a conceptual tool means that the research focus is broader than the focus of study because it allows us to question what is taken-for-granted (Reay,200 ). That is, this contextual and situated knowledge provides insight into how classed inequalities shape clients' experience of social prescribing, as the proceeding section will go on to explore via the stories of Andy, Geetha, Eddie and Tracy.

Andy
Andy, in his early 0s, lives with his wife and their adult son. He owns his home, a semi-detached house, which is situated on a quiet, tree-lined street overlooking a large garden. A former tradesperson, Andy went to university as a mature student, and has worked in the same company for over 20 years. During eldwork, he talks of taking early retirement in the coming months when his mortgage is paid off. It became clear that his paid employment was the source of much of his distress, when in our rst meeting he tells the story of a trip to the GP in late 201 when work had taken him to his 'boiling point'. The visit triggered a diagnosis of T2D, six months sickness leave, and a referral into talking therapies and the social prescribing intervention, which in turn referred him to the gym and nutrition classes.

I obviously got a fright because I think of diabetes and think, "Blimey, you're going to lose your feet," which can happen. So, I got a bit of a fright, and got back into training, got back into wellbeing, got back to the gym, got back to football, and I've been probably the fittest I've been for a long time.
Andy takes no medication for his T2D. Importantly, the diagnosis and proceeding support of the intervention acted as a 'kick-start' which activated and motivated him to re-engage with health maintaining practices, a point to which he frequently returns throughout eldwork: It's put [me]  Andy articulates a sense of control about his future as is evidenced by his plans for early retirement, and he needs little encouragement to engage with investing in his future health. The 'fright' of a T2D diagnosis -which he clearly positions as counter to his expected trajectory -and subsequent link into the gym appears suf ce to 'kick start' him 'in the right direction'. Andy is clear that the gym is a neat t with his habitus when he elaborates the role that tness has already played in his life course. Importantly, the intervention acts as a 'reminder', a 'memory' even, of what he 'should be doing'. Hence, his orientation towards investing in his health is positioned as 'obvious' and therefore 'thinkable' (Nettleton and Green,201 ), by drawing on his learnt disposition to invest in his body. The relatively harmonious transition back into the gym highlights the extent to which habitus is durable: he 'feels at home' there, and as such is a 'born-again gym bunny'. His journey through the intervention therefore starts from the advantageous position of having inherited health capital, which he is able to exchange for the opportunities on offer through the intervention. In addition, Andy's present context also works to enable his participation. He was able to remove himself from the cause of his health issues -his employment -and address them without any nancial consequences by taking six months sickness leave. Furthermore, his employer then facilitated his engagement in the intervention by being exible. In sum, Andy's present context and personal history create a scenario which increases his chances of doing well in the intervention.

Geetha
Geetha, in her late 0s and widowed, came to England from the Indian subcontinent with her husband after she nished university. Like Andy, she owns a semi-detached house on a quiet, residential street. Geetha was referred into the social prescribing intervention in late 201 , shortly after her T2D diagnosis. Like Andy, she clearly signi es that she does not take medication for her T2D, and is committed instead to managing it through 'diet control'. Through the intervention (she was discharged in late 2019), Geetha enjoyed a theatre trip and attending chair-exercise classes, which were hosted by the service provider. When the classes stopped, Geetha visited the provider of ces to enquire when they would restart. On discovering that the classes were discontinued, Geetha searched for alternative activities. Already regularly attending the gym, yoga, and a women's social group, she discovered a new walking group advertised in her GP surgery and encouraged her friends to go along. Such was their enthusiasm about walking, that Geetha and her friends started another walking group.
For Geetha, the intervention acted as a potential avenue to discover new activities, rather than a source of support: I don't need support because I find groups by myself, she said. In fact, it was Geetha's appetite to nd new activities which prompted her to 'join' the intervention in the rst place. She explained that she 'was really bored' and wanted to 'find out what's happening, what's going on around the area'. She wanted to occupy her time following her retirement from working in the voluntary and community sector ( CS). She often talked of searching for volunteering opportunities, for instance to support people with completing bene ts forms. She frequently drew on her previous employment to demonstrate her understanding of the intervention. For example, we bump into her former work colleague at the walking group: I tell her friend that I am a researcher following Geetha around. Geetha laughs and says to her friend 'she asked me who sent me here and I told her I came by myself. I've found everything by myself'. Her friend laughs and agrees, saying she used to send people to things like this and now she comes herself. I ask her how she came across the group and she explains that she just searched for local activities on Googleshe'd been to the swimming pool already today. Geetha says that there are lots of 'interventions', but not so many for older people. I'm surprised to hear her mention 'interventions'; I tell them that that's what I am interested in: interventions.
Field notes: Walking group with Geetha Geetha attended a range of activities alongside a close-knit female network who all displayed a number of middle-classed markers. KG was always introduced by Geetha as her 'friend from the university' to her friends, who often discussed their children and grandchildren who attended private schools or university, or were in professional occupations. Sharing food was often central to Geetha's social activities, as is customary in diasporic communities (Abbots,201 ). At the start of 2020, when Geetha is unable to leave the house following minor surgery, her friendship group proves invaluable in mitigating some of the effects of this disruption. Unaccustomed to being 'stuck in the house', Geetha had a stream of visitors, all of whom gifted her foodstuffs which topped up the meals she had bulk cooked and frozen prior to her operation.
Like Andy, Geetha displays an alignment towards the future. She draws on classed and gendered dispositions to plan for the impending disruption of surgery by forward-planning food and she repeatedly performs a commitment to investing in her health and wellbeing through her proactive searching for additional activities. Her autonomy in pursuing these activities is partly enabled by her possession of capital and her present context of recent retirement, which has accorded her with temporal freedom. Most notable is the role that her immediate social network plays. Importantly, in addition to providing a support network built around reciprocity and sociality, Geetha's social ties act as a form of classed capital, which she is able to mobilise to access further opportunities. For example, walking groups have a positive impact on her health and wellbeing. Furthermore, Geetha's past employment plays a central role in equipping her with the knowledge to domesticate aspects of the social prescribing intervention to complement her knowledge of the CS landscape. She was clear in her conversation with her former colleague that her critical perspective about 'interventions' was one of a professional, not a client. This brings us to our nal point which is that, like Andy, Geetha's personal history and current context created the social conditions of possibility for a smooth trajectory within the intervention. However, for Geetha this manifested itself as a form of classed distinction and critical distance from being a service user.

Eddie
Eddie, in his late 0s, spent much of his childhood in care. He lives alone in a one-bedroom at, which is situated in an island of social housing surrounded by wasteland. Since leaving school at 1 , he worked in various jobs until 201 , when, following a relationship break down, he suffered from depression and anxiety for which he was referred into talking therapies and the social prescribing intervention. Eddie has had T2D since 200 . He manages his diabetes through a combination of medication and by trying to 'eat the right foods', although sometimes 'it gets on top' of him because the 'right foods' 'are expensive to buy'. Eddie rarely refers to his T2D during eldwork, except to share an ironic remark about whether he will abstain from having chocolate sprinkles on his cappuccino when we meet for coffee. His mental health and living in poverty are much more pressing concerns for him.
Eddie's social prescribing trajectory has involved a number of activities and referrals. At rst, these concerned his T2D and mental health issues, but increasingly became about addressing his nances. For instance, the focus of his second link worker meeting was on 'motivating' him to make healthier food choices and the link worker later sent him a food diary and some recipe ideas. Eddie was referred to a local gym, however (and despite several text reminders), he never went. He was encouraged to attend a social group, but after a series of cancellations, the link worker concluded he was 'uncommitted'. Eddie was then referred for counselling sessions, which he regularly attended.
Eddie's photo (Fig. 1) and accompanying words emphasise the weighty reality of his everyday life:

Sometimes when I'm out, you see, I don't like to look ahead. I look down. I don't know why.
The bene ts system appears to play a central role in a shaping Eddie's 'temporal outlook' (Warin et al.,201 ). The excerpt below alludes to his sense of powerlessness:

I can't do anything at this time until I hear from these people [Department for Work and Pensions, a UK government department responsible for welfare and pension policy]. Like, I'm in limbo, you know, I'm not one thing or the other, so I just have to wait until they come with their decisions on yes or no. So, it's like they're controlling my life at the moment.
As eldwork unfolded, Eddie starts to contact the intervention requesting foodbank vouchers on an almost-weekly basis. With the threat of eviction, he becomes increasingly reliant on foodbanks and every time he meets KG he arrives with a shopping bag neatly folded in his coat pocket, a strategy which KG later learned enabled him to avoid carrying a bin bag full of food home. Ultimately however, the intervention questions the frequency and long-term usage of the vouchers and appears reluctant to issue the vouchers so unequivocally. The eld notes below detail an occasion when Eddie's link worker telephoned him after he contacted the intervention requesting a voucher having just collected one the week prior: He looks at the screen, holds up his finger as if to pause our conversation and says it is the link worker. "Actually I don't like going to that one, I usually go to the one up the top … ok, see you in a bit." He tells me with a raised eyebrow that the link worker wants to meet to look over his expenditures with him after the Debt Relief Order (DRO, a means of writing off debts of less than £20,000) is sortedhe doesn't look too pleased. For now though he can go up to the office and get the foodbank voucher, he says.
Field notes: meeting Eddie for coffee Eddie claimed 'Universal Credit', a state bene t which is subject to annual review. With the threat of a bene ts review ever-present, Eddie deploys his limited resources to ensure he collates enough evidence to ensure his claim is successful. For instance, he contacts the intervention to request a support letter, follows up his second talking therapies referral and talks of having to 'put on an Oscar-winning performance' during the bene ts review. The terrifying reality of appealing is all too clear when Eddie attends an appointment at the Citizens Advice Bureau (CAB), a service which offers free advice about nancial, housing and debt issues, to process his DRO. There is a limit to his proactivity:

She [the financial advisor] starts to methodically go through the paperwork pausing to ask for more detail when necessary. Eddie's voice starts to crack a little and she asks if he's ok. 'I'm ok', he says. 'I just suffer from anxiety and depression and it's all getting on top of me', he explains. She reassures him she is nothing to be intimidated by and makes a joke about her terrible spelling.
She asks him if he gets PIP(Personal Independence Payment, available for claimants who can evidence that they cannot work due to ill health). He tells her it was rejected. She sighs and sits back in her chair. 'Did you not appeal?' Eddie tells her that when he read that it was refused, he threw the letter in the bin. She shakes her head and works out that he has until February left to appeal. 'I'm not doing that again' he says.
Field notes: Debt advice with Eddie at CAB Before, during, and after this interaction Eddie's anxiety and shame was evident. For instance, in the meeting he apologised profusely for forgetting his bank statements and chastised himself for being stupid. Afterwards he said he felt 'ashamed and embarrassed'.
Eddie's past, one of trauma and child abuse, provided him with scant inherited capital to exchange for the health opportunities offered by the intervention. Furthermore, Eddie's present context of poverty supersedes his inclination to invest in his future health. As such, the reactive strategies he deploys concern ways to acquire economic capital to alleviate his present context. Eddie's present shapes a temporal outlook of future uncertainty; for instance, his experience at the CAB delineates this fear, as does his photo of his feet. Like his present context, his imagined future is precarious, which is largely due to a bene ts system which xes him as powerless, or 'in limbo'; it relies on him evidencing his lack. Importantly, Eddie does not lack knowledge: for example, he uses humour to mark his re exive awareness of the lack of health credentials attached to chocolate sprinkles. Nor does Eddie lack motivation. Contrary to his link worker's view of him, he is highly proactive and he strategically domesticates the intervention in order to make it work for his circumstances. That is, Eddie's tactical management of the intervention focuses, not on investing in his long-term health, but rather on navigating through the immediate requirements of poverty via shortterm practices of survival. In short, the uncertainty and lack of capital identi ed in Eddie's past and present shape a future orientation which appears at odds with the logic of the intervention.

. Tracy
Tracy lives with her partner and her partner's daughter; they rent a two-bedroom ground-oor at on a steep terraced street. Tracy has multiple serious LTCs, was diagnosed with T2D in 201 and is awaiting major surgery. She takes 1 tablets a day for her LTCs, but her 'goal is to come off all of them'. Until 201 , Tracy worked in a variety of jobs including factory work and cleaning. Tracy's depression and ensuing health issues were, she says, triggered by the sudden death of her mum. She no longer has any contact with the remainder of her family who, she feels, blamed her for her mum's death. She explains,

I was st until my mum died, and because of what happened with them in [place name], and pushing me out and what have you, I used to eat lots of chocolate. At least 0 bars of chocolate every day … …Not a meal, just chocolate, until I was sick of it, and it was every day. How I afforded it, I don't know, but that's when it went bump, bump, bump, and that's why I'm like this.
Tracy frequently talks of confused interactions with health professionals and often mentions her uncertainty and anxiety around her impending surgery: 'I'm nervous about the operation', she says uite abruptly. I ask her what the operation involves. She says she's not sure but prior to the operation she has to eat a special diet for her kidneys. I ask what that involves and she tells me that she doesn't know but has to go to a seminar in a few weeks to find out … …'My problem is that I'm carrying this because I lost

Fieldnotes: Sitting on a bench with Tracy
Nonetheless, Tracy meticulously sticks to her pre-surgery diet regime and loses . stone. She tells KG on four separate occasions that she was never issued a target weight by her dietitian because she reached it before her surgical referral.
Tracy was referred into the intervention in 201 and has been linked into several services and activities, including debt advice, a support group and a local women's charity. Additionally, following a heart attack in 201 , Tracy was referred by secondary care into a local health centre where she enjoys attending the gym and has also accessed a rst aid course, mindfulness sessions and nutrition classes. Below is an extract from a conversation KG had with Tracy about the nutrition class: I ask her if she thinks she might cook any of the recipes and she tells me she will try the frittata. 'Will you cook it for [partner or partner's daughter]?' I ask. At that suggestion she laughs -'all they eat is food from packets and boxes, nothing fresh; I don't know why, they've just always done it'.

Fieldnotes: At the cafe with Tracy
Tracy often spoke about a local women's centre, but said that she never 'dared' go in because she was 'too nervous'. She was particularly excited when a link worker accompanied her to a cooking class there; however, her negative experience of attending alone the following week discouraged her from going again. She explained, 'I haven't got anyone to go with, and everyone else seems to know someone, so I don't know what to do with myself.' Tracy then joined a gardening club at the centre; KG met her after her rst session: [ olunteer] tells me that Tracy has been helping her get the garden ready. Tracy, who still hasn't spoken, says that it's difficult because she doesn't know which are weeds, and she's frightened to pull the wrong thing up. I tell her she'll soon get the hang of it.

Field notes: Gardening with Tracy
A bout of illness caused Tracy to miss the next few sessions and she never returned, explaining it was 'too difficult to go back'. Despite a number of setbacks, like Geetha, Tracy talks often of needing to 'keep busy' and is understood by the link worker as being 'highly motivated'. For instance, she buys a swimming costume ahead of a new swimming pool opening in the area. Her plans are thwarted however, when minor surgery triggers a number of further hospital stays. Sadly, Tracy's setbacks continue; the most signi cant of which is the Covid-19 pandemic resulting in the gym closing down, her social group discontinuing and the postponement of her surgery.
Arguably, Tracy brings limited inherited capital to the intervention; occasionally, she mentioned that she had been previously 'active' and played sport. However, unlike Andy who appeared unre exive in drawing on his embodied tacit knowledge to make sense of his current health practices, there is a tone of insistence in Tracy's accounts about her health. Perhaps indicative of the disruptive effects of the intervention, she appears implicitly aware that her habitus is discursively marked as lacking value. Her orientation towards the future is also insistent and tenacious, albeit within a context of uncertainty. What is particularly striking in this regard is that Tracy does prioritise and is 'motivated' to invest in her long-term health, and is driven to 'play the game' by ostensibly deploying her limited resources. Signi cantly, this seems to be without the support of any social capital. Tracy's response to KG's suggestion that she cook her partner frittata is met with laughter, signifying that her investment is unthinkable and at odds with her household norms. Also signi cant is that Tracy's attempts at engagement are thwarted time and time again. Clearly, her LTCs cause Tracy several setbacks. But importantly, these LTCs are the result of a vicious downwards cycle triggered by the sudden death of her mum (a biographical disruption which she appeared ill-equipped to recover from). As such, Tracy's personal history literally marks her habitus in such a paradoxical way that it both de nes and interrupts her future journey to better health. As she says, her past is with her all the time; it is embodied and visible on her habitus in the form of multiple LTCs. Equally important is the role that uncertainty plays in Tracy's discomfort at various junctures during her social prescribing journey. Her experience of the 'hysteresis effect' highlights how dif cult it is for her to enter unfamiliar elds of practice. While the presence of a link worker in new settings enabled her initial transition there by alleviating the discomfort of unfamiliarity, it was insuf cient to embed her in a new position, such that she never quite gets 'the hang of it'.

Discussion
Through the stories of Andy, Geetha, Eddie and Tracy we have shown how habitus is intersected by past and present experiences which in turn shape individuals' future transitions through a social prescribing intervention. By focusing on the extent to which their classed habitus aligns with this particular intervention, we have illuminated how the possession of capital enables a relatively smooth and straightforward trajectory to better health because it creates the conditions of possibility to engage with the intervention. We thus complicate the assumed linear homogeneity of an interventional logic which is organised around 'activating' and 'motivating' individuals to invest in their future health, and instead we account for health practices as situated in a 'discontinuous, patchy space of practical pathways' (Bourdieu, 1990: 8 ). This analysis challenges the claim that social prescribing can reduce health inequalities, suggesting that instead it has the potential to exacerbate existing inequalities.
Our inclusion of participants in relatively privileged classed positions is crucial for advancing an understanding of health inequalities. Research persistently looks towards marginalised communities to explain inequalities (there are limited exceptions, e.g. Nettleton and Green, 201 ;O'Donnell, 2020;and Wiltshire et al., 2019 offer cross-class comparisons). This continued scrutiny on 'deprivation' ignores the pathways through which capital is accrued thus obscuring how socio-economic and cultural inequalities enable access to health. Like other participants with similarly high volumes of capital, Andy and Geetha's 'distance from necessity' (Bourdieu,198 : 1 ) places their future 'within easy reach' (Warin et al.,201 : 09). As such their practical logic aligns neatly with the intervention's future-focused logic. Through familiarity and recognition they mark themselves as being actively committed towards investing in their future health and wellbeing. In fact, all participants in this study recognised, and therefore tacitly naturalised, the idea that health is a matter of individual investment with little recourse to social context. But as we have argued, some participants' challenging immediate social circumstances took priority over investing in their health. This is not to say that Geetha and Andy experienced no setbacks. However, the negative effects of their disruptions were mitigated by their portfolio of capital, meaning they had more autonomy to adapt to their circumstances. In Andy's case, his stable employment situation accorded him the temporal and economic freedom to invest in his health. This follows O'Donnell's observation that individuals in higher socioeconomic positions have more autonomy to 'remove themselves from the conditions giving rise to their distress and move into a social space where more health-enhancing behaviours were possible' (2020: 1). For Geetha, the disruption of retirement created possibilities to engage in further activities, which she was able to access both by deploying her knowledge of the local landscape and with the support of her immediate social network. As McNay argues, Andy and Geetha's social position accorded them with 'the objective ability to manipulate the potentialities of the present in order to realize some future project ' (2008b: 281).
In part, our ndings, such as that of Eddie's priority to survive his 'living present', echo the orientations to practice identi ed by Warin et al. (201 ) in their ethnography in a disadvantaged community in Australia. Warin et al. develop the concept of 'short horizons' to encapsulate how strategies deployed by participants to cope with living in poverty are shaped by 'narrowed vistas of possibility' (ibid: 10). As they nd, ''Living poor' illustrates what is possible within the constraints of short horizons, the improvisations that people use in their day to day lives to deal with living presents, rather than anticipated futures' (ibid: 1 ). Sadly, Eddie was not alone in this regard. Several participants living on state welfare bene ts were more inclined to concentrate on the immediacy of their precarious social circumstances, often at the expense of health-enhancing practices. Importantly, contrary to this intervention's logic (and more broadly, policy rhetoric around social prescribing which implicitly focuses on behavioural change), it is not that Eddie, and people in similar class positions, lack 'motivation', lack 'personal resilience' (NHS England, 2019a) or have 'low activation' (NHS England, 2019d). Just as Geetha and Andy proactively accrue health capital, Eddie is highly proactive in his orientation to accrue the required economic capital to alleviate his current context of poverty. Like the many creative and improvisational strategies carried out by Warin and uvkovic's participants which allowed them to get by in circumstances of disadvantage, the short-term nature of Eddie's investments were at odds with the 'synoptic time of public health futures ' (2019: 19 ).
Nonetheless, it is important to note that while some participants were constrained by short horizons, it is not that they did not look towards the future. If habitus generates a continuum of possible social trajectories (Reay,200 ) then it is in this regard that our example of Tracy is especially important. Like Geetha and Andy, Tracy is orientated towards investing in a future healthy self. However, her future-orientated practice is conducted in the context of uncertainty and her journey to better health is truncated by a series of setbacks. For Tracy, it is not that investing in an abstract future is 'unthinkable', but rather she engages with this investment from a disadvantaged position, meaning that her efforts are often thwarted. By paying attention to Tracy's experience of these encounters, or shifting positions, we highlight how habitus becomes disrupted when it nds itself in situations to which it is not accustomed. It is notable that Geetha and Andy were relatively unre exive in mobilising their pre-engrained ideas about health and wellbeing. Already positioned well in social space, their transition to new groups and activities appeared ostensibly unquestioned. As Bourdieu (1990) reminds us, when embodied dispositions align with positionality they operate as an unrecognised practical consciousness. In contrast, Tracy, Eddie and several others in similar class positions did not display such security. Their social prescribing journey was in a context of uncertainty and interrupted by several moments of hesitancy, disquiet and heightened re exive awareness that their habitus was ill-tting with its position in the eld. For instance, one participant, who was discharged due to lost engagement, talks at length about the anxiety she experienced around meeting her link worker. For her, even engaging with the intervention was 'unthinkable'. Attending to these moments of disjuncture and tension is especially important for our understanding of class in relation to engaging in social prescribing. Luca et al. employ hysteresis effect to understand how interventions can create 'small disruptions' leading to changing health dispositions (2019: 1 ). Likewise, Hanckel et al. note that interventions aim to 'disrupt and change what are framed as 'unhealthy' dispositions' (2020: no page). However, by accounting for how such disruptions are experienced, our ndings shed light on how inequalities interact with people's trajectories of change. For Tracy, and many others who engaged with the intervention from precarious social positions, this experience at times led to non-engagement. et, contrary to being understood as passive resistance or lack of commitment, non-engagement could be usefully accounted for as a 'hysteresis effect' associated with shifting positions. As the experience of Tracy suggests, attending to such disruptions requires a considerable allocation of link worker time. In this paper, there has not been the scope to explore temporalities shaping the client-link worker relationship. Future research could aim to further understand how the presence of a link worker might facilitate social prescribing journeys by bridging the upward ascensions, cushioning disjunctures and, through continuity, suf ciently embed clients into new encounters.
We suggest that if social prescribing interventions work on the assumption that everyone has access to the capital required for health investment, then it risks exacerbating health inequalities. Instead it should account for the socio-temporal contexts shaping orientations to practice. This means accounting for habitus as an embodiment of the very social relations social prescribing seeks to address. Doing so, means that non-engagement is understood, not as lack of individual motivation, but as the result of habitus being misaligned with the social position required to invest in future health. Without properly accounting for the symbolic and material con gurations shaping accessibility, social prescribing interventions risk sustaining a 'fantasy paradigm' which positions health inequalities as eradicable via local interventions aiming to change individual behaviour (Mackenzie et al., 2020;Scott-Samuel and Smith, 201 ). Echoing Phelan et al. (2010) we underline a need to develop interventions which do not require resources, or at the very least minimise their relevance, and which therefore can be broadly distributed and accessed so as not to perpetuate existing health inequalities. Granted, it is unrealistic to suggest that it is in social prescribing's remit to dismantle intrinsically legitimising ideology which situates health as an individual project irrespective of social context. However, if social prescribing is to address the social determinants of health it must recognise and actively problematize these social relations at play. Failure to do so, has important implications for the reproduction of inequality. It masks the effects of class by discursively positioning those without access to the legitimated capital required for engagement as individually and morally failing to invest in their health and wellbeing.

Conclusion
In some respects, the intervention worked for all of the participants we report on here -even Eddie who was able to utilise social prescribing to access the foodbank. However, while the intervention assisted with negotiating the social determinants of health, it did not, and indeed could not, remove them. In this way, our ndings demonstrate the problems created by the individualisation of understandings of social inequalities in health within the health sector (Mead et al., 2020) and the implausibility of addressing health inequalities via an intervention which emphasises individual lifestyle change (Mackenzie et al., 2020).
This paper extends this analysis by exploring how classed inequalities shape clients' engagement with a social prescribing intervention in the North of England through the contrasting examples of four participants. We have shown how for clients, social prescribing entails a trajectory of social positions across the eld of health, the experience of which is related to habitus and its related volume and composition of capital. Our ndings show that clients who are familiar with the process of cultural health capital accrual fared well in this intervention. In contrast, most disadvantaged participants experienced multiple points of tension and disjuncture. Our explanations thus provide insight into the nuanced ways that structural contexts relate to social prescribing.
Too often is there an underlying assumption in health interventions that individuals are homogenously predisposed to investing in their future health. Here, we have demonstrated the importance of scrutinising the ways in which the context and circumstances of people's lives shape their interaction with a health intervention. Rather than presenting the achievement of better health as an individualised project, health interventions must account for and be exible to the effects of the underlying social inequalities in uencing the accessibility and thinkability of health practices. This involves providing structural opportunities which enable and support individuals to acquire other forms of capital which in turn can be exchanged for health opportunities.
Classed responses to social prescribing exacerbate the inherent contradiction in attempting to tackle structural inequalities through an individualised intervention (Mackenzie et al., 2020). To this end, like Mackenzie et al. we call for a 'de-coupling' (italics original) of the public policy aspiration of reducing health inequalities from the operationalisation of social prescribing. Furthermore, we argue that political claims regarding social prescribing's capacity to reduce health inequalities represent a fundamental contradiction in UK public policy. At a rhetorical level, this seemingly ubiquitous concern for health inequalities in health policy demonstrates a commitment to alleviating inequalities. In reality however, this commitment, or 'fantastical vision' 201 : 20) is, in social prescribing, a mirage that dissolves under close scrutiny, and is not mirrored in broader policy that could effectively tackle structural inequalities.