Background: Collaborative care treatment is widely recognized as an effective approach to improve the quality of mental healthcare through enhanced and structured collaboration between general practice and specialized psychiatry. However, studies indicate that the complexity of collaborative care treatment interventions challenge the implementation in real-life general practice settings. Four Danish Collaborative Care Models were launched in 2014 for patients with mild/moderate anxiety and depression. These involved collaboration between general practitioners, care managers and consultant psychiatrists. Taking a multi-practice bottom-up approach, this paper aims to explore the perceived barriers and enablers related to collaborative care for patients with mental health problems and to investigate the actual experiences with a Danish collaborative care model in a single-case study in order to identify enablers and barriers for successful implementation. Methods: Combining interviews and observations of usual treatment practices, we conducted a multi-practice study among general practitioners who were not involved in the Danish collaborative care models to explore their perspectives on existing mental health treatment and to investigate (from a bottom-up approach) their perceptions of and need for collaborative care in mental health treatment. Additionally, by combining observations and qualitative interviews, we followed the implementation of a Danish collaborative care model in a single-case study to convey identified barriers and enablers of the collaborative care model. Results: Experienced and perceived enablers of the Danish collaborative care model mainly consisted of a need for new treatment options to deal with mild/moderate anxiety and depression. The model was considered to meet the need for a free fast track to high-quality treatment. Experienced barriers included: poor adaptation of the model to the working conditions and needs in daily general practice, time consumption, unsustainable logistical set-up and unclear care manager role. General practitioners in the multi-practice study considered access to treatment and not collaboration with specialised psychiatry to be essential for this group of patients. Conclusions: The study calls for increased attention to implementation processes and better adaptation of collaborative care models to the clinical reality of general practice. Future interventions should address the treatment needs of specific patient populations and should involve relevant stakeholders in the design and implementation processes. * Correspondence: firstname.lastname@example.org Research Unit for General Practice & Section for General Medical Practice, Department of Public Health, Aarhus University, Bartholins Allé 2, 8000 Aarhus C, Denmark © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Møller et al. BMC Family Practice (2018) 19:78 Page 2 of 12 Background Aim Collaborative care is a widely recognized approach to This paper aims to explore the experiences with current improve mental healthcare through enhanced and treatment practices among GPs, clinic staff and care systematic collaboration between general practice and managers and to examine their views on and perceptions specialized psychiatry. Trials have shown better out- of future collaborative care. The paper also aims to iden- comes with collaborative care for patients with mental tify enablers and barriers for successful implementation health problems when compared to usual care [1–10]. of a specific DCCM. The study provides new knowledge However, trial evidence seems difficult to translate on a sparsely researched topic by taking a qualitative into real-life care delivery [8, 11–17]. Danish Collab- approach to exploring experiences with and perceptions orative Care Models (DCCMs) for patients with mild/ of collaborative care in a general practice setting. moderate depression and anxiety were launched in four (of five) Danish regions in 2014. The interven- Methods tion designsof these DCCMswereinlinewiththe Design and setting recommendations in existing literature on collabora- We employed a qualitative approach and used a com- tive care [18–21]. Gask et al.  defines CC treat- bined research design to conduct a multi-practice study ment by four key components: (1) a multi-professional among GPs together with a single-case study of the approach to patient care, (2) a structured management DCCM. plan in the form of guidelines or protocols for interven- General practice clinics in Denmark are privately tions, e.g. antidepressant medication, patient screening, owned by GPs and organised in small units, either as education, counselling, cognitive behavior therapy, (3) single-handed practices (1 GP) or group practices (2–10 scheduled patient follow-up, and (4) enhanced GPs) . Most single-handed practices also have a inter-professional communication, e.g. team meetings, medical receptionist and/or a nurse, whereas group case conferences, individual consultation/supervision, practices usually have both types of clinic staff. More shared medical records, patient-specific written or verbal than 95% of the Danish population is listed with a spe- feedback between caregivers. cific GP (approximately 1600 patients per GP), and all The model explored in this study involved collaboration inhabitants must obtain a referral from their GP to get between general practitioner (GP) and care managers specialist treatment. Treatment is tax-funded, and health (often psychiatrically trained nurses) from specialized psy- services are divided into primary care (including general chiatric units in the Central Denmark Region. Care man- practice) and secondary care (including hospitals and agers offered a 12-week treatment course to patients in specialised psychiatry) with separate organisations and general practice. The course consisted of 5–6individual different financial structures. Psychologists generally op- Cognitive Behavioural Therapy-based sessions; first and erate on a private basis, but GP-referred patients are en- last sessions were shared with the patient’sGP. Other titled to partial reimbursement. Danish GPs treat components were questionnaires for diagnosing and mon- patients with various degrees of mental health problems. itoring, group-based psycho-education, supervision and For the larger patient population with mild/moderate training of GPs, and weekly supervision of care managers anxiety and depression, usual care often consists of a by a psychiatrist. Each care manager collaborated with up limited number of conversational therapy consultations to ten GPs. with the GP and in some cases medical treatment. Only The GPs were responsible for identifying eligible in severe cases of mental health problems, GPs may patients, arranging appointments with the care man- refer patients to treatment in specialised psychiatry. ager and the patient for a shared consultation and subsequently referring patients electronically through Sampling and recruitment existing psychiatric visitation procedures by adding a We conducted a multi-practice study to explore the ex- reference to the specific care manager assigned the periences of GPs and their staff with current mental GP. Moreover, GPs held the treatment responsibility health treatment and to examine their perceptions of throughout the course of collaborative care treatment and need for collaborative care. The practices were se- and provided a treatment room for the care manager. lected among general practices that did not take part in GPs were remunerated for the time spent on clinical the DCCM. To ensure diversity of opinion and broad conferences with the care managers. general practice representation, we applied a purposive The DCCM was designed by specialised psychiatry to sampling strategy employing three main parameters: engage GPs in a collaboration aiming to improve both urban/rural location, practice size and proximity to the treatment of patients and the GPs’ skills in psychi- specialised psychiatry. We thus identified clinics from atric treatment through intensified cooperation across three different geographic areas of the Central Denmark scientific disciplines and sector boundaries . Region. We included a group practice located in an Møller et al. BMC Family Practice (2018) 19:78 Page 3 of 12 urban area in close vicinity of specialised psychiatry, a relate the single-case study experiences with her experi- single-handed practice located in a rural area in long ences with the DCCM in general. distance from specialised psychiatry, and a large group practice located in another rural area a bit closer to Data collection specialised psychiatry. Some GPs were known for their Observational studies involvement in quality development and continuing To explore current clinical practices and establish an un- education, whereas others were not. We included 10 derstanding of the context of mental health treatment, we GPs in our observational studies. Nine GPs participated observed a number of GP-patient consultations and prac- in semi-structured interviews (Table 1). None of the GPs tice nurse-patient consultations in the multi-practice were personally or professionally acquainted with the study and in the single case study before the launch of the anthropologist performing observations and interviews. DCCM. The first author (MM) spent 2–6days ineach For the single-case study, we recruited a GP among included clinic to explore how GPs treat patients with the 15–20 GPs attending the information meeting various degrees of mental health problems. In the hosted by the DCCM project managers in one of the multi-practice study, we conducted a total of approximately participating municipalities. The recruited GP runs a 80 h of observations; these were evenly distributed between single-handed practice located in an urban area with the the GPs. Consequently, more hours of observations were assistance of a receptionist and a medical intern. done in clinics with more GPs. The care manager affiliated with the GP in the To get insight into various DCCM components and to single-case study also agreed to participate in the deepen our understanding of both the context and spe- single-case study. The care manager also collaborated cific barriers and enablers, we followed one patient in with other GPs participating in the DCCM. Her partici- the single-case study throughout the entire collaborative pation was cleared with the DCCM management. A sec- care treatment period through multiple observations and ond care manager was recruited at the DCCM interviews with both GP and care manager. We con- information meeting in the other DCCM participating ducted participant observations of their collaborative municipality. The experiences of the second care man- work through each step of the intervention, including ager helped qualify the findings in the single-case study their initial preparation meeting, the first shared consult- and provided valuable insights into the workings of the ation, the individual sessions between patient and care DCCM. Moreover, since care managers involved in the manager, and the final shared consultation between pa- DCCM collaborated with several general practitioners, tient, GP and care manager. After each consultation, the the care manager in the single-case study was able to care manager and/or the GP were asked to comment on Table 1 Study participants and their participation in research activities Participants Practice type Semi-structured Follow-up Ethnographic Direct Participation interview interview conversation observation in DCCM Single-case study GP 1, practice 1 Single-handed, urban x x x x x Care manager 1 x x x x x Care manager 2 x x x Multi-practice study GP 2, practice 2 Group, rural x x x GP 3, practice 2 Group, rural x x x GP 4, practice 2 Group, rural x x x GP 5, practice 2 Group, rural Cancelled x x GP 6, practice 2 Group, rural Left job x x GP 7, practice 3 Solo, rural x x x GP 8, practice 4 Group, urban x x x GP 9, practice 4 Group, urban x x x GP 10 Group, urban x GP 11 Group, urban x Total 9 2 11 10 3 Abbreviations: GP general practitioner, DCCM Danish Collaborative Care Model Møller et al. BMC Family Practice (2018) 19:78 Page 4 of 12 it. The patients also agreed to participate in observations of the care manager in the single-case study are not ex- and interviews. Findings related to the patient perspec- clusively related to the single-case study. Letting the care tive will be published elsewhere. manager draw on the experiences with the DCCM In both studies, observations of treatment practices in- across various general practice settings helped broaden cluded face-to-face encounters and consultations by the insight into barriers and enablers related to the phone or email. Observations focused on interactions/ DCCM. The same broadening of insights was gained by relations, treatment options, cross-sectorial cooperation, including the perspectives of the second care manager. treatment decisions and GP-patient negotiation thereof. Interview guides for interviews with care managers and All observations were initially jotted down and later GPs are shown as Additional file 1. extended into written field notes [25, 26]. To eliminate potential misunderstandings, observations were followed Analytic procedures up by short ethnographic conversations with the GP, The analysis was conducted in an iterative process to care manager or patient, and sometimes with all of them closely connect data collection and analysis [26, 27]. The in individual informal conversations. thematic analysis began from the initial review of litera- ture and the preparation of research questions, over data Interviews collection, followed by pattern identification, interpret- For the multi-practice study, we developed separate ation and the final writing of this paper. Each interview interview guides for GPs and clinic staff for the transcript was read and reread to perform an initial semi-structured interviews. Interviews focused on expe- coding. Key phrases and themes were compared across riences with and perspectives on existing treatment pos- transcripts. Field notes based on observations in clinical sibilities and limitations both in general practice and in settings and at DCCM recruitment meetings were in- collaboration with specialised psychiatric care. In the cluded to provide a contextual framework for under- single-case study, we performed two semi-structured in- standing the barriers. This procedure identified terviews with the GP and the care manager: one before emerging insights and dominating themes , which initiating the DDCM and one follow-up interview to were discussed and condensed by the research team. provide in-depth insight into their views on enablers and Scope and design of data collection were discussed barriers for the implementation and function of the throughout the course of fieldwork by the first and last intervention. Furthermore, all GPs across the two studies authors. The first author conducted the initial coding were asked which kind of support and treatment options and identification of themes. Final identification of were needed to improve the treatment of patients with themes, analytical choices and the framing of the paper mental health problems and the role of collaborative were discussed by the research team before the writing care in this. of the first draft and in several iterations hereafter. All Interviews were conducted in the clinics during lunch authors have contributed substantially to the writing of break or after work. One practice nurse and one recep- the manuscript. tionist from the rural single-handed practice participated in a group interview. Two practice nurses from the other Ethical considerations two practices participated in ethnographic conversations. The study followed the Code of Ethics by the American Additional data was gathered and observations qualified Anthropological Association . The study was by the GPs and the clinic staff through informal ethno- approved by the Danish Data Protection Agency (file graphic conversations during the first author’s presence number: 2014–41-3207), and data collection and data in the practices, usually between patient consultations handling were performed in accordance with their and during lunch break . Interviews were recorded guidelines. According to Danish law, no ethical approval and transcribed verbatim. Practice nurse and secretary from the regional Committee on Health Research Ethics perspectives are not addressed in this paper. was not needed for this study as no biomedical interven- Interviews with care managers centred on their experi- tion was performed. ences with usual treatment and usual cooperation with In accordance with existing guidelines and research general practice (not related to the DCCM), on their ex- ethics, all GPs, staff, and care managers received written pectations to the DCCM and their actual experiences information about the study and gave verbal consent with the DCCM. Both care managers collaborated with before initiation of the study. All participants were in- several different DCCM-participating practices; this formed about the purpose of the study, anonymity, that offered additional, although second-hand, data on GP participation was voluntary, and that participants could attitudes towards the DCCM. The care managers also withdraw from the study at any time. In addition, infor- provided their views on DCCM organization and imple- mal meetings before initiation of study activities were mentation in various GP settings. Thus, the perspectives held with the participating general practices to inform Møller et al. BMC Family Practice (2018) 19:78 Page 5 of 12 GPs and staff about the study and its implications. handed them over to the project [the DCCM] for a Written information about the study was posted in the while” (care manager). waiting areas of the general practices. The data material produced and analyzed as part of Improved access to treatment and “free” treatment were the study is not publicly available due to ethical consid- important enablers among GPs and a reason for the erations. We constantly strive to ensure the full anonym- single-case GP to join the DCCM: ity of participants, and we respect the verbal agreements made with the participants. As they were not asked for “I joined the project to have an offer to patients who permission to make entire raw data material publically cannot afford a private psychologist and who are not available, we find it ethically inappropriate to make sick enough for treatment at the psychiatric hospital interview transcripts and observational study notes pub- ambulatories” (GP, single-case study). licly available. Data material is available from the corre- sponding author on reasonable request. The DCCM also targeted another GP concern; the exist- ing reimbursement structures for provision of conversa- Results tional therapy to patients with mild/moderate anxiety Three main sub-themes emerged from the analysis and depression. Danish GPs receive reimbursement for across the single-case study and the multi-practice conversational therapy only if the patient receives at study: 1) DCCM enablers, 2) DCCM barriers and 3) least two consultations. If a patient does not return for a Who needs collaborative care? The relevance of target second conversational therapy consultation, the GP is groups. not entitled to reimbursement for the time spent on the first consultation; this can have a negative impact on the DCCM enablers motivation of some GPs: Access to “free” treatment and workload reduction GPs in both studies were positive towards a new treat- “I allocate extra time and then maybe the problem is ment model addressing the growing number of patients solved or the patient doesn’t show up [for the next with mild/moderate depression and anxiety, as these appointment], so I don’t get the reimbursement for the patients are time-consuming: extra time spent…so I think rather despondently that I won’t engage in that” (GP, single-case study). “It would in fact be a relief to many general practices, no doubt about it, if you could [refer patients]” The DCCM does not only accommodate workload is- (GP, multi-practice study). sues concerning the many patients with mild/moderate anxiety and depression; it also tends to encompass the frustrations related to reimbursement structures. “I could easily give up those conversations [conversational therapy]…because we are under quite Quality of treatment a pressure staff-wise, so it [treatment by a care In addition to workload issues, some GPs expected the manager] would be a really nice relief. If I had lots of CC treatment to improve the quality of care because the time I might not give it [conversational therapy] up” DCCM allowed more time for systematic cognitive (GP, multi-practice study). conversational therapy. In the single-case study, the GP reported the impression that the care manager in fact Most GPs expressed frustration about limited referral improved the quality of care: options for these patients and long waiting lists for treat- ment by private practising psychologists. The care man- “Well, she [care manager] has more time, and I think agers added that the DCCM meets a need in general she works more thoroughly than I do. I don’tdoit as practice for patients who are not sick enough for treat- thoroughly” (GP, single-case study). ment in specialized psychiatric ambulatories. A care manager explained: In line with this, a GP in the multi-practice study reflected on the advantages of collaborative care: “There are certainly not any patients [included in the DCCM] with severe depression, and then they have no “It would be really nice to feel that they [care chance of being offered treatment in a psychiatric managers] are competent and hired directly to deal ambulatory. Alternatively, the GP should treat…and with this and to have allocated time slots, where this to my knowledge most GPs have done… these patients person [care manager] could offer half an hour... And with conversational therapy, and now they have this person [care manager] could reach some level of Møller et al. BMC Family Practice (2018) 19:78 Page 6 of 12 overview of our patients, and get some kind of Approaching general practice in pragmatic ways and fit- knowledge about these patients and have a ting the DCCM to the individual GP’s preferences sup- continuous relation to them…that would be ported the implementation of the DCCM. The care fantastic, and it would take off some of the pressure” managers had to adapt to different appointment systems, (GP, multi-practice study). lack of office space, different therapeutic approaches and varying levels of engagement because they worked in Most GPs across our studies agreed that the interest and many different clinics. The care managers also adjusted expertise in mental health treatment vary immensely to the priorities of the GPs. For example, the GPs were among GPs and that improved referral options would to complete standard questionnaires with the patients, benefit both the GPs and their patients: but the care managers took on this task when this did not happen although it deviated from DCCM “protocol” “When you open a referral option, most GPs who do to get things done and avoid conflict: not find it interesting to work with mental health treatment will hurry to refer [their patients]. That is “We [the care managers] do take it [the workload of great, because everybody would be happy… And not GPs] into consideration. Maybe it is overstated… I everyone [every GP] has the competencies required in have defended this [doing the GP’s task] to myself by this field, or an interest…” (GP, multi-practice study). thinking that it is extremely important that we get started well and that we get to know each other, and The lack of interest among some GPs combined with that I don’t appear too rigid and too insisting” the need for improved quality in this field and the (care manager). experience/knowhow of care managers could thus be another enabling factor for collaborative care. The flexible and pragmatic approach taken by the care managers enabled implementation of the DCCM. The Training and supervision by psychiatrist ability to juggle GP needs and adapt DCCM elements to GPs at the DCCM information meetings showed interest different organisations made the care manager an asset in the possibility for supervision and/or training by spe- in collaborative care. cialized psychiatrists. They especially requested brush-up courses on psychoactive drugs and regular supervision in DCCM barriers peer groups lead by a psychiatrist. In the multi-practice Organizational and practical barriers study, GPs also stressed the importance of collaborating The DCCM caused several logistic and organizational with a psychiatrist for specialised advice on patients who problems that (although simple) caused trouble in the do not respond to the usual treatment in general practice. implementation. First, the referral procedures were inad- The same was expressed by the GP in the single-case equately described: study, but eight months after the DCCM launch she had not yet been introduced to brush-up courses or supervi- “It caused me trouble. The first few times I just sion by a psychiatrist. referred [the patient] to the psychiatric hospital with a The GPs’ need for upgrading their psychiatric compe- note saying that this person wanted to participate in tencies might be an enabling factor of the DCCM if a collaborative care, and then they declined because psychiatrist facilitates training and supervision. referral requires a scheduled time for consultation with the care manager…and I don’t think I have been Care managers as implementers properly informed about that” (GP, single-case study). The care managers proved to be translators of the over- all idea of the DCCM. This task was not a described Second, there was lack of information about the standardized element of the DCCM, which left the care practical procedures on how to obtain remuneration: managers to pragmatic maneuvering. Observations of GP-care manager interactions and interviews with the “We have these reimbursement options which I have two care managers indicated that the care managers tried to apply, and then they are rejected, apparently aimed to facilitate a smooth implementation of collab- because I am not registered in the care manager orative care, but they also acknowledged the challenge project. Well. Who do I register with? So I just had to of entering general practice as outsiders: contact them again and claim to be registered in the project and then they accepted…I have contacted the “…it is about how you enter general practice, you enter project manager twice about this, and he hasn’t their territory, and you come as a guest. So you have replied…we need some follow-up on the information to tread cautiously …” (care manager). meeting” (Single-case GP). Møller et al. BMC Family Practice (2018) 19:78 Page 7 of 12 Third, the DCCM increased the GP workload as it The quote could be read as lack of confidence in the ex- required shared consultations with the care manager, pertise of psychiatric nurses, but it also expresses a frus- coordination and booking of consultations, and setting tration in the GP towards what seems to be gatekeeping up appointments for patients. It also proved difficult to by psychiatric nurses when the GP needs assistance by a organise a consultation room for the care manager: psychiatrist. This frustration was shared by most GPs in both the single-case study and the multi-practice study. “Well, there is all the hassle; everything I have Despite an a priori scepticism in some GPs, the GP in attended…information meetings, meetings [with the the single-case study expressed that her clinical cooper- care manager] here and…yes, making sure the ation with the care manager in the DCCM ran smoothly consultation room is ready. There is a lot I have to do and that the care manager seemed to be competent. The extra…” (GP, single-case study). care manager herself experienced a positive attitude towards her in most of the DCCM practices she collabo- In line with this, a care manager stated:“They don’t have rated with, and she found that the GPs actually a spare room, so there is a lot of logistics and planning welcomed interdisciplinary exchange: in it for us as care managers. My calendar, the patient’s calendar and then the consultation room, the “What I hear, at least from three of the GPs that I [psycho-education] classes, conferences. It is a huge talked to, is simply a need for a professional back-and- logistic work. And some patients go to school and some forth; the dialogue, ‘I see the case [the patient] like this go to work, which you must also show consideration and this: How do you see it?” (care manager). for” (care manager). Despite the willingness in the GP to provide a profes- In the follow-up interview, the GP in the single-case sional back-and-forth, the DCCM set-up did not facili- study concluded that she might not want to volunteer tate a closer cooperation according to the findings in the and invest the extra resources in a similar future collab- single-case study. Except for brief ad hoc exchanges of orative care model. treatment plans and coordination of shared consulta- The findings indicate that the DCCM was inad- tions, both the GP and the care manager considered it equately prepared; the project did not get off to a more as a transfer of the patient. In that sense, as also smooth start, and this challenged successful implemen- reported by the single-case GP, the collaborative care tation and continuation of the intervention. project facilitated only a shallow relationship across sectors and disciplines. Cross-sectorial collaboration The GP stated: An important aim of the DCCM was to improve the cooperation and communication between primary and “I don’t cooperate a lot with the care manager in the secondary care through facilitation by care managers. sense that we exchange experiences and stuff … I get to Although most GPs in both studies embraced the idea of know the care manager a bit through the initial and collaboration with specialised psychiatry, several GPs at last consultations. But proper exchange of knowledge the recruitment meeting expressed a priori scepticism and experience on how to handle these patients or towards handing over their patients to care managers. sharing her tricks with me…that doesn’t really Furthermore, some were reluctant to attend training happen” (GP, single-case study). courses facilitated by care managers. This a priori scepti- cism towards the role of care managers in cross-sectorial Both care managers agreed. One care manager concluded: collaborative care can be interpreted as a mental barrier to the DCCM and as a sign of a lack of interprofessional “The way that our shared consultations worked meant respect. In line with this, a GP in the multi-practice that they actually had the characteristics of a transfer study expressed a general frustration towards existing [of the patient] with the GP telling the patient, in my visitation procedures where psychiatric nurses assess presence, that ‘I have told this and this [to the care whether a patient referred by a GP is eligible for special- manager] about your situation’” (care manager). ized psychiatric care: The other care manager said: “It can be frustrating when you refer a patient who doesn’t respond to the treatment, and then it is a “…and my conclusion right now is that it is hard nurse who assesses the patient and rejects the formetosee,inrelationtothesepatients,why I patient without having a psychiatrist involved” should claim a need for a shared consultation” (GP, multi-practice study). (care manager). Møller et al. BMC Family Practice (2018) 19:78 Page 8 of 12 The care managers and the GP agreed that written com- would need their own consultation room, or we should munication would have been sufficient to exchange the receive some kind of reimbursement, or we should rent necessary information and that treatment of patients by out [the consultation room], and then we would have care managers could have been conducted in a psychiatric the trouble with administrating their booking of outpatient ambulatory, but the collaborative care part consultations and things like that. So… otherwise, I would then have vanished. One care manager explained: would be the one paying for the patients’ free treatment, and then we wouldn’t do it. It “They [the patients] might as well get treatment in the [collaborative care] has to be solved [done] somewhere ambulatory, but then there would be no collaboration else, I think…then it would be really great, but yes… with general practitioners” (care manager). they should create an ambulatory for it because it would be swarmed with patients” (GP, multi-practice The GP in the single-case study acknowledged that in- study). creased cooperation across sectors was time consuming, the multi-practice study and she was unsure if she would be willing to invest more time in it. One important barrier thus seems to be that the In addition to organisational and logistical issues, DCCM involved extra (rather than reduced) workload in several other important barriers to the DCCM were general practice without compensating the GPs for this identified: the unclear character and purpose of the additional work. cross-disciplinary cooperation, a priori scepticism in Several barriers to the DCCM were experienced by the many GPs towards care manager-facilitated training, and single-case GP and perceived by the GPs in the limited actual exchange of knowledge and experience. multi-practice study. Although the care managers played a central role in the treatment and collaboration related Sustainability to the DCCM, none of the GPs in the multi-practice The single-case study revealed a number of organisational, study requested elaborated cooperation with a care man- practical, logistic and resource-related barriers to the ager from specialised psychiatry for patients with mild/ DCCM when meeting the realities in general practice. moderate anxiety and depression. The GPs seemed to The care managers were intended to provide treatment of expect benefits from collaborative care in terms of patients in up to 10 general practices . This set-up improved access to high-quality treatment rather than challenged both coordination and logistics, and the care increased collaboration with external partners, including managers experienced a considerable waste of time when a care manager situated in the clinic. Consequently, the commuting between different general practices. Due to support gained by the DDCM was not considered to these barriers and the lack of capacity in specialized psy- sufficiently justify the time spent. chiatric care, the care managers doubted that the model would be sustainable beyond the project period: Who needs collaborative care? The relevance of target group DCCM enablers and barriers were partly related to the “There wouldn’t be enough qualified nurses if you way that the GPs perceived mental healthcare for differ- deploy this model everywhere. So I can’t imagine how ent patient groups, and how the GPs tended to divide this model should be implemented permanently in its these into categories according to the severity of the current form. I really don’t” (care manager). patients’ mental illness and their treatments needs. When asked which of patients with mental health The GPs in the multi-practice study appreciated the problems they found the most challenging to treat, the potential of the DCCM to free time, but some also GPs (across both studies) referred to specific diagnoses, feared that it might “steal time” due to planning and co- and they categorised the patients into two main groups ordination with the care manager. The care managers based on the severity of conditions: 1) a large group of found it time-consuming to get the DCCM model patients with mild/moderate mental health problems started and running (also after the run-in period). and 2) a small group of patients with severe mental The logistical problems of finding a vacant consult- health problems (e.g. severe depression, schizophrenia ation room for the care manager were also mentioned as and other psychotic conditions) and higher risk of som- a potential barrier to the sustainability of the proposed atic comorbidity and complex health problems: DCCM by GPs in the multi-practice study: “There are not as many patients with schizophrenia as “It [treatment by care manager] should not be done patients with anxiety and depression so it is a matter here. Then they [the care managers] would have to be of prioritising resources. The schizophrenic patients are here all the time. Then they [the care managers] left more to themselves; they are a more vulnerable Møller et al. BMC Family Practice (2018) 19:78 Page 9 of 12 group who rarely consult their GP. Patients with registration and reimbursement procedures, limited genu- anxiety and depression consult their GP regularly. And ine collaboration and knowledge sharing, much GP time often they have relatives who support them, whereas spent on practical issues and little availability of consult- the schizophrenic patients are more socially isolated” ation rooms. Another barrier perceived by the GPs in the (GP, multi-practice study). multi-practice study and, to some extent, by the single-case GP was the lack of confidence among GPs in The GPs considered treatment of mild/moderate mental the importance of investing time and resources for health conditions to be a central GP task, but the num- enhanced cooperation across sectors and disciplines on ber of patients is increasing, and the GPs do not have this patient group; most participants questioned the sus- the capacity to handle the volume of patients. The GPs tainability of the DCCM. emphasized that the complexity of the mental/somatic/ As an overall framework for treatment of patients with social problems in patients with severe mental health mental health problems, the DCCM fits well with the conditions makes them an especially vulnerable group; dominant discourse in modern western healthcare pol- they are the most challenging to care for in general prac- icy, which focuses on coherent, cross-disciplinary and tice and have the highest need for improved treatment. cross-sectorial healthcare set-ups. However, it has been argued elsewhere that the societal focus on cooperation “Where are the schizophrenic patients now? The across traditional sector boundaries cannot be expected number of hospital beds have been cut back, they are to be mirrored in everyday life [29, 30]. This may explain discharged, and they disappear into nothing, and we why carefully developed collaborative care models may are not in control of them…it happens too often” work under experimental conditions but not in real-life (GP, multi-practice study). situations, as we also found in this study. Such models collide with daily life in general practice, where GPs When asked about the potential of a collaborative care must make meaningful choices with limited resources; model that integrates psychiatric and general practice all activities are subject to standardised fee-for-service treatment, the GPs agreed that the patients with severe agreements, and all choices are influenced by competing mental health problems and somatic comorbidity have tasks, different patient needs and individual preferences the highest need for cross-disciplinary and collaborative among GPs and patients. treatment. The complexity of health conditions in these If the GPs consider improved referral options and not patients calls for new ways to ensure integrated mental collaboration with a care manager to be the most im- healthcare that meets their somatic, psychiatric and so- portant for this patient group, the DCCM should reflect cial needs, whereas the GPs did not see a need for en- this. The logistic and financial constraints also need to hanced collaboration with specialised psychiatry on be addressed. The GPs were not per se against the con- patients with mild/moderate mental problems. For this cept of collaboration, but they wanted cooperation and patient group, they requested a fast track to high-quality communication to fit with their practice life, and they treatment in cases when usual care fails. This discord- were concerned about potential side effects in terms of ance between the target group of the DCCM and the increased workload. In particular, the introduction of an actual and perceived needs of collaborative care suggests external care manager in general practice revealed logis- a major motivational barrier for successful implementa- tic, practical and financial barriers. tion of the DCCM. The practical daily life of general practice should be taken as a starting point when a revised DCCM is devel- Discussion oped and implemented. Marshall states that “improve- Main findings ment initiatives are sometimes planned on the hard high We did a purposeful sampling for the multi-practice ground, but they are put into effect in the swampy low- study in order to include potential variations in GP per- lands” . Therefore, central stakeholders, such as GPs ceptions and their need for collaborative care. However, and patients, need to be deeply involved in the develop- we found very limited variation. Therefore, the analysis ment of new collaborative models to ensure that they centred on the thematic trends across the two studies. are designed to meet the existing organisational, cultural The introduction of the DCCM for mild/moderate anx- and structural realities . Our findings also indicate iety and depression was promoted by a need in the GPs the importance of developing a DCCM in a systematic for better access to free high-quality treatment and way, including pilot testing as described by the Medical pragmatic manoeuvrings by the care managers. Major Research Council in the guidance on the development of barriers to the implementation of the DCCM, as complex interventions, in order to adjust new interven- experienced by the single-case GP and the care managers, tion prototypes to the clinical reality of general practice were organisational deficiencies concerning patient before presenting a new model [33, 34]. Møller et al. BMC Family Practice (2018) 19:78 Page 10 of 12 The two-armed research design that combines the general practice settings and with numerous patients DCCM single-case study with a multi-practice study during the first six months of the intervention. More- allows the discrepancies between model and reality to over, the model was observed throughout the entire emerge. We wanted to take a bottom-up approach to process (including information meetings, initial meetings the need for collaborative care in contrast to the between care manager and GP and the entire course of top-down developed DCCM in order to discuss and treatment) and through a combination of before and problematise the introduction of new models without in- after interviews. This thorough exploration added volving all relevant stakeholders. strengths and depth to the study by providing broad and multi-facetted perspectives on the functioning of the Comparison with other studies model. Furthermore, many of the experiences from the Our findings are consistent with other studies on the single-case study accorded with the findings from the implementation of collaborative care in both the UK, the multi-practice study. US and the Netherlands, which show that it can be chal- lenging to implement research-designed collaborative Conclusion care in real-life general practice settings [9, 12, 13, 16, The increasing number of patients with mild/moderate 35, 36]. Other studies have found that collaborative care anxiety and depression in general practice supports the can be implemented if financial barriers are reduced, ef- implementation of collaborative care in mental health fective training and facilitation are provided, a common treatment. Additionally, the workload-reducing potential mental health model is developed, and new care pro- of care manager-led treatment did in fact constitute an cesses are introduced . The most important factors actual and perceived enabling factor of the DCCM. found in other studies using the Normalisation Process Nevertheless, there are a number of experienced and Theory as the analytical frame were shared understand- suggested barriers to the sustainability of the DCCM in ing of mental illness between GPs and care managers its current shape. Firstly, the GPs in this study did not and agreement on the division of tasks [4, 12]. Our find- request increased cooperation with specialised psychi- ings support the crucial role of developing a common atric care (represented by care managers) for patients understanding of mental health conditions and their with less severe mental conditions although they see a treatment as well as targeting logistic and financial bar- growing need for improved referral options for this pa- riers . Adding to this, our study provides insight into tient group. Secondly, our single-case study suggests a the GPs’ distinction between mild/moderate and severe number of practical and logistic barriers to the DCCM. mental illness. Treatment of mild/moderate conditions Thus, the prospected barriers and concerns of GPs in can be optimised within general practice and through the multi-practice study who had not experienced the more referral options to psychological therapy, whereas implementation of the DCCM supplement the findings collaborative care is perceived to be more suitable for se- from the single-case study. vere cases. GPs and patients should be key in the development of fu- ture collaborative care models. The pragmatic reality of Strengths and limitations general practice, cross-sectorial differences in the percep- The present study has limitations with regard to general- tions of target groups, (lack of) needs for more cooperation, isability. The general practices in the study may not time constraints and identification of practical and logistical cover all types of practices, and the included GPs were barriers should be considered to ensure successful imple- known to be among the most progressive and active mentation for all involved parties. Therefore, we argue that compared to other general practice clinics. Therefore, there is a need to ask: Who needs collaborative care? we would expect them to have a more positive attitude to new models of care delivery, such as a DCCM. Conse- Additional file quently, our findings may report a more positive attitude to the DCCM than would have been found in the gen- Additional file 1: Interview guides. (PDF 126 kb) eral population of GPs. Financial constraints allowed us to include only one Abbreviations GP clinic participating in the DCCM. Thus, the findings DCCM: Danish collaborative care model; GP: General practitioner reflect the experiences from only one collaborative care setting and may not be representative of collaborative Acknowledgements care experiences in other general practices. However, a The authors wish to thank all the GPs and care managers who generously shared their time and experience for this study. We also owe thanks to the broader perspective on the function of the model was patients and clinic staff who participated in the study although their ensured through the inclusion of two care managers as contributions do not form part of this paper. Finally, we thank the Lundbeck both of these interacted with several GPs across different Foundation for funding the study. Møller et al. BMC Family Practice (2018) 19:78 Page 11 of 12 Funding 4. Coupe N, Anderson E, Gask L, Sykes P, Richards DA, Chew-Graham C. The study was supported by an unrestricted grant from the Lundbeck Facilitating professional liaison in collaborative care for depression in UK Foundation (grant number: R155–2012-11280) and conducted in primary care; a qualitative study utilising normalisation process theory. BMC collaboration with the MEPRICA research group at the Research Unit for Fam Pract. 2014;15:78. General Practice and Section for General Medical Practice, Department of 5. Barley EA, Haddad M, Simmonds R, Fortune Z, Walters P, Murray J, et al. The Public Health, Aarhus University, Denmark. UPBEAT depression and coronary heart disease programme: using the UK Medical Research Council framework to design a nurse-led complex Availability of data and materials intervention for use in primary care. BMC Fam Pract. 2012;13:119. Although no official ethical approval was required, the dataset generated 6. Katon WJ, Lin EH, Von Korff M, Ciechanowski P, Ludman EJ, Young B, et al. and/or analysed during the current study is not publicly available due to Collaborative care for patients with depression and chronic illnesses. N Engl ethical considerations. We constantly strive to secure full anonymity of our J Med. 2010;363(27):2611–20. informants and respect the verbal agreements made with informants. Since 7. Solberg LI, Crain AL, Jaeckels N, Ohnsorg KA, Margolis KL, Beck A, et al. The informants were not asked for permission to make entire raw data sets DIAMOND initiative: implementing collaborative care for depression in 75 publically available, we find it ethically inappropriate to make interview primary care clinics. Implement Sci. 2013;8:135. transcripts and observational study notes public available. Datasets are 8. Byng R, Norman I, Redfern S, Jones R. Exposing the key functions of a available from the corresponding author on reasonable request. complex intervention for shared care in mental health: case study of a process evaluation. BMC Health Serv Res. 2008;8:274. Authors’ contributions 9. Chwastiak L, Vanderlip E, Katon W. Treating complexity: collaborative care MCRM designed the study, collected the data, analysed and interpreted the for multiple chronic conditions. Int Rev Psychiatry. 2014;26(6):638–47. data, and drafted the manuscript. FB supervised the study and participated 10. Archer J, Bower P, Gilbody S, Lovell K, Richards D, Gask L, et al. Collaborative in the design of the study, the analysis of data and the drafting of the care for depression and anxiety problems (review). Cochrane Database Syst manuscript. AM contributed to the analysis of data and drafting of the Rev. 2012;10:CD006525. manuscript. All authors have read and approved the final manuscript. 11. Whitebird RR, Margolis KL, Asche SE, Trangle MA, Wineman AP. Barriers to improving general practice of depression. Perspectives of medical group Ethics approval and consent to participate leaders. Qual Health Res. 2013;23(6):805. The study followed the Code of Ethics by the American Anthropological 12. Knowles SE, Chew-Graham C, Coupe N, Adeyemi I, Keyworth C, Thampy H, Association . The study was approved by the Danish Data Protection et al. Better together? A naturalistic qualitative study of inter-professional Agency (file number: 2014–41-3207), and data collection and data handling working in collaborative care for co-morbid depression and physical health were performed in accordance with their guidelines. According to Danish problems. Implement Sci. 2013;8:110. law, no ethical approval from the regional Committee on Health Research 13. Knowles SE, Chew-Graham C, Adeyemi I, Coupe N, Coventry PA. Managing Ethics was not needed for this study as no biomedical intervention was depression in people with multimorbidity: a qualitative evaluation of an performed. integrated collaborative care model. BMC Fam Pract. 2015;16:32. In accordance with existing guidelines and research ethics, all GPs, staff, and 14. Hermens ML, Muntingh A, Franx G, van Splunteren PT, Nuyen J. Stepped care managers received written information about the study and gave verbal care for depression is easy to recommend, but harder to implement: results consent before initiation of the study. All participants were informed about of an explorative study within primary care in the Netherlands. BMC Fam the purpose of the study, anonymity, that participation was voluntary, and Pract. 2014;15:5. that participants could withdraw from the study at any time. In addition, 15. Hauge-Helgestad A, Johansen KS, Hansen J. Behandling af mennesker med informal meetings before initiation of study activities were held with the angst og depression. Kortlægning af behandlingsfeltet og diskussion af participating general practices to inform GPs and staff about the study and perspektiverne ved indførelse af collaborative care [Treatment of patients its implications. Written information about the study was posted in the with anxiety and depression. A mapping of the field of treatment and a waiting areas of the general practices. discussion of perspectives on the implementation of collaborative care. In The data material produced and analyzed as part of the study is not publicly Danish.]. DSI. 2012. https://www.kora.dk/media/763524/behandling-af- available due to ethical considerations. We constantly strive to ensure the full mennesker-med-angst-og-depression.pdf. Accessed 14 Dec 2017. anonymity of participants, and we respect the verbal agreements made with 16. Overbeck G, Davidsen AS, Kousgaard MB. Enablers and barriers to the participants. As they were not asked for permission to make entire raw implementing collaborative care for anxiety and depression: a systematic data material publically available, we find it ethically inappropriate to make qualitative review. Implement Sci. 2016;11(1):165. interview transcripts and observational study notes publicly available. Data 17. Katon W, Unutzer J. Collaborative care models for depression: time to move material is available from the corresponding author on reasonable request. from evidence to practice. Arch Intern Med. 2006;166(21):2304–6. 18. Eplov LF, Lundsteen M, Birket-Smith M. Shared care for ikke-psykotiske Competing interests sygdomme. Anbefalinger på baggrund af en systematisk The authors declare to have no competing interests. litteraturundersøgelse [Shared care for non-psychotic mental illness. Recommendations based on a systematic review. In Danish]. Danish regions. 2009. Publisher’sNote 19. Richards DA, Hill JJ, Gask L, Lovell K, Chew-Graham C, Bower P, et al. Clinical Springer Nature remains neutral with regard to jurisdictional claims in effectiveness of collaborative care for depression in UK primary care published maps and institutional affiliations. (CADET): cluster randomised controlled trial. BMJ. 2013;347:f4913. 20. Brinck-Claussen UO, Curth NK, Davidsen AS, Mikkelsen JH, Lau ME, Received: 20 December 2017 Accepted: 18 May 2018 Lundsteen M, et al. Collaborative care for depression in general practice: study protocol for a randomised controlled trial. Trials. 2017;18(1):344. 21. Curth NK, Brinck-Claussen UO, Davidsen AS, Lau ME, Lundsteen M, References Mikkelsen JH, et al. Collaborative care for panic disorder, generalised anxiety 1. van der Feltz-Cornelis CM. Ten years of integrated care for mental disorders disorder and social phobia in general practice: study protocol for three in the Netherlands. Int J Integr Care 2011;11 Spec Ed:e015. cluster-randomised, superiority trials. Trials. 2017;18(1):382. 2. Huijbregts KM, de Jong FJ, van Marwijk HW, Beekman AT, Ader HJ, 22. Gask L, Bower P, Lovell K, Escott D, Archer J, Gilbody S, et al. What work has Hakkaart-van Roijen L, et al. A target-driven collaborative care model for to be done to implement collaborative care for depression? Process major depressive disorder is effective in primary care in the Netherlands. A evaluation of a trial utilizing the normalization process model. Implement randomized clinical trial from the depression initiative. J Affect Disord. Sci. 2010;5:15. 2013;146(3):328–37. 3. Muntingh A, van der Feltz-Cornelis C, van Marwijk H, Spinhoven P, 23. Central Denmark Region. Revideret ansøgning til SATS puljen til styrket Assendelft W, de Waal M, et al. effectiveness of collaborative stepped care samarbejde mellem behandlingspsykiatrien og almen praksis (shared care). for anxiety disorders in primary care: a pragmatic cluster randomised Projekt “Afprøvning af collaborative care-modellen i region Midtjylland” controlled trial. Psychother Psychosom 2014;83(1):37–44. [revised grant application regarding increased collaboration between Møller et al. BMC Family Practice (2018) 19:78 Page 12 of 12 psychiatry and general practice (shared care). Project: "test of the collaborative care model in the Central Denmark region. In Danish]. Central Denmark region. 2012. https://www.rm.dk/api/NewESDHBlock/ DownloadFile?agendaPath=%5C%5CRMAPPS0221.onerm.dk%5Ccms01- ext%5CESDH%20Data%5CRM_Internet%5CDagsordener%5CUdvalg_ vedroerende_nae%202013%5C08-01-2013%5CAaben_ dagsorden&appendixId=39306. Accessed 14 Dec 2017. 24. Pedersen KM, Andersen JS, Sondergaard J. General practice and primary health care in Denmark. J Am Board Fam Med. 2012;25(Suppl 1):S34–8. 25. Bernard HR. Research methods in anthropology. Qualitative and quantitative approaches. Thousand Oaks: Sage; 1994. 26. Hammersley M, Ethnography AP. Principles in practice. New York: Routledge; 1995. 27. Pope C, Ziebland S, Mays N. Qualitative research in health care. Analysing qualitative data. BMJ. 2000;320(7227):114–6. 28. American Anthropological Association: Statement on Ethics. Principles of Professional Responsibility. 2012.http://ethics.americananthro.org/category/ statement/. Accessed 30 Oct 2017. 29. De Certeau M. The practice of everyday life. Berkeley: University of California Press; 1988. 30. Jöhncke S, Svendsen MN, Whyte SR. Sociale teknologier som antropologisk arbejdsfelt [Social technologies as field of interest in anthropology. In Danish]. In: Hastrup K, editor. Viden om verden. En grundbog i antropologisk analyse [Knowledge about the world. A guide to anthropological analysis. In Danish]. Copenhagen: Hans Reitzels Forlag; 2004. p. 385–408. 31. Marshall M, de Silva D, Cruickshank L, Shand J, Wei L, Anderson J. What we know about designing an effective improvement intervention (but too often fail to put into practice). BMJ Qual Saf. 2017;26(7):578–82. 32. May CR, Johnson M, Finch T. Implementation, context and complexity. Implement Sci. 2016;11(1):141. 33. Craig P, Dieppe P, Macintyre S, Michie S, Nazareth I, Petticrew M. Developing and evaluating complex interventions: the new Medical Research Council guidance. BMJ. 2008;337:a1655. 34. Moore GF, Audrey S, Barker M, Bond L, Bonell C, Hardeman W, et al. Process evaluation of complex interventions: Medical Research Council guidance. BMJ. 2015;350:h1258. 35. Eghaneyan BH, Sanchez K, Mitschke DB. Implementation of a collaborative care model for the treatment of depression and anxiety in a community health center: results from a qualitative case study. J Multidiscip Healthc. 2014;7:503–13. 36. Speyer H, Christian Brix Norgaard H, Birk M, Karlsen M, Storch Jakobsen A, Pedersen K, et al. The CHANGE trial: no superiority of lifestyle coaching plus care coordination plus treatment as usual compared to treatment as usual alone in reducing risk of cardiovascular disease in adults with schizophrenia spectrum disorders and abdominal obesity. World Psychiatry. 2016;15(2):155–65. 37. Katon W. Collaborative depression care models: from development to dissemination. Am J Prev Med. 2012;42(5):550–2.
BMC Family Practice – Springer Journals
Published: May 30, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera