Drug fatalities and treatment fatalism: Complicating the ageing cohort theory

Deaths related to drug ‘misuse’ remain at an all- time high in the United Kingdom (UK). Older heroin consumers are particularly at risk, with the highest rates of deaths among people aged 40– 49 and the steepest rises in the over- fifty age bracket. Accordingly, a popular theory for the UK’s increase in drug- related deaths, made by the government, and propelled in the media, is that there is an ageing cohort of heroin users with age- related health complications predisposing them to an overdose. However, drawing on in- depth interviews with those people deemed to be most at risk, this article works to complicate this theory, with participants citing a shift in (a) experience and responsibility, (b) route of administration, (c) desired effects, (d) acceptance of their drug use and ‘user’ status and (e) valuing health. Disrupting age as a given risk factor, this article turns attention away from the individual and these ‘natu-ral’ processes to what participants describe as a singular, punitive, and inflexible treatment system and its intersect-ing structures. Approaching


INTRODUCTION
Since 1993 (when the recording series began), there has been a threefold increase in deaths related to drug poisoning in the UK, with rates rising rapidly between 2012 and 2015. While recent statistics have shown some stabilisation in younger age brackets, for people aged fifty and over, death rates have continued to rise (ONS, 2020). This has led to a popular theory, proposed in several government reports (e.g., PHE, 2016PHE, , 2017, as well as academic papers (e.g., Gao et al., 2016;Pierce et al., 2015), that there is an ageing cohort of heroin users with age-related health conditions who are inflating these statistics due to a predisposed risk of dying from an overdose. Worryingly, however, seeing these deaths as age-related risks de-politicising them as something that are naturally occurring, distancing them from the realms of state and societal intervention. For example, responding to how this upsurge in deaths has coincided with a shift in national policy 1 (Home Office, 2010, 2012, Public Health England, perhaps defensively, states: '[we] did not establish a direct relationship between the policy focus on recovery and DRDs [drug-related deaths] ' (2016: 15). With this, the ageing cohort theory pushes the site of responsibility (and even blame) onto the individual -the irresponsible drug user who should really know better (especially at their age!). It is an all too easy link that neglects and flattens out different experiences that suggest age could function protectively (Drug & and Alcohol Findings, 2017). Complicating this picture, the study here draws on accounts from older people who use heroin to both disrupt age as a given risk factor and relocate this risk elsewhere.
While there is no doubt that older people are dying in higher numbers than younger people, this is by no means a simple matter of biological ageing and its associated illnesses, but rather can be complicated by what participants in this study describe as treatment despondency and fatalism. Participants describe feeling unheard, ill-understood, stigmatised and even punished, which means they engage only superficially -hiding their illicit drug use or saying things to please treatment workers and systems -or leave altogether because the 'fight' (as one participant explains below) for the treatment they want has become too onerous. Here, this weariness manifests as an embodied lack of care or fatalism at having things done to them, summed up with the recurring statement: 'it is what it is'. With this, the study builds on a long line of research in critical drug studies that demonstrates how drug treatment systems get under the skin and constrain agency (most notably, Bourgois, 2000;Fraser, 2006;Fraser & valentine, 2008;Harris and Rhodes, 2013), especially as this most recently relates to abstinencefocused recovery models (see, e.g., Fomiatti et al., 2019).
Taking up care as a lively concern for being that brings matters of life and death together with meaning and value (Mol, 2008;Mol et al., 2010), it is crucial that treatment is made more meaningful and responsive to older service users to prevent this fatalism that can only too easily tip over into fatality. At its most basic, we know that people receiving opiate substitution treatment for heroin dependency are less likely to die from an overdose than those who are not (Alho et al., 2020;Hamilton & Stevens, 2017;Pierce et al., 2015;PHE, 2016PHE, , 2017Sordo et al., 2017;White et al., 2015). Following a feminist ethic of care as a material and affective doing and obligation (Puig de la Bellacasa, 2017), it is vital, therefore, for treatment to register and respond to older people's lives made with substances and their reasons for using them in order to sustain and improve these ways of being. This extends a burgeoning area of critical drug scholarship on care as a more-than-human practice in acknowledging the role of illicit and licit drugs in people's situated modes of living and caring for themselves and others (see, e.g., Dennis, 2019a;Dennis & Farrugia, 2017;Duff, 2015;Duncan et al., 2019;Farrugia et al., 2019;Gonçalves et al., 2016;Race, 2018;Rhodes et al. 2019;Van Schipstal et al., 2016;Vitellone, 2017). Care, as I propose here, is about striving to know and nourish these existing modes.
There are likely many factors in this rise with the principle one being an ageing cohort of 1980s and 1990s heroin users who are experiencing cumulative physical and mental health conditions that make them more susceptible to overdose. (2016: 14) In that same year, the government's independent Advisory Council on the Misuse of Drugs (ACMD) also state: The ACMD can assert with a good degree of confidence that the ageing profile of heroin users with increasingly complex health needs (including long-term conditions and polysubstance use), social care needs and continuing multiple risk behaviours has contributed to recent increases in drug-related deaths.
(2016, no pg.) These findings are again, more recently, reiterated by PHE: An ageing cohort of 1980s and 1990s heroin users is now experiencing cumulative physical and mental health conditions. Older heroin users also seem to be more susceptible to overdose because of long-term smoking and other risk factors. (2017, no pg.) This ageing cohort theory has proved particularly popular with the media, who have dubbed this group of heroin users, now in their 40s and 50s, the 'Trainspotting generation', after the famous book and film following a group of young heroin users in Edinburgh during the 1990s heroin boom. Reigniting its 4 | DENNIS pertinence, these increases in deaths coincided with the release of Trainspotting 2 in January 2017, making such headlines an instant success in broadsheets and tabloids alike. Headlines included 'Trainspotting generation most likely to die from drug use' (The Guardian, Siddique & Perraudin, 2017) and 'Trainspotting generation drives record drugs deaths' (Financial Times, Cornish, 2017). Such headlines then returned the following year to reflect local disparities: 'Trainspotting generation of heroin addicts dying in Britain's seaside towns' (Daily Mail, Riley, 2018).
As a necessary rebuttal, this article supports a wariness towards such a theory that could naturalise these deaths (implying that little can be done; see, e.g., Eastwood, 2019), place blame on the individual (Stevens, 2019), silence the complexities in people's experiences that might be working to protect them from harm and overdose (Drug and Alcohol Findings, 2017), and turn attention away from wider structural and societal issues such as austerity-driven cuts (Drummond, 2017), the recommissioning of drug services (ACMD, 2017), growing health inequalities (Hamilton, 2020;Heyman et al., 2019;Liddell, 2019;McPhee et al., 2018), moralistic drug policy (Stevens, 2019) and global prohibition, making heroin purity levels unpredictable (as seen in HM Government, 2020).

AIM, APPROACH AND METHODS
The interviews drawn on in this article form part of a larger multi-modal ethnographic study, which seeks to explain and explore drug-related deaths as a sociomaterial failure to matter. Following a feminist technoscience approach to care and responsibility, 'matter' and 'mattering' are understood as entangled social and material processes that inextricably tie 'of matter' to 'to matter' (Barad, 2007;Haraway, 2016;Puig de la Bellacasa, 2017). This is not to say that social factors simply interact with biological ones, nor is it to say that biological matters are constructed by social forces, for example, in a social construction of drug-related deaths (e.g., Cruts, 2000). Rather, these social processes are already embedded in the biological and vice versa (Haraway, 1991). Donna Haraway's (2016) notion of response-ability is particularly important to this project (explored further below) for how it implicates our ability to respond to people's lives as a fundamental part of what they can become. Approaching life and death in this 'lively' way, then, invites a more critical account of our knowing and responding practices that are inherently implicated.
To contextualise the aim of the interviews, it is necessary to show how they fit within the wider study, which commenced in 2018 and is still ongoing. The study is three-staged. First, I carried out observations and in-depth interviews at care sites in London, where people who use drugs are 'made to matter' (see, e.g., Dennis, 2019b). Second, I explored where mattering practices may be 'failing' by conducting interviews at a drug service, and observations and interviews at an organisation that advocates for people who use drugs (see, e.g., Dennis, 2020). Third, I seek to intervene in these practices by enacting alternative ways of performing care through creative workshops and exhibiting the outputs. Centring on participants' experiences of overdose, I focus here on the second phase of the research.
From fifteen interviews conducted with current heroin consumers, twelve were with participants aged over forty. From this group, I interviewed eight men and four women, aged forty-one to sixtyone, who identified with a range of ethnic and social groups. Most participants smoked heroin but would also sometimes inject. Participants were recruited via peer networks at a south London drug service. The interviews lasted approximately 1.5 h (ranging from 40 min to 2 h) and followed a loose topic guide covering drug use histories, current drug practices, experiences of harm and overdose (direct and indirect), and treatment experiences and 'goals'. The interview transcripts were analysed using grounded thematic techniques. Ethical approval for this study was granted from the University and NHS ethics boards. 4 | 5

DRUG FATALITIES AND TREATMENT FATALISM
In what comes next, I draw on these interviews with service users to show how, contrary to government reports and growing popular opinion, age may actually be working to protect them from overdose-related deaths. With this finding, I, instead, turn attention back onto our treatment systems that may be part of a wider failure to respond to the lives of older people who use drugs.

Questioning age as a given risk factor
Most of the participants in this study had been using heroin for more than twenty years, since their late teens or early twenties. What became clear is that they saw these early years as the dangerous years and now actually felt safer. A past, almost ubiquitous occurrence of overdose gets summed up in James's 5 explanation of a need for 'an agreement': We had an agreement among each other. You know, when people hear about drug users putting people out in the stairwells they think 'oh that's horrible', but the thing is, you've got to look at it as, when you OD [overdose], the last thing you want to do is get your friends in trouble, so we had an agreement.
Participants had lived through these dangerous times, where it was felt necessary to have an agreement in place for what to do in the event of an overdose. With this, James flips the usual moralising gaze towards 'the friend' onto the legal structures that meant the overdosed person in the stairwell may too have agreed to this unjust situation where calling an ambulance could trigger a law enforcement response. Participants describe now feeling distanced from this close proximity to death and the brutal pragmatism it demanded, citing a change in their (a) experience and responsibilities, (b) route of administration, (c) desired effects, (d) acceptance towards their drug use and 'user' status and (e) valuing health.
Beckie says: In the early days of my addiction I couldn't give a fuck, but now I'm a bit more old and a bit more mature. For Beckie, now in her mid-forties, she has responsibilities, which means she is careful with how much heroin she uses and how frequently. She speaks about her heroin use like one might talk about other recreational activities, only allowing herself to 'play' once she has attended to her responsibilities and has the time and money to spare. Drawing on her devastating experience of having her child removed by social services she describes becoming determined to care for the other 'babies' in her life ('I learnt that the hard way').
[…] You can never get enough in you in one go to put you over'.
After many years of injecting, participants often found it hard to locate and raise a viable vein for injecting into, while fragile capillaries could easily burst. But, as Andy points out, although this change to smoking may have initially been due to necessity, it was continued due to a new, learnt desire for its slower, more relaxed pleasure: 'I've lost all my veins, and I kind of enjoy smoking anyway, I kind of like it, it's a bit more relaxed'. In this sense, it is not necessarily that an older, responsibilised drug user desires the gentler, less-risky route of smoking, but that the smoking itself engenders this desire. There is a link here between the route of administration and the responsibilisation it affords, where the purpose, method and effects become co-producing.
James contrasts these new, calmer practices with the fast-paced urgency of his younger years: Whereas when I was really bang at it and injecting, it would be around at a mate's house. There'd be three or four of us and we'd be throwing our money in together. We'd all be getting in as quick as possible and then sitting back. Then it'd be the next hit. Whereas now it's just much more… I think my drug use is way more civilised now.
Drawing on notions of civility and citizenship, James, like Beckie, highlights the difference between then and now. It is no longer only about the speed in which one reaches the end result (injected straight into the venal system), which is seen as an almost immature and crass interest ('okay it's in your system or whatever'), but, much like a connoisseur of fine wine, there is new-found enjoyment in the drawn-out process of preparation, ingestion and taste: I suppose, and I enjoy the relaxation process of it. To inject it, like, okay it's in your system, or whatever, but there is nothing else to do, that's it. I kind of enjoy the process of smoking it.
(Andy, 51) Smoking is seen here as a more skilled and creative practice that allows time for the user to appreciate the drugs' effects and what it might enable, in the body but also socially.
A fourth way participants like Beckie, Andy and James felt they were now more distanced from a drug-related death was that they had come more to terms with their drug use and 'user' identity. Andy says: 'I guess, I'm 51 now and maybe I'm coming to terms with it a bit more, you know?'. For James, he had got bored of going 'around and around on this wheel' of abstinence-work-relapse: Becoming abstinent, getting a job, relapsing and went around and around on this wheel for about 10 years until probably four or five years ago the penny dropped. I don't know why, but it was, 'I'm not doing any harm to anybody. I'm not a thief, what's the problem?' It was like a weight lifted off my chest.
With this, James, like others, decided to continue taking heroin or, rather, to not try so hard to stop (which had caused self-esteem and confidence issues). This acceptance that some participants had reached was felt to be protecting them against these social and emotional harms, as well as the physical ones, including overdose (e.g., following periods of abstinence, see ACMD, 2019; Bukten et al., 2017;Strang et al., 2003;Strange, 2015). Therefore, by seeking stabilisation in their opiate use, participants sought to avoid these ebbs and flows in tolerance levels and emotional wellbeing.
Following a similar sentiment, participants spoke of their drug use as part of a repertoire of healthvaluing and promoting practices: 'Using the [cannabis] oil, I know it's helping me stay well. I do batter my body, but I know that I have to repair it too' (Kimona, 50). Speaking more closely to the pleasures | 7 DRUG FATALITIES AND TREATMENT FATALISM of these substances that can be forgotten in a treatment system focused on reducing harms, especially as it relates to drug-related deaths, Sofia (56) says: But the thing is, they don't understand that people do not indulge to do harm to themselves, they indulge because they actually like it, so the last thing that they want is to actually kill themselves.
In fact, participants strove to get assurances on their heroin that an illegal market makes difficult: 'The overdose was purely because he didn't know what he was buying' (James, 51). For Beckie, this meant keeping the same dealer to avoid 'deals' that were dangerously adulterated: With age and experience, then, comes a number of factors that may actually be working to protect older people from drug-related overdoses and dying prematurely. But, if, as we know, this risk still exists, then, where might it be coming from?

Re-locating the risk
You know, you can listen to Public Health England and they say it's because we're all getting older and we all, kind of, have all of these other issues, like COPD [chronic obstructive pulmonary disease] and everything else, which is partly true, of course. But, I think there is another reason as well, and I think it's the conversation that we've just had, with, you can't really be open.
Summed up beautifully by Andy, the rise in drug-related deaths among older people is about more than age. There is a distinct lack of treatment options and flexibility that is manifesting as an inability to respond to older people's needs, leaving them feeling silenced and closed off ('you can't really be open'), giving way, in some cases, to despondency and a sense of fatalism. This is what several participants refer to as 'it is what it is'. By paying attention to these experiences, this section hopes to think about this problem of age as a problem of treatment response-ability, where our often inflexible and limited treatment options are failing to respond to older people who have used drugs for a long time. This is not to place blame on individual workers or even the intentions behind certain services or treatments but to draw attention to the hidden and often unintended effects of such systems. Participants here share their concerns over a treatment system that is (a) singular (lacking choice and diversity in approach), (b) punitive (making it difficult to talk openly) and (c) inflexible (especially in its dosing regime).
Where treatment for heroin dependency is driven by an ambition for abstinence through opiate substitution therapy (OST), comprising, overwhelmingly, of the choice of two medications, methadone or buprenorphine (Home Office, 2012; NHS, 2012), older service users, like Andy, who have lived through many treatment episodes and repeated attempts to abstain from heroin using these two options, yearn for a new approach: 'You know, just saying, like, we require you to be drug free, but I've 8 | DENNIS tried, you know, it's not [working]. Can we try something else before I snuff it?' Recognising that his age is against him, he is reaching out for something different, an invitation that is never fully taken up.
Several participants, including Andy, hope for a legal source of heroin (diamorphine) or morphine so they do not have to continue using illicit heroin and exposing themselves to the kinds of harms Beckie and James talked about (above). Upset with the way methadone and buprenophine are promoted and other opiate substitutes withheld, Sofia says indignantly: '[the manager] must have some kind of share in methadone. We'll give you methadone by the bucket full, but he won't give you morphine'.
Participants often felt infantilised and that their years of experience and expertise were neglected: 'I'm a grown up and yet drug service worker will tell me what's best for me' (Kimona, 50). These feelings of frustration were particularly stark in relation to professionals, especially doctors, who did not have the embodied, experiential knowledge of what it was like to take these substances every day and develop habits with them and what they made possible. For example, in a rare example where morphine had been previously prescribed (as part of a clinical trial), Sofia tried repeatedly to get it reinstated, only to be offered an impossibly low dose. 'Don't insult me!', she shouts, acting out one of these consultations. Feeling unheard and even purposely silenced, Sofia describes feeling 'resentful' as she is forced, as she sees it, to continue using illicit heroin: I'm resentful because I'm fifty-six, I know what works for me. I know that heroin [diamorphine] or morphine is not in itself harmful, what is harmful is using stupid street drugs so please […] I'm not going to die of morphine, just work with me, give me what I need.
With this, Sofia relocates the risk (of death and harm) from the substance to the treatment system that denies her legal heroin, putting her life, as she sees it, into the hands of the 'street drug', and neglecting 'what works' for her and what she 'needs'. She feels let down and angry: 'It was like "you think you're in charge and you don't know jack shit". I'd like to run my own recovery but it's like from the top down'. I am very, very resentful about the fact that somebody who has got very little clue can, you know, dictate to what I think, and I know, would be better for my wellbeing, and I've got to defer and go 'yes sir, three bags full sir, yes [names the doctor], yes arsehole.
(Sofia, 56) But perhaps more worryingly, as we see in other parts of the interview, is how this anger and frustration can quickly move into despondency and despair: 'You're not helping me, and I'm going to make do, and I want to get stoned because I'm now fed up, and I want some relief'. It is in this sense that such feelings of detachment might not only lead to desperate acts of 'relief' but may, as we see (below) in Anya's case, push participants to leave treatment altogether (with its known attached risks, noted above).
A second concern voiced by participants over their treatment was that it was punitive. For example, using heroin 'on top' of an opiate substitution prescription could result in them having to take their prescription under supervision (in the pharmacy, observed by a pharmacist). Because of this, participants felt unable to talk about their illicit drug consumption (one of the ironies of a drug treatment system where service users are both subjects of care and legal sanction).
I'd love to be able to have a grown-up conversation and just be like [speak frankly about his drug use] … Like, I'm fifty-one, for God's sake, like. I've been around a long time. I don't need to be punished. I understand it [his heroin use].

DRUG FATALITIES AND TREATMENT FATALISM
This silencing is frustrating not only for participants but also for their keyworkers. Andy says 'none of us have that conversation [about 'on top' use] because once you've had that conversation, it opens up all the safety measures and difficult conversations' (e.g., triggering a drug test and, with this, a stricter dosing regime, like supervised consumption). Some keyworkers therefore preferred not to know when a client was using heroin/drugs 'on top' of their prescription. This not knowing, then, can be understood as a kind of unspoken knowing. But this is a learnt (un)knowing developed through and reliant on a specific service user-worker relationship, which can be easily lost, as we see in Anya's case, when she gets a new worker: That's why I stopped using the methadone for the first time because I was years and years on it and then my [new] keyworker, every time I was coming here, she was asking for a [urine] sample and, of course, they were not coming out 'clean'. She said, 'Next time, if it comes out the same, I'm doing it supervised.' I said, 'Look, I'm not going to the pharmacy [to take it under supervision]. It's degrading'.
(Anya, 41) Despite her protests, Anya was put onto a regime of supervised consumption. Sofia mocks this system, putting on a dictator's voice: 'we've got ways to make you comply'. But rather than complying, refusing to be degraded as Anya sees it, Anya ended up leaving the service.
As Anya's case highlights, it is not only that this system can be punitive but, in having to hide one's 'on top' heroin use, it fosters a closed, opaque (un)knowing that encourages variability in how individual keyworkers handle such drug use, making prescription regimes uncertain and reliant on this relationship. Maintaining this relationship therefore becomes a crucial part of sustaining one's current or desired prescription dose and regime, which exposes this care relationship to powers of discrimination and manipulation, as well as easy disruption. As we have seen, if a keyworker leaves a service or the relationship is ruptured, or even if new rules are brought in, service users face changes to their prescription that could disrupt how these medications are consumed. And, where these practices have become embedded in people's lives and habits, a disturbance to this regime can be shattering. It is this potential for life-changing disruption that participants feel is not always fully understood, or worse still, disregarded.
As I say, I used to go and teach, I used to go and do whatever I wanted to do, you wouldn't know [I was on a morphine prescription], and these people, they set themselves up as judge and jury, and I've lost… no, I haven't lost the will to live, I've lost the will to fight, to bring my point of view.
(Sofia, 56) Stopping herself just short of saying that she's lost the will to live, Sofia expresses her frustrations at a treatment system that once gave her the freedoms to 'do whatever [she] wanted to do' to one where she now feels judged, punished and constrained.
It was not just a singular and punitive (and potentially manipulative and discriminatory) system that distanced older service users from their treatment but an inflexible one. Participants spoke of reducing their prescription dose (often at the encouragement of workers), only to find it hard to then increase it back up again. This was especially the case for the two participants who had been prescribed morphine (following a clinical trial). Sofia says that 'in here, they won't go backwards'. Another participant, Erik, confirms: 'They took us down, but they wouldn't put us back up.' For Erik, this even meant he was contemplating asking for methadone instead (a surprising step when he preferred morphine in every other way).

DENNIS
But, as aforementioned, this is not to say that an inability to respond to older heroin users' needs is due to an individual unwillingness (or even service-level unwillingness). In fact, many participants spoke highly of individual workers and the treatment service (as 'one of the better ones'). Rather, participants were aware of the policy and organisational constraints put on treatment services, which meant payments were based on 'successful completions', that is, those leaving treatment drug-free. As James sums up: Say, somebody comes into the service, they're self-harming and they're using shitloads of drugs. Then they cut out the self-harm and they drastically reduce their drug use. The services are not rewarded for that in any way, shape or form. They're only rewarded for this cold business of successful completion.
It is not surprising then that service users are encouraged to reduce their substitution medication as services try to achieve these 'successful completions'. But where younger service users may be more able to make such reductions and take on such advice (having lived for less time with these substances and habits), older users felt they had already tried everything and wanted something new.
These issues in treatment policies and structures, mostly to do with a narrow, punitive and inflexible substitution therapy, left participants feeling alienated, leading to feelings of frustration but also fatalism. As Anya puts it: For me, I don't like it, but I didn't have any options. I had to accept it, and that's what I keep doing, so we'll see. I don't know what's going to happen and if I'm going to continue like this or not, but, for now, it is what it is.
This phrase, 'it is what it is', gets repeated by two other participants and signals to a treatment fatigue and growing acquiescence in the decree of fate, which is a concern for how close it comes to fate's other meaning of death. Speaking of what she calls a 'fight to be heard' and have her treatment changed to include a morphine prescription, Anya sums up this sentiment: 'At the moment, it's very grey. I can't see anything past a metre in front of me. I can't see it just yet. Maybe it will change, but, right now, it's a bit bleak'.

DISCUSSION: IMPROVING RESPONSE-ABILIT Y, PREVENTING FATALISM/IT Y
Shaping response-abilities, things and living beings can be inside and outside human and nonhuman bodies, at different scales of time and space. All together the players evoke, trigger, and call forth what -and who -exists. (Haraway, 2016: 16) Thinking of workers, service users, drugs and treatment systems as players in what Donna Haraway describes as a string figuring of knowing and doing, our abilities to respond to different forms and entities become intrinsic to what and who can exist. Striking a chord here, treatment epistemologies not only render certain ways of being with drugs know-able but also, through this, possible. It is this twodirectional work of ontology and epistemology that makes our treatment systems so crucial to intervening in drug-related deaths. Based on a feminist ethic of 'response-ability', these findings expose the potential | 11 DRUG FATALITIES AND TREATMENT FATALISM repercussions of participants feeling closed off and closed down (described as a fatalism, 'it is what it is') from treatment networks that are experienced as singular, punitive and inflexible. This sentiment is summed up by Andy's point that 'you can't be open'. Furthermore, seeing drug-related deaths as a causal effect of ageing and age-related conditions neglects the social processes that make premature deaths more inevitable for some than others (for example, following entrenched lines of socioeconomic disadvantage, Hamilton, 2020;Heyman et al., 2019).
With reciprocity at its heart, an ethic of response-ability calls for us to pay greater attention to drugrelated deaths as a failure to respond to these long-inhabited ways of being with drugs that demand a different approach. This is not about being led by the patient, as in patient-led care, which puts undue pressure on the individual to express their needs, but is about responding to the whole network where 'need' is collectively produced (Mol, 2008). Staying with the string figuring metaphor -a game of return and relay -this puts equal responsibility on services, workers, policymakers, researchers and other publics who must improve their capacity to respond to these ever-changing patterns of drug use and their effects. As we have seen, age is not an automatic risk factor for heroin-related deaths. In fact, there are many ways that age could be protecting people from harm and overdose. Rather, then, there is a much more complicated picture of multiple social, political and material 'players', where the risk of an overdose is contingent on the response-ability of workers, treatment policies and medication, which of course intersects with a much larger web of historical and political-economic forces. Staying with the specificities of the treatment site, we have seen how such processes have worked to pressurise people to reduce their prescription or leave treatment altogether.
By seeing drug-related deaths as a failure in response-ability, there is much we can do to intervene. Treatment needs to be flexible, rather than procedural and punitive, in order to be able to learn from and respond to older people's ways of being with drugs and not block, redirect and discipline them (see, Malins, 2017, on a desire-led treatment system). Sofia, Anya, Erik, Andy and others all describe a rupture to their lives and life achievements (e.g., employment, education and family life) due to a change or reduction in their opioid medication. This adds to a growing criticism of abstinence-focused recovery policies that fail to account for the role drugs play in people's lives (see, e.g., Fomiatti et al., 2019;Fomiatti, 2020;Harris & Rhodes, 2013). More specifically, then, the kind of 'open' treatment practice this article calls for is a way of responding to heroin users' modes of knowing and doing despite what James calls these 'cold' and distancing policies and regulations (based on drug-free recovery). In very practical terms, this may look like making more opioids available for substitution treatment, flexibility in how and when they can be consumed, and a continuity in keyworker relationship. In a disturbing twist of fate, this may be the kind of flexibility that the deadly virus COVID-19 has brought with it as services have tried to limit social contact. Within weeks of the UK's first lockdown, new regulations, unimaginable a few months earlier, were implemented, which meant service users, some for the first time, could collect their OST medication to consume at home (PHE, 2020). We wait with cautious optimism to see how this 'natural experiment' (Finch, 2020) may unfold and the care it makes possible.
The findings and recommendations in this article add to long-made arguments within the harm reduction field for more situated care (Rhodes, 2002(Rhodes, , 2009). But to accept this invitation here is to also give something of ourselves over, at an embodied level. As Karen Barad puts it: 'Each of "us" is constituted in response-ability. Each of "us" is constituted as responsible for the other, as the other' (2012: 215, original emphasis). Bodies are cared for through a 'collective knowing and doing, an ecology of practices' (Haraway, 2016: 34;see also, Stengers, 2010) and are thus 'held together' but also put 'at risk' in these networks (Mol, 2008;Mol et al., 2010;Martin et al., 2015;Puig de la Bellacasa, 2017

CONCLUSION
These times called the Anthropocene are times of multispecies, including human, urgency: of great mass death and extinction; […]; of refusing to know and to cultivate the capacity of response-ability; of refusing to be present in and to onrushing catastrophe in time; of unprecedented looking away. (Haraway, 2016: 35) With reciprocity at its heart, a feminist ethic of response-ability calls for us to pay attention to drugrelated deaths as a failure to respond to the needs of people who use drugs. I have done this by looking at treatment provision, but this is only one among many possible sites. As people who use heroin are getting older, more people are dying, but as these findings suggest, this does not have to be the case. Highlighting the dangers of extrapolating theories from large datasets, this small qualitative study looks at what can be learnt if we hone in on people's stories and lived experience. Choosing to turn towards, rather than away from, this catastrophe of drug-related deaths, and move in closer, the study has found that, for some, age could be working, conversely, to protect them from harm and overdose. This is happening through a change in desire for heroin's more relaxing effects rather than an overwhelming 'hit', smoking rather than injecting, coming to terms with their drug use rather than getting caught up in cycles of using and abstaining, and seeking to end their illicit opioid use through alternative legal options. If anything, then, we need to be doing more to learn from and assist with these modes of care that are already being (and trying to be) employed by people who use drugs (see, e.g., Duff, 2015;Farrugia et al., 2019;Gonçalves et al., 2016;Harris, 2020;Race et al., 2021). Instead, current treatment practices, often designed around younger, newer users, and thus drugs being recover-able, are leaving older service users feeling ignored, or worse still, punished and constrained by limited and inflexible OST medications and regimes. By looking away, we risk producing despondency, detachment and despair, and with this, a giving up and over to what is easily dismissed as fate. Therefore, understanding the importance of living and dying in these lively terms of reciprocity, this article asks us to pay better attention to these ecologies of care and the processes by which the needs of older drug users are defined and maintained.

ACKNOWLEDGMENTS
I would like to thank the Wellcome Trust for funding this research. I am very grateful to the Sociology of Health and Illness reviewers and editorial team for their insights and support. I am extremely thankful, as always, to all the research participants who gave up their time to speak to me. I am especially indebted to the Service user coordinator who put me in touch with many of their peers. This article is dedicated to this person (who remains anonymous for confidentiality reasons) and the cause they fight so tirelessly for: service user involvement in research and service provision.

DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.

| 13
DRUG FATALITIES AND TREATMENT FATALISM ORCID Fay Dennis https://orcid.org/0000-0002-2441-6313 ENDNOTES 1 More details on this policy shift from 'harm reduction' to 'recovery' can be found elsewhere (e.g. Dennis et al., 2019;Floodgate, 2018; and a critical analysis of its potential impact on drug-related deaths can be seen in Boyt, 2014;Stevens, 2019) 2 Because of the different ways drug-related deaths are measured by the devolved counties in the UK, I focus here on