Reducing harms through interactions: Workers orienting to unpredictable frames in a low-threshold project for people injecting drugs

0 This is the accepted manuscript of the article, which has been published in International Journal of Drug Policy, 2020, 82, 102828. https://doi.org/10.1016/j.drugpo.2020.102828 © 2020. This manuscript version is made available under the CC-BY-NC-ND 4.0 license http://creativecommons.org/licenses/by-nc-nd/4.0/ REDUCING HARMS THROUGH INTERACTIONS: WORKERS ORIENTING TO UNPREDICTABLE FRAMES IN A LOW-THRESHOLD PROJECT FOR PEOPLE INJECTING DRUGS Johanna Ranta, Tampere University, Finland Keywords injection drug use; harm reduction; low-threshold services; frames of interaction; mobile ethnography Abstract Background: People using injection drugs and living in complex, vulnerable and stigmatised life situations can face many kinds of barriers when accessing social and healthcare services. Thus, they are often encountered in easily reachable low-threshold services aimed at harm reduction. However, little is known about how clients’ multiple drug-related needs are actually encountered in the everyday practices of harm reduction work. Methods: This ethnomethodologically oriented interaction analysis examines harm reduction work in a Finnish low-threshold project, the aim of which is to support clients broadly in their drug-related life situations. The data consist of 12 audio-recorded client-worker meetings observed in 2017 by utilising mobile ethnography. In this context, workers are expected to react to clients’ needs in situ; they must take different kinds of orientations within the interactional frames that the clients have set for the meetings. This study asks: 1) In what kinds of frames are client-worker conversations conducted in the project? 2) How do the workers orient to these frames? 1 Results: The results point out that participants orient to various frames – practical, therapeutic, educational and casual – during the meetings. The workers respect the frames set by the clients and maintain them by taking variable interactional orientations. By doing so, they are able to respond to the clients’ individual and acute needs, and they conduct a client-centred harm reduction approach through interactions. Conclusions: The study highlights that when workers utilise situationally reactive interactional approaches, they conduct client-centred harm reduction work through interactions. When people using injection drugs are given much interactional freedom, they can be powerful actors in institutional contexts. Thus, harm reduction is not merely about reducing the concrete risks of drug use; it is also about social interactions that show respect for the interests of the people visiting the service. Introduction Harm reduction policies are widely used to control drug-related harms both in Finland and globally. The concept is defined in various ways, but the common aim is to offer people a better quality of life despite their drug use by reducing individual and societal harms and risks related to drugs. For instance, methadone maintenance treatment, needle exchange and supervised injection services are effective in reducing the risks associated with injection drug use (Clarke et al., 2016; Guinness et al., 2009; Krüsi et al., 2009; Palepu et al., 2006; Potier et al., 2014). In addition to health risks, a harm reduction approach is also beneficial in reducing social harms, such as homelessness (e.g., Briggs et al., 2009; Heather & Petersen, 2009; McNeil et al., 2012; Perälä, 2012). Harm reduction is based on human rights and is aimed at strengthening the social equality of people using drugs (Hunt, 2004; Hurme, 2004; Tammi, 2007). Low-threshold services offer good surroundings in which to actualise harm reduction work (Edland-Gryt & Skatvedt, 2012; Lee & Zerai, 2010), as they aim to reach ‘structurally vulnerable’ people in complex life situations (McNeil et al., 2016). In practice, ‘low threshold’ usually refers to easily accessible services that can be visited 2 anonymously and without an appointment and do not require the client to be sober or aiming at abstinence (Lee & Zerai, 2010). These services are allocated for people who are in risk of facing ‘high thresholds’ and ‘treatment barriers’ in social and healthcare services; for example, they might be expected to go through bureaucratic processes, to fit organisations’ criteria and to commit to organisations’ rules to get support despite their vulnerable positions in society and weakened abilities to function (e.g., Notley et al., 2012; Virokannas, 2019). The dynamics between clients and workers affect clients’ experiences of approachable, ‘lowthreshold’ service environments. In addition to social and healthcare professionals, peer workers with their own experiences of drug use can act as significant interaction participants in low-threshold services (Marshall et al., 2015). No matter their background, workers are expected to offer clients trustful relationships that are based on understanding how injection drug use affects everyday life (see Krüsi et al., 2009). This is important, for instance, to support clients dealing with shame due to the stigmatisation of injection drug use (see Krüsi et al., 2009; Rhodes et al., 2007). Although the importance of client-worker relationships has been noted, research on everyday interactions in harm reduction services has been limited (see Petersson, 2013; Ranta & Juhila, 2019). This study aims to fill this gap by examining 12 naturally occurring client-worker meetings (audio recorded in 2017) that took place in a Finnish low-threshold project with a harm reduction approach (henceforth referred to as ‘the project’). The work is targeted comprehensively at clients’ drug-related social needs. The low-threshold project’s institutional task of harm reduction does not strictly limit participants’ everyday actions. Instead, clients can start discussions they deem important when they spontaneously arrive at the project’s joint facilities or spend time there during functional group activities. This interactional freedom allows workers to react to clients’ acute needs in situ. Nevertheless, these encounters can be challenging for the workers, as they cannot predict the required interactional orientation. Still, they are expected to respond immediately to avoid violating the expected interaction order (see Juhila & Hall, 2017, p. 67) and to use their discretion regarding clients’ 3 needs by orienting, together with the clients, to different kinds of frames that clients set in interactions (see Goffman, 1986). This study assumes that by orienting to these frames, workers are actualising a comprehensive harm reduction work in action. By utilising ethnomethodologically oriented interaction analysis, this study asks: In what kinds of frames are client-worker conversations conducted in the project? How do the workers orient to these frames? Institutional task and unpredictable frames Drew and Heritage (1992, p. 3) argue that the essential features of institutional interaction are: 1) one or more participants represents a formal organisation and 2) the interaction is always ‘task-related’. They (1992, p. 22) write: ‘Institutional interaction involves an orientation by at least one of the participants to some core goal, task or identity (or set of them) conventionally associated with the institution in question. In short, institutional talk is normally informed by goal orientations of a relatively restricted conventional form’. Furthermore, Juhila (2000, p. 192) stresses that an institutional task is ‘something that has been set via legislation and administrative orders and instructions for social work in certain organisations’. Traditionally, workers define the goals for meetings within the limits set by the service’s task (Juhila, 2000). Both workers and clients are aware of this task and tend to orient to it in interactions (Drew & Heritage, 1992, p. 22). Workers often need to conduct this explicit task based on law, which defines precise boundaries for their working practices. For example, in child protection services, social workers have a legal responsibility to assess whether a given home is a safe place for a child to live. Correspondingly, in psychiatric services, doctors have a legal duty to assess whether a client with psychotic symptoms needs a referral for involuntary treatment. Nevertheless, not all services have obligatory law-based tasks to conduct. In the project, the aim is to offer clients voluntary support with a low threshold – a place to visit without strict, pre-defined terms. The project’s only but extensive institutional tasks are aimed firstly at reducing the harms of drug use and secondly at increasing the participation of people using drugs. These loose tasks allow clients 4 themselves to define the goals and themes of institutional encounters. In other words, clients set frames for the interactions (Goffman, 1986) in which they invite the workers to participate. Hall and Slembrouck (2014, pp. 65–66) define a frame as: . . . a socially recognised reference for the activity talkers are engaged in: it is both a (shared) definition of a situation and a framework for action (what to do), interpretation (how to interpret actions) and participation (who does what). It comes, among other things, with a set of expectations about likely and allowable interactional behavior and contributions, including the kind of topics talked about and the ways in which the issues raised will be interactions attended to and become consequential. Frames are always present when people meet each other, and they are negotiated in interactions (Hall & Slembrouck, 2014, pp. 65–66; Peräkylä, 1989, pp. 118–119, 127). Our actions define the meanings that we give to the things surrounding us (Husso et al., 2012; Peräkylä, 1990). Different entities within these actions can be defined as frames through which we orient to the interaction and interpret it. We define a surrounding situation differently in each frame, and each frame constitutes its own reality. We observe our surroundings and choose to move constantly from one frame to another and to act within those frames. (Goffman, 1986; Peräkylä, 1989, pp. 118–119; Peräkylä, 1990, pp. 16–17, 19.) Frames can also change during the interaction, as they can be accepted, displayed or even resisted by other participants (Hall & Slembrouck, 2014, pp. 65–66). Frames have been studied previously in several institutional settings (see Hall & Slembrouck, 2014; Husso et. al., 2012; Peräkylä, 1989; 1990). For instance, Peräkylä (1989) noted that the care of dying patients in hospital interactions is defined through practical, medical, lay and psychological frames. In addition, Husso et al. (2012) found that workers use practical, medical, individualistic and psychological frames when making sense of domestic violence in healthcare interventions. Although frames are always present in institutional contexts, analysis of frames in the everyday interactions of services for people using drugs barely exists (cf. Lehto, 1991). 5 This study pays particular attention to the discussions in which clients set the frames for interactions. Compared to traditional institutional interactions, this can be seen as ‘role play’, where the client assumes the worker’s role or identity – alongside being an object for the workers’ actions (Hall & Slembrouck, 2014, p. 65; see Drew & Heritage, 1992, p. 28; Juhila & Hall, 2017, pp. 67–68). As workers have certain duties and responsibilities towards clients in need of help, they are expected to join the frames in situ by positioning themselves according to required orientations within these frames (see Peräkylä, 1990, p. 22). For example, if clients want to discuss delicate issues, the workers are expected to orient to these situations with sensitive responses. All interactions are somewhat unpredictable. However, in services with law-based tasks, workers can better prepare themselves for encounters with clients based on the institution’s frames. In lowthreshold services without strict institutional tasks, encounters are usually more unpredictable. This creates a special context for both institutional interactions and harm reduction. Although workers are often familiar with the phenomena they discuss with clients, they cannot predict the frames clients will set at each meeting and how they should orient to those frames in interactions. To understand the complexity of these encounters and workers’ abilities to orient to different frames for the sake of their clients’ needs, client-worker interactions need to be examined in detail. Data and method Finnish harm reduction project as a research context In Finland, services aiming at reducing drug-related harms, such as needle exchange and opioid substitution treatment, are common nowadays (e.g. Tammi, 2007; Perälä, 2012). In recent years, fixed-term projects aiming at reducing homelessness and other social harms related to drug use have strengthened the national range of harm reduction services. The context of this study is a lowthreshold project with a harm reduction approach, organised by a national NGO. The project’s four workers focus especially on reducing drug-related social harms and increasing the participation of people using injection drugs. The clients’ life situations differ: while most of them currently use 6 injection drugs, others are committed to rehabilitative opioid substitution treatment. In either case, the clients’ current or previous long-term drug use still affects their everyday life. Thus, they use the project’s services. The client-worker encounters take place in the project’s facilities, which are open three days a week for clients to eat, sleep, watch TV, converse with workers and other clients, attend to financial matters and so on. The clients spend time in a large, jointly shared living room for all participants. If privacy is needed, the workers’ office can be used. Group meetings conducted by the workers are arranged weekly or less often. The kitchen group, in which the participants cook, eat and discuss everyday issues together, is arranged in the project’s kitchen, which is located next to the living room. Another example of a group activity is a group for women in which the conversations take place while participants perform various physical activities, such as cosmetic care, body care or dancing. The data corpus and participants The data of this study consist of 12 meetings in the project’s facilities (see Table 1). The data were gathered over six months in 2017 by utilising mobile ethnography, also known as ‘shadowing’ (e.g., Czarniawska, 2007; Quinlan, 2008). The researcher observed and audio recorded the client-worker conversations, which are the core of the analysis. In addition, she followed the participants’ movements and made field notes based on their actions. The field notes provide context for the conversations; they describe actions, feelings and surroundings that cannot be traced from the recordings. The recorded data are characterised as naturally occurring interactions: the meetings would have happened regardless of this study, and the researcher aimed at maintaining the role of an observer. 7 Table 1. The data and participants of this study Ethical considerations The guidelines of the National Advisory Board on Research Integrity (TENK) in Finland were respected during this research. The Ethics Committee of the Tampere region stated that there were no ethical problems in this study. The clients’ vulnerable positions were acknowledged when making ethical considerations. The priority was that all the clients were invited to participate. However, in collaboration with the workers, the researcher assessed that if a client was known to have acute paranoia, the client’s mental health would have been put at risk due to the paranoid thoughts that audio recording could have caused. Thus, these clients were not actively recruited to participate. The researcher spent 40 hours in the project before audio recording. This made it possible to carefully familiarise herself with the project’s everyday actions, to discuss the purpose of the research and to form trustful relationships with the participants, aiming to keep the recorded client-worker interaction flow as natural for the institutional context as possible. If the researcher’s presence clearly affected a verbal interaction, that interaction was edited out of the data. All participants signed written research consent forms in which they were informed that all personal identifiers would be removed, participation was voluntary, they could refuse to participate at any time and the support they were receiving from the project would not be affected if they chose not to participate. 12 audio recordings, 678 minutes 16 participants: 11 clients, 5 workers (4 from the project, 1 from another organisation) Three clients and two workers participated in both group meetings and non-appointment meetings. GROUP MEETINGS Data: 3 audio recordings (2 kitchen groups, 1 group for women) Length: 515 minutes in total from 164 minutes to 187 minutes per meeting Participants: 8 different clients, 3-5 per meeting 3 different workers, 1-2 per meeting NON-APPOINTMENT MEETINGS Data: 9 audio recordings (discussions with individual clients) Length: 163 minutes in total from 5 minutes to 75 minutes per meeting Participants: 5 different clients, 1 per meeting 4 different workers, 1-3 per meeting 8 Techniques of the analysis Ethnomethodologically oriented interaction analysis (e.g., Hall et al., 2014; Peräkylä, 1990) is applied to study client-worker interactions in the low-threshold project with a harm reduction approach. In more detail, Erving Goffman’s (1986) concept of frame is utilised as an analytical tool. The following questions are examined: 1) In what kinds of frames are client-worker conversations conducted in the project? 2) How do the workers orient to these frames? In analysing the data, the first interpretation was that the workers dealt with clients’ various needs that were not easily categorised under one thematic title – for example, housing, financial matters or drug use. Thus, the assumption was that, in the project’s practices, harm reduction was an extensive and inclusionary, not an exclusionary, approach. This confirmed that the project’s institutional task was loose, and it allowed clients to start the kinds of conversations they assessed as relevant to them. In other words, the clients set the frames for the discussions in which they invited the workers to participate. Moreover, it was noted that the workers used variable interactional approaches while orienting to these frames in situ. The research questions are based on these perceptions. The ATLAS.ti 8.0 programme was utilised as a coding tool to find the structure for more detailed interaction analysis and to recognise all the frames in the data. While conducting the coding, one frame was interpreted to have ended when another frame was set for the conversation. One code represents one appearance of a frame in the data. In total, four frames were identified: practical (27 codes), therapeutic (25), educational (25) and casual (25); these appear in the data with almost equal frequency. All frames include different themes and meanings relevant to harm reduction. The practical frame is related to reducing practical harms in clients’ everyday lives, such as in financial matters and housing. The therapeutic frame focuses on reflecting clients’ traumatic life experiences and other sensitive personal issues. The educational frame is based on delivering educational information regarding harms caused by drug use related to, for example, everyday routines. The 15 8 Molly: It was good. It’s my favourite. I thought that I could make rice pudding. I mean, it includes that rice thing. And jam. So, could I make it myself . . . 9 W3: Yeah, you can. 10 Molly: With porridge rice, or what is it? Is it porridge—? 11 W3: Porridge rice and then, for example, a bit— 12 Molly: And then jam. Okay. By starting this conversation, Molly invites other participants to discuss her eating habits. The ‘agreement’ Molly has made (i.e., ‘always have to eat something’) indicates that she knows she needs to take her situation seriously as eating is a basic human need (line 1). After Sandra’s clarifying question regarding Molly’s eating rhythm (line 2), W3 confirms Molly’s comment on her behalf with a delicate question form (line 3). Molly confirms this by repeating W3’s line and continues by giving accounts of how she has tried to follow her agreement without good results (line 4). W3 responds by keeping up the educational frame and advising Molly on how to make eating a routine (line 5), simultaneously allying with Molly’s opinion that she should eat more often. Molly’s positive response indicates that she approves of the advice (line 6). She highlights her approval by confirming that she likes rice pudding and by asking if she could make it herself (line 8). W3 confirms Molly’s vision (line 9), thus encouraging her to make the pudding. The encouraging tone continues while W3 advises Molly and teaches her the details of the pudding ingredients (lines 10, 11). After adding one ingredient, jam, Molly neutrally approves of the advice (line 12). The frame continues: 13 W3: There it is then. 14 Molly: I’ll make it then [with laughter], so that I’ll begin to eat every day. 15 W3: Well, if it’s something that you like to eat, then eat for god’s sake, ‘cause you need all the energy in the world. 16 Sandra: Yeah. 17 W3: Absolutely. 16 18 Molly: Mm. 19 W3: I wouldn’t usually advise anyone to eat rice pudding, that it’s good and healthy, but if it’s one of the few things that you can and like to eat, so . . . 20 Molly: It is and . . . 21 W3: Then hold on to it. 22 Molly: But I really try to eat once a day. After W3’s confirming and approving answer (line 13), Molly makes the decision to make pudding to accomplish her goal concerning her eating rhythm (line 14). Although this decision could be a minor issue in another context, it holds significant meaning in Molly’s everyday life. W3 takes a worrying orientation by speaking out his opinion and by advising Molly to eat the pudding, which he strengthens with an underlining idiom, ‘for god’s sake’, and a supplementary argument, including emphasising words: ‘cause you need all the energy in the world’ (line 15). After Sandra has supported W3’s opinion (line 16), W3 strengthens his previous argument, aiming to point out that it is unnegotiable (line 17) – again, to establish that he supports Molly in her decision and to express his worried approach. Saying that he would not recommend eating pudding in other circumstances can be seen as W3’s way of assuring Molly that he takes the issue seriously and that he understands her situation (line 19). After Molly confirms that eating pudding is a good solution for her (line 20), W3 allies with Molly by advising her to stick to her decision (line 21). In the last line (22), Molly assures the workers that she takes the advice about her eating seriously. Casual frame Oliver has arrived at the project’s facilities aiming to catch up with W4 over a game of chess. Since W4 has time for it, they start a game. As playing chess is a very casual activity, Oliver sets the casual frame for the interaction. Oliver and W4 discuss their chess moves in the living room: 1 W4: I shouldn’t move this from here. [32 sec pause; W4 thinks about her next move] Oh no, oh no, oh no. [6 sec pause] Well . . . 17 2 Oliver: Take your time. There’s no rush. 3 W4: This is good practice for it. 4 Oliver: Mm. 5 W4: This forces to practice it. Erm . . . [24 sec pause] Oh no, oh no. [16 sec pause] There, or there, so . . . [14 sec pause] Are you as reflective in real life as you are in this game [with laughter]? 6 Oliver: I’m quite pensive. 7 W4: [5 sec pause] For me, this is good practice to think, ‘If you make this kind of move, what happens at the other end?’ [with laughter]. In the casual frame, the participants do not orient to the traditional hierarchical roles of ‘client’ and ‘worker’. This can be seen when W4 orients to the discussion as Oliver’s equal co-player in the game (line 1). Furthermore, while responding to W4 with some advice (line 2), Oliver is stepping outside of the client’s role; advising is usually included in the worker’s interactional orientation, as is the case in the practical and educational frames. Next, W4, as a co-player, makes a common and casual comment regarding how playing chess is affecting her ability to ‘take one’s time’ (lines 3, 5). While W4 continues thinking about her next move, she asks a question concerning Oliver’s character outside the game (line 5), as anyone playing with Oliver might. After Oliver responds positively to W4’s interpretation (line 6), W4 takes a personal orientation while sharing her thoughts concerning herself as a private person. Instead of acting as a worker, W4 orients to the situation as a peer or a friend who shares her personal thoughts and weaknesses with another friend as she reveals that the reflective nature is something that she needs to rehearse (line 7). The game continues: 8 Oliver: I know what I’m doing. They say that a rook would be worth more than a knight or a bishop, but it’s a matter of opinion of which piece you like to play with. Which do you like to play with, Helen [W4], with a knight? 18 9 W4: I like it, somehow, it’s probably ‘cause I like horses so much [In Finnish, the knight piece is called the ‘horse’] anyway. 10 Oliver: I don’t like horses at all. 11 W4: But then, I have to say that these are really, these are going from one side to another, this rook and this . . . 12 Oliver: Yeah, they are. 13 W4: . . . just that if I knew how to use them, so . . . 14 Oliver: Yeah. 15 W4: . . . probably, then, there would be some clever or good moves. Oliver keeps up the casual frame while comparing the chess pieces and invites W4 to participate in the frame by asking about her favourite piece (line 8). As W4 explains why she likes the knight most, she reveals her personal affections as a private person (line 9). This indicates ‘ordinary’ talk, which appears more often in interactions between friends than in institutional contexts. Thus, it can be interpreted that W4 is orienting to the situation as Oliver’s peer or friend. Oliver’s response implies that he has taken the friend orientation as well since he dares to give his honest opinion about horses (line 10) even though he is aware that W4 ‘likes them so much’ (line 9). The conversation ends with lines (11, 13, 15) in which W4 casually analyses, as Oliver’s co-player, good gaming tactics, which Oliver neutrally confirms (lines 12, 14). Discussion The aim of this study was to examine in what kinds of frames client-worker conversations are conducted in a low-threshold project with a harm reduction approach and how workers orient to these frames. The results point out that the focus and aim, the workers’ orientations and the implications for harm reduction varied within four frames: practical, therapeutic, educational and casual (see Table 2). 19 Table 2. Summary of the results. In the practical, therapeutic and educational frames, the workers aim at harm reduction by taking orientations typical of social and healthcare workers (e.g., Hall et al., 2014). In the practical frame, they aim to reduce practical harms, for example, by preventing clients’ social problems from escalating and eliminating possible ‘barriers’ in other services by contacting them together with the clients (see Notley et al., 2012; Virokannas, 2019). In the therapeutic frame, harms related to mental health and functional ability are reduced by supporting clients in complex situations caused by drug use and traumatic life experiences. In the educational frame, the workers aim to fix harms that are realised as, for example, an inability to maintain routines, such as eating and other basic human needs. In the casual frame, the workers take orientations that are common among peers outside institutional hierarchies, such as citizens or friends. By treating clients as equal participants, the workers aim to Frame Focus and aim Workers’ interactional orientations Implications for harm reduction Practical Aiding with practical issues, e.g. contacting other services, financial matters, housing Taking a traditional worker’s role by advising, escorting, assuring, confirming, motivating, allying Preventing drug-related social problems, e.g. homelessness, from arising or escalating Therapeutic Reflecting on delicate personal issues, e.g. traumatic life history, difficult current situation Taking a traditional worker’s role by speaking out, sharpening, allying, normalising, judging other’s behaviour, hinting Achieving better mental health and functional ability despite drug addiction Educational Delivering educational information, e.g. the effects of drug use on everyday life and routines Taking a traditional worker’s role by allying, confirming, encouraging, advising, teaching, worrying, speaking out, assuring Improving life management to take care of basic needs despite drug addiction Casual Discussing ‘ordinary’ everyday life issues not related to drug use, e.g. hobbies, TV shows Stepping outside a traditional worker’s role by being a coplayer, peer, friend, private person Increasing participation and reversing the stigma connected to drug use 20 increase their participation by reducing the stigmatisation and shame related to drug use (see Krüsi et al., 2009; Rhodes et al., 2007). According to Hurme (2004), harm reduction is ‘pragmatic, reactive and situational work’. This study supports this argument as it shows in detail what happens when harm reduction as a loose institutional task allows participants interactional and situational freedom. Although the frames appearing in the project’s encounters have some similarities to the frames in other institutional settings (e.g. Peräkylä 1989; Husso et al. 2012), compared to other research, the salience of the frames in this study is that a) they are always set by the clients, and b) they are unpredictable for the workers. Still, the workers act as situationally reactive ‘interaction chameleons’, changing their orientations from one to another within each frame for the sake of the clients’ needs. The findings show that this kind of harm reduction work is interactionally demanding for workers. It requires both professional interaction skills (especially in the practical, therapeutic and educational frames) and the personal capacity to act in social encounters (especially in the casual frame). Most of all, it demands strong discretion of what kinds of interactional approaches should be used and when. To respect clients’ needs, harm reduction workers must be situationally reactive: they need to give space for clients’ frames by accepting and maintaining them, and they need to change their orientations in the ways that each frame requires. This changes the traditional institutional clientworker power dynamic (see Drew & Heritage, 1992; Juhila & Hall, 2017) and enables a particularly client-centred approach – one of the core aims of harm reduction. Compassionate and open approaches to interaction are important avenues through which to strive for harm reduction’s aim of social equality (see Hunt, 2004; Hurme, 2004; Tammi, 2007) and increase socially excluded people’s confidence towards society and social and healthcare services (see Krüsi et al., 2009; Perälä, 2012). The results should be interpreted through the study’s institutional context; the existence of the analysed frames cannot be generalised to all harm reduction work. Additionally, the audio recordings represent moments between individual clients and workers. Therefore, the found frames do not 21 include all of the thematic content that appears in the project’s everyday encounters. Moreover, as any client could walk into the project’s facilities at any time, the recording was cut off if the arriving client was not a participant in the research and had not granted research permission. As some of the recordings were short, it was not always possible to evidence how often the frames changed with each client. Hence, the analysis was based on more detailed interactional nuances – that is, how orientations change during an individual frame. This kind of focus revealed that, despite their vulnerable and complex life situations, having much interactional freedom can make people using injection drugs powerful actors in institutional contexts. Conclusions Services in which clients define the goals of the meetings imply client-centred environments that enable workers to react to variable drug-related harms in situ. The possibility to ask for help with acute needs is extremely important in services for people in vulnerable life situations: they often need multiple social and healthcare services, but they are not always able to ask for help because of structural barriers or the shame they are experiencing. The presence of variable frames implies that people injecting drugs benefit from free forums where they can ask for the kind of support they prefer. Flexible practices enable client-centred encounters, but facing clients’ complex and individual needs makes harm reduction work demanding. Giving workers the chance to define their own working practices improves their ability to act based on professional ethics. However, there is a risk of moral distress if too much responsibility for clients’ vulnerable situations is placed on harm reduction services, as not all problems can be solved with such services alone. Thus, resilient collaboration with social and healthcare services is essential. This study highlights that harm reduction is not just about reducing the concrete risks and harms of drug use; it is also about social interactions that show respect for the interests of the people visiting the service. As interactional orientations have a significant meaning in harm reduction work, more research concerning client-worker encounters is needed. 22 Acknowledgements The research was supported by the Academy of Finland [grant number 307661] and the Alfred Kordelin Foundation [grant number 160373]. Declarations of interest: none. References Briggs, D., Rhodes, T., Marks, D., Kimber, J., Holloway, G. & Jones, S. (2009). Injecting drug use and unstable housing: Scope for structural interventions in harm reduction. Drugs: Education, prevention and policy, 16(5), 436–450. Clarke, K., Harrisa, D., Zweiflerb, J.A., Lasherb, M., Mortimerb, R.B. & Hughesb, S. (2016). The significance of harm reduction as a social and health care intervention for injecting drug users: An exploratory study of a needle exchange program in Fresno, California. Social work in public health, 31(5), 398–407. Czarniawska, B. (2007). Shadowing and other techniques for doing fieldwork in modern societies. Malmö: Liber. Drew, P. & Heritage, J. (1992). Analyzing talk at work: An introduction. In P. Drew & J. Heritage (Eds) (1992). Talk at work: Interaction in institutional settings (pp. 3–65). Cambridge: Cambridge University Press. Edland-Gryt, M. & Skatvedt, A.H. (2012). Thresholds in a low-threshold setting: An empirical study of barriers in a centre for people with drug problems and mental health disorders. International Journal of Drug Policy, 24, 257–264. Goffman, E. (1986). Frame analysis: An essay on the organization of experience. Boston: Northeastern University Press. 23 Guinness, L., Vickerman, P., Quayyum, Z., Foss, A., Watts, C., Rodericks, A., Azim, T., Jana, S., & Kumaranayake, L. (2009). The cost-effectiveness of consistent and early intervention of harm reduction for injecting drug users in Bangladesh. Addiction, 105, 319–328. Hall, C., Juhila K., Matarese, M. & van Nijnatten, C. (Eds) (2014). Analysing social work communication: Discourse in practice. London: Routledge. Hall, C. & Slembrouck, S. (2014). Boundary work. In C. Hall, K. Juhila & M. Matarese & C. van Nijnatten (Eds) Analysing Social Work Communication: Discourse in Practice (pp. 61–78). London: Routledge. Heather, S.L. & Petersen, S.R. (2009). Demarginalizing the marginalized in substance abuse treatment: Stories of homeless, active substance users in an urban harm reduction based drop-in center. Addiction Research and Theory, 17(6), 622–636. Hunt, N. (2004). Public health or human rights: What comes first? International Journal of Drug Policy, 15, 231–237. Husso, M., Virkki, T., Notko, M., Holma, J., Laitila, A. & Mäntysaari, M. (2012). Making sense of domestic violence intervention in professional health care. Health and Social Care in the Community, 20(4), 347–355. Hurme, T. (2004). Jumping out of harm’s way. Harm reduction in Finnish drug policy: conceptual problems and contradictions. Nordic Studies on Alcohol and Drugs 21(1), 17–27. Juhila, K. (2000). Organisaation sanelemaa toimintaa vai syvällistä vuorovaikutusta? Institutionaalinen tehtävä ja ammatilliset tavoitteet päihdekeskusteluissa [Institutional task and professional aims in conversations about alcohol abuse]. In A. Jokinen & E. Suoninen (Eds) Auttamistyö keskusteluna. Tutkimuksia sosiaalija terapiatyön arjesta (pp. 191–219). Tampere: Vastapaino. 24 Juhila, K. & Hall, C. (2017). Analysing management of responsibilities. In K. Juhila, S. Raitakari & C. Hall (Eds) Responsibilisation at the Margins of Welfare Services (pp. 57–79). London: Routledge. Krüsi, A., Small, W., Wood, E. & Kerr, T. (2009). An integrated supervised injecting program within a care facility for HIV-positive individuals: A qualitative evaluation. AIDS Care, 21(5), 638–644. Lee, H. S. & Zerai, A. (2010). ‘Everyone deserves services no matter what’: Defining success in harm-reduction-based substance user treatment. Substance Use & Misuse, 45, 2411–2427. Lehto, J. (1991). Juoppojen professionaalinen auttaminen: Tutkimus lääkärien, sosiaalityöntekijöiden ja poliisien juoppouteen kohdistamasta työstä ja siihen kohdistuvien professionaalisuusodotusten vaikutuksesta [Professional help for drunkards: A study of the work of physicians, social workers and police officers with drunkenness and of the effects of expectations of professionalism in this work]. Tutkimuksia 1. Helsinki: Sosiaalija terveyshallitus. Marshall, Z., Dechman, M.K., Minichiello, A., Alcock, L., Harris G.E. (2015). Peering into the literature: A systematic review of the roles of people who inject drugs in harm reduction initiatives. Drug and Alcohol Dependence, 151, 1–14. McNeil, R., Guirguis-Younger, M., Dilley, L.B., Aubry, T.D., Turnbull, J. & Hwang, S.W. (2012). Harm reduction services as a point-of-entry to and source of end-of-life care and support for homeless and marginally housed persons who use alcohol and/or illicit drugs: A qualitative analysis. BMC Public Health, 312(12). McNeil, R., Kerr, T., Pauly, B., Wood, E. & Small, W. (2016). Advancing patient-centered care for structurally vulnerable drug-using populations: A qualitative study of the perspectives of people who use drugs regarding the potential integration of harm reduction interventions into hospitals. Addiction, 111(4), 685–94.