Organisational prerequisites for coordinating the return-to-work process for people with multimorbidity and psychosocial difficulties

Abstract Purpose The return-to-work (RTW) process for people with multimorbidity and psychosocial difficulties can be complicated. This study explores the organisational prerequisites for coordinating these patients’ RTW processes from the perspective of coordinators in different clinical areas in Sweden. Material and methods Six focus group interviews were conducted with 24 coordinators working in primary healthcare (PHC), psychiatric and orthopaedic clinics. The data were analysed thematically, inspired by organisation theory. Results Coordinators described varying approaches to people with multimorbidity and psychosocial difficulties, with more hesitancy among PHC coordinators, who were perceived by other coordinators as hindering patient flows between clinical areas. Most organisational barriers to RTW were identified in the healthcare sector. These were long waiting times, physicians drawing up inadequate RTW plans, coordinators being involved late in the sickness absence process, and lack of rehabilitation programmes for people with multimorbidity. The barriers in relation to organisations such as Social Insurance Agency and Employment Services were caused by regulations and differing perspectives, priorities, and procedures. Conclusion Our findings indicate what is needed to improve the RTW process for patients with complex circumstances: better working conditions, steering, and guidelines; shorter waiting times; and a willingness among coordinators from different clinical areas to collaborate around patients. Implications for rehabilitation RTW coordinators need sufficient physical and psychosocial working conditions as well as clear leadership. In order to avoid inequalities in access to RTW support, better systems are needed to identify patients who would benefit from rehabilitation and RTW coordination. There is a need for multilevel collaboration between clinical areas so that patients with multiple healthcare contacts and prolonged sickness absence can obtain support during the RTW process.


Introduction
Return-to-work (RTW) coordinators have been introduced in many countries. Their role is to support the RTW process for people on sickness absence (SA). They are often employed by large workplaces, insurance companies, or the healthcare system. The way they select which people on SA will be offered this support differs considerably in line with the demands of their organisations. Previous research has seldom looked at the organisational factors that influence the work of RTW coordinators, focussing instead on what effect RTW coordinators have on RTW rates. Some literature reviews have found no effect [1,2], while others have found moderate evidence that RTW coordination results in a higher probability of RTW [3][4][5][6]. Other studies have scrutinised the professional competence of RTW coordinators and found information gathering, organisational/administrative skills, communication skills, conflict management, professional credibility and factual knowledge to be important [7][8][9][10]. Few studies have focussed on the organisational prerequisites and conditions that are necessary for coordinators to succeed with the RTW process. Studies that do look at organisational factors find that RTW coordinators improve the amount of communication and collaboration that takes place both within their own organisations [11] and externally with other organisations [7, [12][13][14], although some studies [11,14] show that most coordinators do not engage sufficiently in external collaboration. An Australian study found that the support and resources received by RTW coordinators varied from one type of organisation to another. Several struggled with part-time positions which required them to "wear many hats" [13]. A Swedish study [15] investigating facilitators of and barriers to the coordination of RTW for patients with common mental disorders in primary healthcare centres (PHC) found a number of organisational factors to be important. Some of these were: access to a coordinator network with a process leader; managerial mandate; time to reflect upon and develop the coordination; an introduction to and supervision of the coordinator role; and a suitable physical work environment. Other important factors at PHCs were leadership engagement; routines and a team-based approach to RTW; and staff's understanding and appreciation of the coordinator role.
Most studies of RTW coordinators focus on people with one diagnosis, even though people with multimorbidity and psychosocial difficulties are more likely to be on SA [16]. It has been suggested that people with multimorbidity and psychosocial difficulties benefit particularly from coordination of the RTW process, as their more complex circumstances may make communicating and dealing with many contacts difficult [11,17]. At the same time, these more complex RTW processes may present their RTW coordinators with considerable challenges. The aim of this study was to explore how coordinators in Swedish healthcare experience their work with people with multimorbidity and psychosocial difficulties and what they see as the organisational prerequisites for coordinating the RTW process for this group. The study involves coordinators in PHC, psychiatric, and orthopaedic clinics and focusses on rehabilitation and RTW processes, internal and external collaboration, steering, and leadership.

The Swedish coordinators and their mission in the RTW process
In 2020, after more than a decade of state-financed trials, Swedish healthcare providers became obliged to offer patients on SA coordination of the RTW process if needed [18]. The purpose of this new coordinating role is to help patients to enter working life, return to, or remain in work [19]. The coordinators have three main missions: � Giving support to individual patients � Coordination within the healthcare services � External collaboration with, for example, the employer, the Public Employment Service, the Social Insurance Agency (SIA), and the social services.
Their mission includes promoting equality in the RTW process and identifying risk factors in patients' work, home and social life that might prolong SA [19]. Coordinators work in accordance with a so-called "rehabilitation chain," which means that patients' work ability is assessed in increasingly broader terms as time passes, which affects their eligibility for SA benefits. When the study's data collection took place, work ability was assessed in relation to any job on the labour market after 180 SA days. After 365 days, SA benefits were only granted in cases of severe illness.

The regional organisation of coordinators
The Healthcare Administration in Region Stockholm, Sweden, provided primary healthcare (from 2017) and psychiatry (from 2019) with resources to form inter-professional knowledge teams in insurance medicine. These teams give clinicians in the various clinical areas courses in insurance medicine and form networks for the coordinators. Due to the large number of coordinators in primary healthcare, this knowledge team formed eight smaller networks with process leaders to support coordinators in their respective geographical areas. The process leaders constitute a regional team who supervise the overall development of the coordinating function. In 2017, coordinators were also appointed in all the region's six orthopaedic clinics. However, when the study took place in 2020, these had not been provided with a process leader or a knowledge team covering orthopaedic care.

A theoretical understanding of organisations
To understand what is going on within and between organisations, Ahrne's [20] organisation theory is useful. He argues that organisations determine conditions for action by providing staff with resources, rules and power-factors that give professionals varying degrees of flexibility or limitation for action.
Ahrne describes professionals as "organisational centaurs," as they act both on behalf of their organisation and their own human mind. The centaurs, here referred to as professionals, are in constant interaction within and between organisations. Irrespective of who they interact with, there is a tension between the human mind and organisational demands to think similarly and follow rules. An organisation often creates activities that are independent of the person conducting them, supported by clear rules that demand as little human involvement as possible from the individual. This can lead to greater standardisation of work and consequently also less professional discretion [21]. At the same time, professionals often claim to know best how to carry out their tasks and the conditions that are required [20]. How much they can influence this is linked to organisations' ability to replace an individual or a group of professionals.
Professionals have varying opportunities to influence the authority and demands of their organisations. It is often the authorised representatives who formulate tasks, routines, and rules about who to help and who to collaborate with; other professionals are expected to comply with these. Those who do not identify with all the expectations tend to seek more flexibility and the way they act is often determined by the human mind, knowledge, or other motivating factors.
Organisations (particularly new ones) with unclear rules, fewer resources or a weak internal culture are vulnerable to centrifugal forces, such as professionals having differing opinions about their mission and work. One important centripetal force is institutional demands such as laws, guidelines, or obtaining funding. Coalitions of professionals in an organisation can claim autonomy or influence by negotiating resources or forming external alliances in order to achieve goals outside their own control. While one organisation may mobilise resources to achieve strength in relation to others, others may mobilise their resources to defend their own interests. As negotiation implies uncertainty, an organisation or coalition needs to show a united face, especially when others appeal to the human mind or knowledge aiming to destabilise the unity. The meeting between professionals from different organisations is determined by the power resources and authority of each organisation [20].

Research design
The present study has a qualitative research design and is based on focus group interviews with coordinators in three clinical areas: PHC, psychiatric, and orthopaedic clinics. Focus groups are a collective procedure that promotes a lively, complex, and unpredictable dialogue [22] and are valuable when little is known about the research question at focus and for gaining participants shared understanding and experiences of a certain phenomenon [23].

Participants and recruitment
All 82 coordinators who in the spring of 2020 were employed in the study setting, i.e., PHC, psychiatry and orthopaedic clinics in Region Stockholm, were invited to participate with the purpose to reach as robust data saturation as possible [24]. They received a letter about the study and two reminders via email.
Groups of 3-5 coordinators were created, three with coordinators working in PHC, two in psychiatric clinics, and one in orthopaedic clinics. By dividing the coordinators up according to healthcare area, we aimed to avoid imbalances within groups arising from the fact that there were so few orthopaedic coordinators compared to PHC coordinators. Of the 27 coordinators who notified their interest in participating, 3 withdrew shortly before the interview due to cold symptoms during the Covid-19 pandemic. In total, 24 coordinators with various occupational backgrounds participated. Most were physiotherapists and occupational therapists, some were health social workers, and others had more uncommon occupational backgrounds. Two were men and 22 were women, which reflects the gender distribution of coordinators in Sweden [25]. With the exception of one group of less experienced psychiatric coordinators, the participants had worked as coordinators for between 1.5 and 6 years (see Table 1). All were between 40 and 60 years old and had previous relevant working experience from their original occupations.

Interview guide
A semi-structured interview guide was developed, based on previous literature about RTW coordination, e.g., indicating difficult organisational prerequisites [13,15], a vague role description [15], and a lack of research on RTW coordination with people with multimorbidity and psychosocial difficulties, although this group are more likely than others to be on SA [16]. The questions were open-ended to stimulate discussions among participants, and the four main topics were: (1) selection of patients and how frequently they encounter patients with multimorbidity and psychosocial difficulties, (2) barriers to a well-functioning rehabilitation and RTW process, (3) working conditions, and (4) the professional role and competence.

Procedures and context
The moderator (first author) aimed to create a relaxed dialogue between the group participants, helping them to feel comfortable and share their different views [26,27]. An observer (second author) added complementary questions and took notes about the non-verbal interaction. The moderator was a researcher with prior training and experience in conducting focus group interviews, while the observer was a research assistant with prior training and experience in group counselling in healthcare. Both were certified health social workers, with several years of work experience with people with psychosocial difficulties and social insurances.
The face-to-face focus group discussions were conducted from late August to early October 2020 in meeting rooms at the Academic Primary Healthcare Centres' facilities in Region Stockholm. They were conducted in Swedish, lasted 77-100 min excluding a coffee break, were recorded, and transcribed verbatim.
Before the discussion began, both authors and participants briefly presented themselves and their backgrounds. The observer (second author) was known to the psychiatric RTW coordinators as she previously worked as one herself, and currently worked in the regional psychiatric knowledge team in insurance medicine at the time the interview took place, arranging activities for the coordinators and other healthcare staff. When the interviews were scheduled, all participants were informed that the second author would be present, and at the beginning of the interviews, the participants were encouraged to bring up issues that could be useful for developing the knowledge team's activities. The first author was known to the participants as an RTW coordination researcher whom they had been in previous contact with at research seminars as well as regarding data collections in other studies. Most participants were also known to each other in the focus groups, being members of the same network for several years. There was thus a familiarity among the focus groups participants.

Ethical considerations
Information about the participants, their colleagues, and workplaces which could identify them was omitted in the transcriptions. Several ethical considerations have been taken to ensure that participants are not identified during the analysing and writing up of the article. Because some characteristics are very unusual among the RTW coordinators in the specific region, in their specific clinical settings in particular, we have not disclosed the most uncommon ones. This refers especially to very uncommon occupational backgrounds and gender aspects. The two men who participated could easily be identified, whereby we in general describe the coordinators in gender neutral terms, and several participants has been provided with gender neutral names. All names are fictive. Ethical approval was granted by the Swedish Ethical Review Authority (Dnr 2020-00403).

Methods of analysis
Data were thematically analysed in six steps according to Braun and Clarke [28]. Because we aimed to understand differences and concordances between the three clinical areas the coordinators worked in, these contexts constituted an additional analytical layer in each step of the analysis. We were inspired by Wilkinson [27], who stresses that social interaction is at the very core of the focus group interview, which includes a dynamic negotiation of the realities of the group and its context. The empirical analysis thus focusses on the interactive co-construction of the participants' social reality in their different contexts. After familiarising themselves with the data and a joint discussion of the possible themes which might be identified, the authors separately conducted a first coding of all transcriptions for multimorbidity and psychosocial complexity, healthcare processes and rehabilitation. The authors then discussed their coding and their notes about such issues as differences between groups or between clinical areas. The aim of this was to ensure a common perspective of the most important themes. The second author then performed an analysis more specifically related to the aims of this study. This was carried out for each group discussion, and a comparative analysis of the three clinical areas was performed. In the next phase of the analysis, the first author structured the codes into more clearly defined subthemes. The first author then checked the themes and subthemes with the data. At this point, some clarifications were made and nuances between healthcare fields were further elaborated. The analysis was jointly discussed to reach consensus on themes and subthemes as well as to strengthen reliability and validity [29,30].

Results
Three main themes were found: (1) varying approaches to people with multimorbidity and psychosocial difficulties, (2) barriers in the healthcare sector, and (3) barriers between sectors and their organisations.

Varying approaches to people with multimorbidity and psychosocial difficulties
When the coordinators were asked how they selected patients and how common multimorbidity and psychosocial difficulties were, most psychiatric coordinators said that their patients generally had multimorbidity, long-term SA and socially complex life situations. All three PHC coordinator groups answered spontaneously that they were not expected to work with such patients, arguing that these cases take too much time.
If you take on a person like that, who's more complex and far from the labour market, and maybe doesn't have unemployment or sickness benefit and there's nothing more, they take … Sometimes you take someone like that on when you've got the time, and then it just grows. If you've got three or four patients like that, you can't take anything else on. (Andrea, PHC coordinator, group 1) Others said they are unable to help these patients, or simply that they had been told only to select patients with either common mental disorders or musculoskeletal disorders who had been on SA for a maximum of three months. Later in the discussions, however, several PHC coordinators said that most of their patients do in fact have multimorbidity and many also have social problems. In a dialogue about whether their way of selecting patients is good or not, one PHC coordinator, who was used to working with complex cases in previous professional roles, said that this limited selection is difficult to adapt to: Anyway, that's what I've been told by the network leader. And it's been difficult for me, because I've previously worked in a different professional role in social psychiatry. So that I might … yes, I keep on thinking all the time, so that I stick to this role. And maybe I got a little too narrow, or whatever, only having so few patients. (Helene, PHC coordinator, group 2) The PHC coordinators also noted that nearly all their patients had common mental disorders, seldom musculoskeletal problems. Besides patients with these diagnoses who have been on SA for a maximum of three months, several said they also selected patients who had a job or were close to getting one, and were motivated to RTW. Motivation for work was also mentioned by some psychiatric coordinators as a criterion for their help, although they generally did not refuse patients support. The PHC coordinators discussed at length how they selected patients and the process leaders explained the reasons for their decisions: Lisa: So, when it's already happened … when people have already got sick leave injuries of various kinds. Then it's also very difficult when there are … or if you … or conflicts, that it's a conflict that hasn't been resolved and no one has tackled it.
Lisa: Then it's almost impossible to come back. One must deal … Eva: Exactly, time is probably crucial, that you've been at home for a long time.
Ren� e: But I also think it's difficult if the home situation is very challenging, so you never the chance to recover.
Ren� e: So that there's … even though you're on sick leave you don't get the opportunity to recover, so it'll never really be the right time to return, because you haven't built up any reserves. (PHC coordinators/ process leaders, group 5) The process leaders mainly discussed the social and other problems surrounding some patients, and while Ren� e pointed out that their opportunities for recovery were unequal because of their differing home circumstances, Lisa maintained that this was not something they could do anything about. The groups mentioned a range of social problems which are barriers to RTW. The most common of these are economic difficulties, homelessness, having children with special needs, and problems returning to work because of a stressful home situation. The majority of the PHC coordinators said that they keep a distance from patients with obvious private or social problems not connected to their work situation, as they did not want to raise hopes about what they could do for them. However, several said they work consultatively with such cases by making phone calls or providing physicians or patients with information about social support.
Both PHC and psychiatric coordinators said they have considerable freedom to decide whether to take on a case or not. One coordinator maintained that it would lead to more equality of care if every patient on SA were systematically referred to a coordinator, instead of this being dependent on whether their particular physician decides to make a referral or not.
The psychiatric coordinators said that their patients often have multimorbidity and psychosocial difficulties. However, unlike the PHC coordinators, they did not see this as an obstacle to providing coordination and support. They stressed the need for the patient actively to want to work and be self-sufficient, but unlike the PHC coordinators, the psychiatric coordinators did not require the patient to be close to getting a job.
The orthopaedic coordinators selected patients differently, with several using triaging via the call-centre at the hospital which the patient contacted. If the patients have easily-solved questions about SA, the call-centre is trained to deal with them. Long-term and more complex SA cases are passed on to the coordinators. As a result of this triaging, the coordinators claimed that most of their cases were complex. Unlike the coordinators in PHC and psychiatric clinics, however, they almost never have face-toface meetings with the patients or attend meetings with other stakeholders. They use the telephone to advise patients how to proceed or who to contact about a variety of matters.

Barriers in the healthcare sector
When analysing the coordinators' conversations about their experience of the organisational prerequisites for and barriers to the rehabilitation and RTW process, it became clear that most barriers were located in the healthcare sector.

Overall differences between the clinical areas
There were significant differences between the PHC, psychiatry and orthopaedic clinics in how coordinators were recruited. The psychiatric and orthopaedic coordinators were mostly internally recruited from the clinic and were therefore already known to colleagues when they started as coordinators. The orthopaedic coordinators argued that this was a precondition for being able to liaise with the physicians and successfully fulfil their role.
Hanna: We believe that it's easier to come from inside the clinic, being familiar with the doctors, because it's … Alex: And knowledgeable in orthopaedics, but already having a role, that you are a person [to colleagues].
Hanna: And that's the advantage for us, that we have … They have some kind of confidence, or quite a lot of confidence in you from the beginning. That has decreased the threshold. If we had come from outside … no.
Alex: No, that would've been very difficult. (Orthopaedic coordinators, group 6) All worked part-time as coordinators besides being employed as physiotherapists. They stressed the importance of being physically located at the clinic where they worked for being able to form close and flexible collaborations with physicians and other colleagues. PHC coordinators, on the other hand, were often employed by rehabilitation centres and hired by one or several PHCs. This meant they had to explain their function to the general practitioners (GP) and build working relationships with them at the PHCs.

Physicians as barriers or facilitators
Working as a coordinator was described as being like walking down a corridor in which the doors to the physicians' consulting rooms were closed. The PHC coordinators in particular described this as a barrier. Another barrier was the lack of continuity, with patients often seeing a number of GPs during their SA. This is a widely discussed problem in Swedish primary healthcare.
Physicians were said to obstruct the coordinators' work in a variety of ways. The PHC coordinators said that they were constantly having to remind GPs about their function and that they should refer cases early in the SA process. Some GPs were described as cooperating as little as possible, although the coordinators understood this was due to lack of time. It was also felt that some senior physicians issue inadequate SA certificates month after month, failing to draw up rehabilitation and RTW plans or to make sure that the patient has received proper treatment or contact with a coordinator. Several groups discussed the inequities in care and support arising from individual physicians' attitudes. One suggested that all patients on SA should be referred to the coordinator for a brief assessment about the need for individual support, saying "So it becomes a more natural part of the treatment and don't fall on the doctor, because I feel that lead to inequalities in care" (Andrea, PHC coordinator, group 1) However, GPs was also described as being important facilitators, acting as role models for other GPs. Two PHC groups felt that they could see a generational shift, with younger, non-specialist physicians showing more interest in performing their sickness certification well and being more open to collaboration.

The importance of a team and colleagues
The coordinators described people with multimorbidity and psychosocial difficulties as often having many contacts inside and outside healthcare. This required more working hours from coordinators-time they did not have. Working in inter-professional teams, especially psychosocial healthcare teams, was therefore deemed important because they produce holistic assessments and better prioritising and collaborating around patient needs. One group discussed the difficulties when many professionals try to handle messy situations within their own silos, and one coordinator said: It's often very messy in people's life. That's how it is. And they may have eight ongoing healthcare contacts, well maybe not eight, but they've quite many. And they put forward everything in every room all the time, and so on. Then it's so good having team check-ins with the patient and the involved contacts. (Eli, Psychiatric coordinator, group 4) While the psychiatric coordinators talked about the advantages of team collaboration at their clinics, the PHC and orthopaedic coordinators often lacked this. They were more dependent on knocking on doors and other professionals responding to their calls. Many felt isolated in their work, having no one to turn to with problems. The network of coordinators was thus not only a forum for consultation and gaining knowledge but it also compensated for their sense of isolation in the workplace.

Delayed and insufficient care and rehabilitation
Most groups emphasised that a main barrier to the rehabilitation and RTW process is delays due to long waiting times for investigation and treatment in the healthcare services. Patients with complex symptoms and circumstances, particularly psychiatric patients, are often investigated by one unit at a time, for months or even years, before they finally end up in another long queue for treatment or a rehabilitation programme. One coordinator describes the complex process for a patient with multiple difficulties: She had some addiction, ended up at the Affective Disorder Clinic, however, who said she needs to be at the Addiction Clinic instead. After this, she was sent forward to the Personality Disorder Program at another department. Yes, and eventually she's back, not at the Affective Disorder Clinic, however, but at the Anxiety Disorder Clinic. At this time about a year has passed./ … /And then she was referred for a neuropsychiatric assessment, that is, at another department. But it actually worked, and we got her back to work. For her, it became obvious that "I cannot wait to get my diagnosis." (Kim, psychiatric coordinator, group 4) Lack of continuity in rehabilitation was also common, with patients being transferred between rehabilitation programmes for different diagnoses. Several groups said there were plenty of treatments for patients with fatigue symptoms and stress-related diagnoses, yet it was difficult to find rehabilitation programmes for patients with multiple diagnoses and complex problems.
The RTW process was also delayed by patients being referred to a coordinator late in the rehabilitation chain, when some PHC coordinators said that they were not able to do anything to help them.

Collaboration problems between clinical areas
When discussing multimorbidity in patients, some PHC coordinators said they should not be involved if a patient is also registered with a psychiatric clinic, even if there was no coordinator at that clinic. Both PHC and psychiatric coordinators described collaboration problems between PHC and psychiatry, between coordinators as well as other staff. They also found it wrong that health social workers in PHC only worked with psychotherapeutic treatment, at the expense of addressing the social problems that hindered patients' RTW. Several PHC coordinators said that their clinic had no health social worker, although others did have such an expert they could consult about issues such as social support systems.
Both orthopaedic and psychiatric coordinators described patients being denied RTW coordination when they had been discharged to PHC. PHC coordinators refused these patients because they had been on SA too long, even though there was an ongoing RTW process.
Alex: Yes, this is a concern. Because primary healthcare has decided in its organisation to intervene early. We highlighted this yesterday at our network meeting via [name], who can talk more with the PHC competence team and so on. That it doesn't work when you transfer patients who are far down the [rehabilitation] chain.
Hanna: No, they're ready orthopedically, they end with us. But they're not back at work and need support. And they may have been on sick leave for six months, so when they come to the coordinators there it's, "no, that one has been on sick leave … " Alex:"We don't take on people who are older … " Hanna:" … so long, so we don't take it." And then … Alex: Because they've actually really said that they … It's like three months, then it's … yes. To limit themselves, because they don't have the resources of course. And we talked to … We have the specialist healthcare [competence team], they're in the building.
Hanna: Academic … Alex: Academic Specialist Centre also, in our network. And they've got chronically ill people who are … These are people who've been on sick leave as long as you like. And they don't get help even though they've got quite far in their return to work. So they don't get an overlap when they go to primary healthcare. (Orthopaedic coordinators, group 6) The orthopaedic coordinators described how each healthcare area organised its own coordinators and did not coordinate with each other around patients. They said that the regional healthcare administration supported each coordinator organisation in various ways but did not want to address barriers to patient flows between them or their causes, i.e., differing opinions about which patients should qualify for help from a coordinator. As the orthopaedic coordinators lacked a process leader or anyone in a position of authority to promote their interests, they had, as the previous quote indicates, found an ally-a person who was about to start a new knowledge team covering several specialist clinical areas.

The importance of steering and leadership
As indicated above, coordinators from the different clinical areas had varying amounts of access to leading functions. What all groups had in common was that they found it wrong that they had not received a clear mission or a target group of patients, neither from the regional healthcare administration nor from their managers. On the other hand, they also said that the lack of steering gave them a great sense of freedom to shape their work.
Andrea: Yes, exactly, but still I've got a huuuge say in how I do my work.
Kiran: But I think that's very important too, because we all work differently, and we must be flexible with [patients'] employers and be able to have meetings with them too. So you need this flexibility and responsibility to be able to fulfil your professional role well.
Andrea: It's not possible to have this fully booked diary and these seven visits that psychologists are supposed to have, is it seven a day? I always have vacancies next week, absolutely not always fully booked, so then I can spontaneously, "we can meet in two days." I can't have a three-month waiting list because the patient should be back [to work] by then. So there are very short lead times, so to speak, from when I get the patient to when I book them in. (Andrea gets support from the rest of the group) (PHC coordinators, group 1) The conversations revealed a great deal of confusion, variation and inequality of care as a result of insufficient steering. PHC and psychiatric coordinators said that their process leaders and knowledge teams had been important in helping them find a way to do their work. The orthopaedic coordinators also wanted a knowledge team and process leaders. They had no one who was in a position to formulate routines or raise important issues with PHC coordinators and process leaders such as how to improve collaboration around patients.
All PHC groups agreed that managers were crucial for explaining the routines for SA work and the coordinators' function to physicians. They were thus the key to successful coordination. One manager, for example, stopped GPs having regular meetings with the coordinator as this would have negatively affected the number of patient appointments they could fit in.
All groups described lacking support in prioritising work tasks, and managers who did not understand the coordinators' work and role: And so it just gets like this, "But you kind of do everything, or what do you actually do?" I even got that question from one of the PHC managers, "You've got a phone and a laptop, but what do you actually do? What good do you do … what use are you to us?" (Charlie, PHC coordinator, group 2) The PHC groups described how they had to prove that they were important for the workplace. This was linked to the fact that they were often hired by PHC but were employed at rehabilitation centres. This gave rise to unclear managerial responsibility for their work and a feeling that the PHC manager could suddenly end their contract. Coordinators in all the healthcare areas maintained that their managers did not understand what they needed to do a good job. Several had barely received any administrative or workplace introduction and some lacked a workspace of their own where they could make phone calls.
One group discussed the phenomenon of managers receiving funding for employing a coordinator but then telling them to work as a physiotherapist instead, or using the funding for other purpose at the clinic.

Barriers between sectors and their organisations
Outside the healthcare sector, institutional regulations and differing organisational perspectives, priorities and procedures all pose organisational barriers to RTW coordination.

Differing organisational perspectives, priorities, and procedures
Coordinators described how different stakeholders have different perspectives, roles, priorities, goals and procedures which are not harmonised and thus delay the rehabilitation process. Physicians had their focus on sickness and disability when writing sickness certificates, while the SIA focussed on work ability, and the coordinators focussed on the indicators of health to promote rehabilitation and RTW process. According to the PHC and psychiatric coordinators, officers at the SIA and Public Employment Service often dealt with one issue after another, often waiting several months for one diagnostic investigation before putting the patient in a queue for the vocational rehabilitation they all know will be necessary in the end. In such cases, the coordinators tried to get officers to work with the different steps simultaneously to speed up the process. One psychiatric coordinator group discussed how they adopted a gently authoritarian approach to influence how the case worker did his or her work: Kim: I'm thinking the new ones [SIA officers], you can still try to make them … "But even if you put them in the queue for a joint investigation now, it takes quite a long time. Now we're in this queue, you can do it at the same time. What do you think?" Then I think they deal … then, you usually have to do some deals.
Eli: You have to be gently authoritarian in your contacts.
Kim: Yes, so a little bit of dealing. (Psychiatric coordinators, group 4) Several groups discussed the fact that officers dealt with SA cases on their own, without ascertaining whether any coordinator was involved from whom they could obtain more information. This hindered collaboration and more rapid processing. There were different views about how much coordinators could influence officers at these authorities. The process leaders discussed the role of the coordinator in such cases: Eva: And I think you can help, because you can put pressure on both the Public Employment Service and the Social Insurance Agency to make something happen. I think I've got that power anyway. The patients may have had to wait for ages, but then I push, and then something happens.
Iris: That's great to see.
Eva: It's actually happened several times and then … because a patient can't do it, they may have tried an awful lot and been banging their heads against a brick wall, but there are still strings we can pull.
Iris: Yes, because I mean we speak a bit of the same language and it's not easy for a patient to speak that language to the Social Insurance Agency, or all the terms in the medical world and so on. But we speak the same language after all. (PHC coordinators/process leaders, group 5) High staff turnover rates at the SIA were also mentioned, with patients often being assigned new officers. This was seen to give rise to unpredictability because different officers often made different assessments.

The rehabilitation chain and the regulations
Both PHC and psychiatric coordinators described barriers to collaboration between authorities because of regulations being interpreted differently by different stakeholders. The coordinators described the rehabilitation chain itself as problematic, e.g., when it does not correspond to a patient's individual rehabilitation process. Deniz says: I think the most difficult ones are those who've got an employer, but they're ill enough not to be able to work full-time and then the Social Insurance Agency withdraws their right to sickness benefit saying "Then you must apply for some other job." I kind of never get used to that. (Deniz, psychiatric coordinator, group 3) Especially in PHC, coordinators found it difficult to help patients without a job. They often had to put a great deal of effort into making another authority act in line with the regulations.

Discussion
This study explores the organisational prerequisites for coordinating the RTW process for people with multimorbidity and psychosocial difficulties from the perspective of RTW coordinators in PHC, psychiatric and orthopaedic clinics. The findings show some hesitancy among PHC coordinators to work with these patients, which can be understood as caused by the instructions the PHC process leaders and the knowledge team gave them about selecting patients-which in turn was said to be based on findings from a research report [31] commissioned by the Healthcare Administration in Region Stockholm. This report was a randomised controlled trial in seven PHCs. It found that, compared to a control group, coordinating interventions led to a reduction in SA for patients with long-term SA in all the four years studied, yet with most significant effects in the first two years for patients with common mental disorders. There were some benefits for patients with longstanding pain, and fewer for those with a combination of the two. The report concludes that this group of multimorbid patients is probably in greatest need of RTW coordination and that interventions to target them should be developed [31]. The present study found, however, that PHC coordinators selected patients before reaching 90 SA days, which indicates that people with longer SA, multimorbidity and psychosocial difficulties are at risk of being denied RTW coordination. This has been identified by other researchers [32] exploring PHC coordinators work with patients with common mental disorders as an ethical issue in terms of unequal inclusion and support. The more complex the case, the less likelihood that the patient will receive the support of a coordinator, even though it has been shown that coordinators are especially important for this group as they are more likely to face challenges and obstacles in the rehabilitation and RTW process [11,17,31,33]. From a patient perspective, the consequence can be unequal access to RTW coordination for those with more difficulties and more complex RTW processes.
The findings show that the PHC coordinators' patient selection constitutes a way of organising their work, which not only has consequences for their own work but also for coordinators from other clinical areas-who in turn experience problems with patient flows and collaborating with PHC. The role of the coordinator is to coordinate contacts between clinical areas but this has not been a focus of research hitherto. Our study thus complements previous studies by revealing that there is insufficient collaboration over RTW interventions between coordinators in different clinical areas-in other words, between those whose mission it is to do exactly that. This seems to be partly due to insufficient regional steering, which means each coordinator network has considerable freedom to decide about their work and defend their own interests in relation to others [20]. Both PHC and psychiatric coordinators said that they can chose whether to take on more complex cases. This means that they apply professional discretion [21] even though the law stipulates that all patients in need of coordination have the right to receive it [18].
Rather than following general rules for selecting patients some coordinators said that they were always able to do something to support the RTW process also for people with psychosocial difficulties and long-term SA, thus showing more of what Arhne describes as human involvement [20]. There are several possible explanations for the different approaches, that some PHC coordinators in contrast to the psychiatric and orthopaedic coordinators express hesitancy getting involved in such cases. The orthopaedic coordinators' lack of time and resources, for example, had led to a standardisation of work [20], meaning they only worked consultatively with patients by telephone rather than taking a more active role in coordinating external contacts. Furthermore, as they lacked a process leader and monetary resources for their loosely formed network, they did not have what Ahrne [20] define as an authorised representative to negotiate their interests in relation to the PHC coordinators, meaning that the PHC had more power to defend its priorities. The PHC coordinators' priorities are based on a standardised patient selection procedure rather than on standardised tasks and appear to offer selected patients time and effort while other patients receive no support at all.
Another possible explanation for the differing approaches to working with people with long-term and complex SA is that, as several PHC coordinators said, they did not feel they had the right competence and therefore avoided such patients. A previous study [15] shows that coordinators tend to base their work practice on the competence acquired through their particular professional training and background. This leads to variations in how the coordination is conducted. This in turn poses the question of whether coordinators need additional training to give them a competence base that covers the range of issues patients may need support with during their rehabilitation and RTW process (see for example [17] for patients' perspective). It has been found that many healthcare professionals avoid psychosocial factors while working with work disability [34]. This is due to a lack of knowledge of biopsychosocial concepts and a belief that psychosocial concerns are not within their scope of practice [34], which is reflected in some PHC coordinators argument that it would be better if health social workers took on the more complex cases involving psychosocial problems.
The findings reveal organisational barriers both within the healthcare sector and between other sectors and their organisations. Because coordinators lack formal decision-making power, they try to negotiate with the SIA and Employment services officers, influencing established routines by appealing to their human engagement in cases [20]. The major barriers in the rehabilitation process were however seen to be located in the healthcare sector. This is an interesting finding given that it is often SIA officers who are blamed for problems with the SA process, for example, by physicians [35]. In fact, some coordinators understood SIA officers' criticism of inadequate SA certificates and rehabilitation plans, describing poor collaboration with physicians and physicians not making proper rehabilitation and RTW plans as the main barriers. Another barrier described was the lack of sufficient rehabilitation programmes for people with multimorbidity and psychosocial problems, a finding in line with these patients' own experiences [17].
Our findings also confirm those of Holmlund et al. [15], namely that the following organisational factors are important for coordinators in Swedish PHC settings: access to a coordinator network with a process leader; leadership engagement; managerial mandate; introduction of the coordinator role; a suitable physical work environment; and routines for the RTW process at the clinic. The lack of steering, leadership and routines outlined in the present study, sometimes led to a lack of mandate in the rehabilitation process, making the coordinators as spectators of the disability process, rather than in charge of the coordination and other actions that was needed in order to promote RTW. Holmlund et al. [15] also stress the importance of a team-based approach to RTW and staff's understanding and appreciation of the coordinator. In the present study, the team-based approach was integrated in the daily work of the psychiatric coordinators, while the PHC coordinators talked about being isolated and walking down a corridor with closed doors. The fact that many PHC coordinators were contracted by PHCs but employed by rehabilitation centres may make it more difficult to be included or to establish teams. Being contracted also seemed to involve what Ahrne [20] calls being more replaceable; it made PHC coordinators feel they needed to prove themselves worthy of staying at the PHC.

Implications for practice
Several improvements are warranted at organisational level to support the prerequisites for RTW coordination. The coordinators need sufficient physical and psychosocial working conditions, as well as clear leadership that helps coordinators to prioritise among tasks, as well as to strengthen the collaboration and SA routines at the clinic. In order to avoid inequalities in access to RTW support, better routines and systems are needed to support that patients come into contact with a coordinator early in the SA process and to identify patients who would benefit from RTW coordination. The findings also suggest working in inter-professional team can help holistic assessments and better prioritising and collaboration around patients with multiple needs.
There is also a need for multilevel collaboration between clinical areas so that patients with multiple healthcare contacts and prolonged SA can obtain support during the entire RTW process. Other implications for practice, derived from our findings regarding obstacles that delay the rehabilitation process, is that rehabilitation programs that are adjusted for people with multimorbidity and psychosocial difficulties are warranted.

Methodological discussion
An important aspect of focus group discussions is that participants should feel comfortable sharing differing or conflicting views with each other [27]. This was demonstrated for example when some coordinators maintained that they are not able to do anything for people on long-term SA, while others held contrasting views [36]. The construction of homogenous focus groups according to healthcare area and a separate group for PHC process leaders aimed to support free discussion without having to adjust to the reality of coordinators in other settings. It also avoided an imbalance due to there being so few orthopaedic coordinators in the Stockholm Region and gave rich data as the differences between the settings became clear, detailed, nuanced, and varied [24].
We cannot be sure that participants were reluctant to air criticism related to their knowledge team as a consequence of that the observer (second author) had previously worked as psychiatric RTW coordinator herself and presently were working in the psychiatric knowledge team in insurance medicine. Apart from this potential limitation, the second author's in-depth knowledge of the organisational context and the coordinators' daily work is a strength when it comes to interpreting the findings. The continuous discussion between the authors strengthens the reliability and validity of the research process and findings [29,30].
The fact that the present study took place in only one region is a strength with regard to understanding the specific organisational prerequisites. However, it was also a limitation because it weakens the transferability of findings: the results may have differed in other regions.

Conclusions
This is the first study focussing on the organisational prerequisites for RTW coordination in healthcare settings, further exploring how these prerequisites influence the work with people with multimorbidity and psychosocial difficulties. Coordinators describe different approaches to working with this group in the different clinical areas. The organisational barriers to the rehabilitation and RTW process were most striking in the healthcare sector, although barriers also were found between sectors and authorities. These were caused by regulations as well as differing organisational perspectives, priorities and procedures.
The findings indicate that, to improve the RTW process for the patients addressed in the present study (particularly those on prolonged SA), it is important for coordinators from different clinical areas to work together with patients in a coordinated way. Better steering is needed regarding the coordinator's introduction, material and psychosocial work environment, implementation of the function at the clinic, and motivating physicians to collaborate more with the coordinator. There is also a need to discuss how equality in coordinating support can be secured, to ensure that all patients who need RTW coordination actually receive such an offer.