Scholarly article on topic 'Shared care: the barriers encountered by community-based palliative care teams in Ontario, Canada'

Shared care: the barriers encountered by community-based palliative care teams in Ontario, Canada Academic research paper on "Psychology"

CC BY
0
0
Share paper
OECD Field of science
Keywords
{""}

Academic research paper on topic "Shared care: the barriers encountered by community-based palliative care teams in Ontario, Canada"

Health and Social Care in the Community (2012) doi: 10.1111/j.1365-2524.2012.01060.x

Shared care: the barriers encountered by community-based palliative care teams in Ontario, Canada

Lily DeMiglio ma and Allison Williams PhD

School of Geography & Earth Sciences, McMaster University, Hamilton, ON, Canada

Accepted for publication 9 December 2011

Correspondence

Lily DeMiglio

McMaster University, School of Geography & Earth Sciences, 1280 Main Street West, Hamilton, Ontario L8S 4K1, Canada E-mail: demigllf@mcmaster.ca

What is known about this topic

• There is wide support in the international literature for integrated healthcare service models such as shared care because of their cost-effectiveness and team approach to care.

• There is little evidence specific to community-based palliative care teams (CBPCTs) working in a shared care model in Canada and the barriers that CBPCTs face.

What this paper adds

• This qualitative case study indicates that CBPCTs have difficulty establishing shared care partnerships because of multiple barriers, such as securing funding for team members, and the influences and values of stakeholders.

• The identification of barriers faced by CBPCTs is a first step in informing how to better support interprofessional shared care collaborations in community settings.

Abstract

To meet the complex needs of patients requiring palliative care and to deliver holistic end-of-life care to patients and their families, an interprofessional team approach is recommended. Expert palliative care teams work to improve the quality of life of patients and families through pain and symptom management, and psychosocial spiritual and bereavement support. By establishing shared care models in the community setting, teams support primary healthcare providers such as family physicians and community nurses who often have little exposure to palliative care in their training. As a result, palliative care teams strive to improve not only the end-of-life experience of patients and families, but also the palliative care capacity of primary healthcare providers. The aim of this qualitative study was to explore the views and experiences of community-based palliative care team members and key-informants about the barriers involved using a shared care model to provide care in the community. A thematic analysis approach was used to analyse interviews with five community-based palliative care teams and six key-informants, which took place between December 2010 and March 2011. Using the 3-1 framework, this study explores the impacts of Institution-related barriers (i.e. the healthcare system), Interest-related barriers (i.e. motivations of stakeholders) and Idea-related barriers (i.e. values of stakeholders and information/research), on community-based palliative care teams in Ontario, Canada. On the basis of the perspective of team members and key-informants, it is suggested that palliative care teams experience sociopolitical barriers in an effort to establish shared care in the community setting. It is important to examine the barriers encountered by palliative care teams to address how to better develop and sustain them in the community.

Keywords: barriers, community-based palliative care, interprofessional care, qualitative research, shared care, teamwork

Introduction

Given the complex needs of patients at the end of life, the input and expertise of a variety of medical, nursing and allied health professionals is often required. For this reason, and to provide holistic palliative care, a team approach is often recommended (Saunders 2000, Billings 2002, Ferris et al. 2002, Crawford & Price 2003, Meier & Beresford 2008). Although the composition of teams vary, at a minimum they include healthcare pro-

viders with specialist training in palliative care, such as physicians, nurses and social workers, the latter, who usually have expertise in psychosocial spiritual and7or bereavement counselling (Crawford & Price 2003, Marshall et al. 2008). In addition, palliative care teams often vary in terms of the settings in which they practice (Saunders 2000); some teams solely serve hospitals and others work in community settings, such as nursing or long-term care facilities, private residences and hospices.

To better coordinate care and build capacity, community-based palliative care teams (CBPCTs) enter into collaborative partnerships with primary healthcare providers, such as family physicians and community nurses. Primary healthcare providers often lack adequate training in palliative care and CBPCTs assist by sharing their specialist knowledge and skills. A shared care model is established through interprofessional collaborations of this kind whereby healthcare providers assume mutual responsibility for a patient (Moorhead 1995) while capacity building (Mazowita 2004).

International research from Australia, the United States, the United Kingdom and in several European countries such as Italy and Spain has shown that palliative care teams working in integrated service models, including the shared care model, have been widely adopted and effective. Various models have been successful at improving the satisfaction of primary healthcare providers, patients and families while also reducing healthcare costs, such as by decreasing hospital usage at end of life (Hearn & Higginson 1998, Costantini et al. 2003, Higginson et al. 2003, Aggarwal et al. 2006, Brum-ley et al. 2007, Herrera et al. 2007, Zimmermann et al. 2008, Neergaard et al. 2010, Howell et al. 2011). On the other hand, this body of literature has also addressed the challenges and barriers encountered by palliative care teams, such as funding (Billings 2002) and interprofessional collaboration issues (Crawford & Price 2003, Neergaard et al. 2008, 2010, O'Connor & Fisher 2011). In the Canadian context, there is some research specific to the success of palliative care programmes and teams (Bruera et al. 1999,2000, Burge et al. 2005, Fainsinger et al. 2007, Marshall et al. 2008, Lawson et al. 2009, Howell et al. 2011); however, there is a dearth of research on CBPCTs working in a shared care model and the barriers that they face. Consequently, there is a need to fill this gap by examining CBPCTs working in a shared care model from a Canadian perspective. The research presented herein makes a contribution through specifically examining the barriers these teams confront, and by so doing, informs the development of this type of service delivery model in jurisdictions which have yet to adopt it.

This research is timely given that, similar to other developed countries, Canada is experiencing more people dying in community settings than in hospitals because of healthcare restructuring (Wilson et al. 2009); further, the number of deaths are projected to increase as a result of population ageing (Statistics Canada 2010). Taken together, these factors demonstrate that community-based palliative care services are, and will be, increasingly needed. The goal of this study was to explore the perspectives of CBPCTs and key-informants about the barriers involved in providing palliative care

in the community using a shared care model. We identify the barriers in establishing a shared care model of palliative care based on the experiences of five CBPCTs and the views of six key-informants located in Ontario, Canada. Before outlining the study's methods, it is important to provide an overview of the relevant healthcare context.

Healthcare delivery context in Canada

Provinces/territories in Canada have authority over the delivery of healthcare services while the federal government finances health-care and monitors adherence to the principles of the Canada Health Act (Marchildon 2005). There is 100 per cent public payment for all medically necessary hospital and physician services. Most of the provinces/territories in Canada have regionalised health-care whereby health services are planned and delivered according to geography; the province of Ontario is divided into 14 geographically delineated Local Health Integration Networks (LHINs) (Marchil-don 2005). In each LHIN, there are Community Care Access Centres (CCACs), which coordinate community-based healthcare services (Witmer 2000, OACCAC n.d.). Ontario is unique in that service providers, such as forprofit and not-for-profit nursing agencies, compete for CCAC home-care service contracts through a competitive bidding process (Abelson et al. 2004, Woodward et al. 2004).

As part of Ontario's End-of-Life Care Strategy, hospice palliative care networks were implemented in each of the LHINs in 2005 to determine LHIN-specific priorities and to select the most appropriate service delivery models (Dudgeon et al. 2007, Seow et al. 2008). This devolution of responsibility has led to the implementation and endorsement of CBPCTs working in a shared care model in only some of the LHINs; as a result, there is no consistency in palliative care service delivery models across the province. The LHIN area of concern herein has endorsed CBPCTs working in a shared care model, in part based on the success demonstrated by a pre-existing CBPCT serving a region of the LHIN area since the mid-1990s.

The specific location and name of the LHIN has been withheld to protect the confidentiality of participants. While the LHIN serves a vast population, each of the five teams is geographically bound, whereby services are limited to patients living within the boundaries of certain regions within the LHIN. As the five teams currently only serve part of the LHIN area, a large number of patients in the LHIN receive usual care with no access to the expertise of a CBPCT. The hospice palliative care network for the LHIN area anticipates the implementation of an additional six CBPCTs in the areas currently receiv-

ing usual care; the timing of the introduction of these teams has not yet been determined.

Methods

This qualitative study is part of a 1-year longitudinal case study comprised of three phases, all of which examine the barriers and facilitators involved in the implementation of CBPCTs pursuing a shared care model of palliative care service delivery. This study is based on findings from the first phase (i.e. interviews with key-informants) and the second phase (i.e. CBPCT focus groups) specific to the barriers encountered by CBPCTs working in a shared care model. The third and last phase, not reported on here, will encompass stakeholder perspectives. Case study research often involves the investigation of a phenomenon or issue and its contextual influences (Yin 2009). Here, the case study is bounded by the geography of the LHIN, and the phenomenon of interest is the delivery of palliative care services by CBPCTs working in a shared care model.

Participants

Ethics approval from McMaster University was in place prior to data collection, which was conducted between December 2010 and March 2011. Recruitment involved one of the researchers initially meeting with each of the five CBPCTs to present a brief overview of the objectives and design of the case study, including participation logistics. Each of the five CBPCTs was formed at different times and has a unique size and make-up (see Table 1). All of the CBPCTs were comprised of service providers such as physicians and nurses (see Table 2); however, two also included administrators (i.e. hospice directors and administrative assistants). Letters of information and consent forms were distributed at each of these initial meetings, and all team members were invited to participate in a focus group interview, which was scheduled for a later date. In total, 33 team members across the five CBPCTs were invited to participate and 28 accepted the invitation in addition to two medical/nursing residents (i.e. on short-term rotation were included). All team members who showed up to the focus group session were included in the study. In addition, eight key-informants with either expertise on shared care models (e.g. researchers), or indirect involvement in the delivery of palliative care services using a shared care model (e.g. coordinators), were invited to participate in one-on-one interviews (see Table 3). A purposeful sampling strategy (Patton 2002) was used to choose informants who would be able to share their knowledge to help the researchers gain a deeper understanding of shared care models while also enhancing

Table 1 Community-based palliative care teams and focus group participant descriptors

Number of Number of focus

Year identified group

Team formed members participants

1 1996/1997 8 7*

2* 2001 3 3

3 2003/2004 8 7

4 2008 11 10*

5§ 2009/2010 3 3

*The range in the number of identified members varies across teams because of differences in the following: (1) the size of the geographical area served (i.e. smaller areas require fewer members); (2) availability of/accessibility to additional supports from community partners (i.e. sources from hospice, hospital etc.); and (3) individual team members' work status (i.e. parttime vs. full-time etc.).

includes participation from medical/nursing resident on rotation with team at time of interview.

*Three core members comprise Team 2, and while several allied health professionals work alongside the team, they were not identified as core team members.

§Team 5 is the newest team, and additional team members are expected to join.

Table 2 Discipline of focus group participants

Discipline Number

Administration 4

Case Management 5

Medicine 9

Nursing 8

Social Work (includes psychosocial spiritual and 4

bereavement)

Table 3 Area of expertise of Key-Informants

Academia Planning

Key-informant and research Clinical and policy

1 X X

3 X X

4 X X

6 X X

credibility (Baxter & Eyles 1997). Letters of information were sent to key-informants via e-mail requesting that those interested in participating reply directly to the researcher. Six of the eight key-informants invited to be part of the study chose to participate.

Data collection

All interviews (n = 5 focus groups and n = 6 key-informant interviews, see Tables 1-3) were semi-structured in nature and digitally recorded to be transcribed verbatim. For consistency, one researcher conducted all of the interviews. The CBPCTs were asked open-ended questions about their: development, team dynamics, as well as barriers and facilitators to providing palliative care via a shared care model. Each CBPCT was interviewed once at their home base headquarters. The focus group interviews ranged from 60-90 minutes in duration. In total, 30 people took part in the focus groups. Six key-informants were also interviewed at a location of their choosing and interviews ran approximately 60 minutes. Key-informants were asked about their experience with, and understanding of, shared care, as well as what they perceived to be the barriers and facilitators to providing palliative care in the community using a shared care model.

Analysis

The coding of focus group and key-informant interview transcripts involved the application of a political science framework, commonly referred to as the '3-1 framework', which explores the role of institutions, interests and ideas in the policy process (Garrett & Weingast 1993, Hall 1993). This framework has been employed elsewhere in healthcare research as a means of situating and explaining the lack of reform in primary health-care in Canada (Hutchison et al. 2001). It has also been applied in examining health system reforms in Europe (Oliver & Mossia-los 2005), rehabilitation services for children with disabilities (Wiart et al. 2010) and children's mental health policy (Waddell et al. 2005). In a similar manner, it is used here to group the sociopolitical barriers that have influenced the capacity of CBPCTs to provide healthcare services in a shared care model. In doing so, it will inform the prospects for the development of additional teams in the LHIN of concern and elsewhere. Institutions are collegially referred to as 'the rules of the game' and here include financing and funding mechanisms, and healthcare system policy legacies (Hutchison et al. 2001) or structural inheritances (Williams et al. 2010). Interests are understood here as the perceptions, viewpoints and motivations of stakeholders. Finally, ideas are categorised to include the values and beliefs of stakeholders, in addition to information and research evidence. The analysis strategy involved reading through each of the transcripts thoroughly to become familiar with the data. Thematic coding was performed during subsequent readings (Cope 2010); it was predetermined that thematic coding would involve categorising text as an institution-, interest-or idea-related barrier.

Findings

Common institution-, interest- and idea-related barriers were apparent across all five CBPCTs and all six key-informants. The most frequently cited barriers related to each of the 3-I's, and excerpts that best illustrate these barriers are presented. Below, excerpts from key-informant interviews are denoted 'KI'. Excerpts from CBPCT members who participated in focus group discussions are denoted 'TM'.

Institution-related barriers

The two main types of institution-related barriers encountered by CBPCTs involve (i) funding mechanisms and (ii) the organisation of palliative care service delivery based on geography at provincial (macro) and LHIN-wide (micro) levels.

Funding

As previously mentioned, medically necessary hospital and physician services are covered as part of Canada's universal healthcare system. This constrains CBPCTs, as their services are being provided outside of the hospital and not all team members are physicians. Essentially, the way the healthcare system funds providers is not designed to support interprofessional collaborations such as shared care. Thus, teams must often resort to funding their non-physician team members through a combination of sources, which are often piecemeal and not secure. The following exchange between team members captures the issue of vulnerable funding:

Our roles not being full-time is a barrier, not having an administrative assistant is a barrier... the fact that we're paid by the hospital and if the hospital with budget cuts sometimes sees us a bit of a frill. (TM19)

So I guess a barrier would be a sense of security because every year the annual budget comes up, there is always a threat. (TM18)

We're always a bit worried. Are they going to cut our programme? (TM19)

According to a key-informant, part of the delay is lack of funding for the non-physician roles, including nurses, psychosocial spiritual and bereavement counsellors. Despite being a critical component of the team, the psychosocial spiritual and bereavement advisor positions have often fallen by the wayside because of lack of funding. Team members have deemed the number of supported hours for bereavement as insufficient. Key-informants acknowledged that failure to provide bereavement support services could, in turn, be more costly in the

long term as families are left to cope with unresolved issues, possibly leading to depression and anxiety.

Geographical organisation of service delivery As mentioned, not all LHIN areas in the province have adopted CBPCTs and the shared care model specifically. As one team member explained, differences in service delivery models across LHINs is sometimes problematic, especially for patients who move from neighbouring LHINs and expect homogeneity. There is a lack of consistency across the province, but also within the LHIN itself, as the areas that do not currently have teams are receiving usual care. As a result, primary healthcare providers are only able to access the CBPCTs depending on whether the patient is located within one of the five serviced geographical areas. One team member spoke about the challenge of getting requests for consultations from a neighbouring town that is currently receiving usual care:

[We] don't have [the] budget to have [the social worker] drive an hour and a half down [there] to see a family... Have we gone down there? Sometimes yeah, for the really tough, tough cases we have helped out; but honestly, until we get more funding just for paying mileage. it makes it hard. (TM3)

While it is common for patients to cross care settings, not all of the teams are able to go into hospices, long-term care homes, retirement homes and hospitals. Some team members are contractually restricted from practicing in certain settings. Depending on their employer, some team members are prohibited from entering settings other than private residences. A key-informant was concerned that this places a limitation on shared care:

. you have a family doctor who has a family practice and also works in long-term care and they can get that support for their home-based clinic but they can't get that support for their residents in long-term care and I think it sends a really bad message about sort of the whole philosophy of it. I think it speaks to the credibility of the programme, of the philosophy. when we say I can't help you with Mr. Jones but I can help you with Mrs. Smith. (KI5)

The inability to cross settings was also perceived as a barrier to the continuity of care and in fulfilling the goal of providing care to patients, regardless of where they are located.

In sum, the institution-related barriers illustrate that, despite the fact that palliative care teams are community-based, they are still very much part of, and greatly influenced by, the larger healthcare system.

Interest-related barriers

The next category of barriers is focused on the interests of stakeholders. More specifically, these barriers stem from

family physicians and community nurses that are invited to work alongside CBPCTs and to participate in shared care. Collaboration with these primary healthcare providers is imperative to the functioning of the shared care model; however, the lack of 'buy-in' from these providers owing to interests specific to autonomy and time was a commonly cited barrier.

Autonomy and time

Key-informants speculated that primary healthcare providers are concerned with maintaining their autonomy and pointed to the lack of 'buy-in' being attributed to threats to their independence:

[Some] feel either threatened because they feel that if somebody comes and sees what they're not doing, that's a threat to them professionally. others it's: 'I already know enough, you don't need [to show me] and: I just don't have enough time for you'. (KI5)

The preservation of time was also viewed as one of the interests held by primary healthcare providers. According to key-informants, primary healthcare providers, many of whom are already overworked, may be concerned that participating in shared care will involve additional time.

Participants also connected the hesitation of family physicians to work with CBPCTs, and thereby build their knowledge in palliative care, as a barrier related to their busy work schedules. Because of their hectic practices, family physicians may prefer for the team to 'takeover' care as a means of saving time on both providing care through home visits and capacity building.

Is that physician prepared to increase their skill level? And the reality is, for many, they're quite happy to have to not go beyond their office to have those decisions made for them because they are really very, very time consuming. (TMj2)

The lack of 'buy-in' from family physicians disables the CBPCT to work in a shared care model, often placing the experts on the team in a difficult situation in that there is no family physician to share the care with. The main difference between family physicians and community nurses seems to be the degree of autonomy. Family physicians have more autonomy in decision-making, whereas community nurses must abide by the rules and regulations stipulated by their employer. The participants noted that community nurses are diverse in terms of the extent of their involvement with teams. However, the same interest-related barrier associated with lack of time that was apparent with family physicians often arises with community nurses:

Learning takes more time and they need to be supported to do that. So our nurses get paid per patient or per visit.

Again there's no incentive there to spend more time learning if you're not going to meet your quota or not get paid, so depending on how the nurses are paid can have an influence. (TM17)

Team members recognised that some community nurses were eager to build their palliative care skills and believed that others had not yet seen the value in working with CBPCTs. Participants raised the issue that establishing shared care not only takes time but it also requires a willingness of primary healthcare providers to engage in skill building and interprofessional practice, the latter which may require forfeiting one's sense of autonomy.

Thus, the interests of primary healthcare providers were perceived to influence whether or not they engage in shared care with CBPCTs. As illustrated above, participants explained that owing to varying reasons, the interests of primary healthcare providers sometimes do not align themselves with working with CBPCTs, and as a result, autonomy and time operate as barriers to forming shared care partnerships.

Idea-related barriers

Two recurrent idea-related barriers were identified: (i) a misperception of the shared care philosophy and (ii) the inability to provide supporting evidence of a CBPCT's success.

Misperceptions about shared care

Conflicting ideas about the meaning of shared care was perceived as a barrier by CBPCTs and corroborated by key-informants. There was a lack of understanding of the shared care philosophy among primary healthcare providers, and specifically family physicians, based on incongruent beliefs or values towards collaborative care. Collaboration among providers is an essential component of the shared care model, as explained by a keyinformant:

. responsibilities for the care is shared between partners. 'It's not my problem or your problem, it's you know, ours', that we allocate the work according to respective schools, experiences, comfort and availability so it's a flexible arrangement... so that the sum of the whole is greater than the sum of the parts. (KI2)

While some attributed the misunderstanding because of a lack of interprofessional training of family physicians, others explained that it might be more to do with the fact that family physicians are accustomed to working independently and therefore do not know how, and7or have the desire, to engage in a shared care partnership. Participants explained that rather than perceiving shared care as a way to augment care, primary

healthcare providers might misinterpret it as a criticism or interference in the care plan of patients:

Sometimes it's [that] people don't know what they don't know, so they think they do palliative care very well so they don't want to refer to shared care. (KI6)

Some participants speculated reasons for the hesitancy of providers to enter into a shared care model. For example, the inability or unwillingness to work with a team was attributed to a lack of interdisciplinary training in medical education. Many team members were optimistic that owing to ongoing enhancements in interprofessional education, recent medical graduates might be more willing to participate in shared care. On the other hand, some noted that while primary healthcare providers, such as family physicians and community nurses, agreed to enter into shared care collaboration, they often failed to uphold their responsibilities. For example, some community nurses failed to keep the CBPCTs informed about changes in patients' statuses, while some family physicians failed to prescribe medications and make home visits. One team member argued that family physician home visiting is an integral part of shared care:

.you can't manage patients on the phone, like. you can't just say, '[the] patient has more pain, let's go up on hydromor-phone'... (TM29)

Failure of primary healthcare providers to uphold their responsibilities undermined the shared care philosophy and forced CBPCTs to assume full responsibility of patients. This not only compromised team resources, and their ability to care for additional patients, but also forced the CBPCTs to 'takeover' care.

Regardless of primary care providers' participation, teams explained that they were often willing to work outside of the shared care model for the benefit of the patient and family:

Some of the [family physicians] want to be involved peripherally; they don't want to do the palliative part but they still want to be the family doctor . but some don't want any involvement and you try to meet the needs of that doctor but [we're] always making sure that we're not forgetting that the patient has needs. (TM15)

When teams 'takeover' care, the shared care model ceases to exist and opportunities for capacity building diminish. A key-informant cautioned that 'takeover' care is not sustainable; without capacity building, the limited resources of the CBPCT will eventually be exhausted.

Lack of information/evidence

Although team members recounted numerous incidents of how they helped to avert patients from seeking care at a hospital's emergency department, the inability to pro-

vide evidence of the team's impact was the second commonly cited idea-related (i.e. information7evidence) barrier. Team members and key-informants expressed frustration with the inability to evaluate performance, which they felt could validate their existence. While the majority of participants equated success with improving the quality of life of patients and families, as well as capacity and relationship building with primary healthcare providers, they acknowledged that these measures of success were incongruent with evidence valued by policy-makers and funding partners:

Good utilization of the team could decrease ER [emergency room] visits, which is a huge buzzword right now. I really think [CBPCTs] can decrease ER visits but I don't know how to measure that. (KI5)

On the other hand, several of the CBPCTs expressed concern over not being able to collect statistics because of a shortage of resources, including administrative support and time. Clinical work took precedence over collecting statistics. At the same time, a key-informant highlighted the fact that to obtain more resources for teams, it is important to provide evidence to funding partners.

To summarise, the idea-related barriers include conflicting beliefs about what it means to work in a shared care model and the inability to collect and make available evidence to gain support for CBPCTs.

Discussion

While it is acknowledged that those barriers related to the Canadian healthcare system will not apply to other countries, it is likely that other countries may face similar challenges in developing community-based palliative care services, and especially those efforts which involve shared care. This study used the 3-I framework to identify the barriers of providing palliative care in a shared care model based on the perspectives of members from five CBPCTs and key-informants. A limitation of this study is that it did not include the primary healthcare providers who work with CBPCTs in a shared care model; their experiences would be valuable to explore to enhance the comprehensiveness of this study. Specific barriers were classified according to whether they related to institutions, interests, or ideas. A closer examination of the findings highlights that the barriers encountered by CBPCTs are interconnected. More specifically, it is evident that institutions establish the foundation from which other barriers are informed. In fact, it is the 'rules and regulations' engrained in the healthcare system, such as funding mechanisms and the geographical organisation of healthcare service delivery which present core challenges for CBPCTs that pursue the shared care model. In an effort to overcome barriers, the CBPCTs involved in

the study have employed various negotiation strategies, some of which will be briefly discussed here.

Because of the funding mechanisms that privilege medically necessary physician and hospital services, funding for non-physician CBPCT members is not secure. In an effort to mitigate this institution-related barrier, several communities have sought alternate funding opportunities to support these roles. Some communities in the LHIN of concern are attempting to build and maintain teams by reaching out to community philanthropic groups, hospitals, Community Health Centres and Family Health Teams. This piecemeal tactic may, in turn, place limits on the type of care settings in which team members can provide services. Connected to this issue of funding was the idea-related barrier concerned with the inability to produce evidence to demonstrate the benefits of CBPCTs as a means to secure funding. Although it is recognised that evaluating the outcomes of palliative care services can prove to be difficult (e.g. McWhinney et al. 1994), a number of the CBPCTs involved in the study were collecting information for the purpose of collating statistics. Yet, teams that lacked administrative support because of a shortage of funding felt limited in their ability to balance clinical and administrative demands.

The manner in which primary healthcare providers are funded also informs the interest-related barriers of autonomy and time. The majority of family physicians in Ontario work independently, or as part of small practices, and are reimbursed on a fee-for-service basis by the ministries of health in each province/territory (Hutchison et al. 2001), whereas community nurses are paid an hourly wage or per visit. CBPCT members and key-informants speculated that the lack of 'buy-in' for the shared care model from primary healthcare providers was as a result of an infringement on professional autonomy and time. The unwillingness of some family physicians to enter into shared care partnerships with CBPCTs may include preference for independent decision-making, as well as the inability or unwillingness to allot time to both working alongside a CBPCT and engaging in capacity building. In recognising this interest-related barrier, some CBPCTs have approached nursing agencies and family physicians on a one-on-one basis to demonstrate via meetings and presentations the advantages of working in a shared care model. Several of the CBPCTs have been successful in inviting community nurses to participate in weekly and bi-weekly rounds to build their palliative care skills through this approach. The teams perceived these types of outreach activities as a means of helping primary healthcare providers to gain a better understanding of the shared care philosophy, which is hinged on capacity building and collaboration. At the same time, this also enables

CBPCTs to address the idea-related barrier regarding the misperception of the shared care philosophy.

While various shared care models have been successful in other specialties, including chronic disease management (Vrijhoef et al. 2002), mental health (Hobbs et al. 2004, Chinna Samy et al. 2007) and obstetrical care (Lombardo & Golding 2003), there is little evidence that supports success in palliative care, particularly in the Canadian context. The lack of success may be attributed to the fact that unlike other specialties employing the shared care model, community-based palliative care involves attending to patients in care settings outside of hospital and physician offices, often in private residences. As noted by participants, the unwillingness of family physicians to commit to making home visits was considered an interest-related barrier to establishing a shared care model. As a way to move forward, the adoption of Family Health Teams in the province of Ontario (Hutchison 2008) may produce more primary healthcare providers that are willing to work with CBPCTs, as these providers are not solely paid on a fee-for-service basis and it would be an extension of the collaborative model in which they already work.

In spite of the constraints related to institutions, interests and ideas, prospects are still promising for CBPCTs in Ontario. The decrease in hospital deaths, in addition to population ageing, may dictate the need for CBPCTs. Given the fact that provinces/territories have authority over the delivery of healthcare services, direction for the promotion and funding of CBPCTs in Ontario will have to come from the provincial ministry of health. Until then, without the political will to move CBPCTs forward, efforts to implement changes will be incremental (Hutchison et al. 2001, Tuohy 2004) in nature. A team member attributed the incremental nature of moving CBPCTs forward to the fact that shared care involves building relationships with primary care providers, which takes time. In the meantime, efforts to develop and sustain CBPCTs will have to rely on the ability of existing CBPCTs to advocate for change and to negotiate the barriers they face.

Conclusion

The qualitative analysis revealed that CBPCTs are constrained by the funding mechanisms for team members; the views of primary healthcare providers; and the inability to generate evidence of their effectiveness. These barriers are interconnected, impacting both the development and sustainability of teams. Yet, the need for CBPCTs in Ontario is apparent given both changes in demographics and trends specific to the location of death. The uptake of CBPCTs will be incremental, and they will be required to continue to negotiate barriers

and work among other types of palliative care service delivery models. Given that community-based care is becoming increasingly common around the world, other regions and countries may benefit from learning about the challenges encountered by the CBPCTs in this study as they attempt to develop community-based palliative care services through the implementation of shared care. This study is primarily concerned with the views of CBPCT members and key-informants and serves as a baseline to identify barriers encountered by teams. Further enquiry may involve learning more about the perspective of primary healthcare providers who enter into shared care partnerships in palliative care. There is also further need to learn more about how CBPCTs negotiate barriers to ultimately inform the development of this type of service delivery model elsewhere.

Acknowledgements

We would like to express our sincere gratitude to the participants for their contributions to the study. LD is funded by a doctoral research award from the Canadian Institutes of Health Research (CIHR). AW is supported by a CIHR OWHC/IGH Mid-Career Scientist Award in Women's Health. This work is supported by the CIHR ICE grant [HOA-80057] titled 'Timely Access and Seamless Transitions in Rural Pallia-tive/End-of-Life Care', which is funded through CIHR Institute of Cancer Research and CIHR Institute of Health Services and Policy Research.

Conflict of interests

The authors declare that there is no conflict of interests.

References

Abelson J., Tedford Gold S., Woodward C., O'Connor D. & Hutchison B. (2004) Managing under managed community care: the experiences of clients, providers and managers in Ontario's competitive home care sector. Health Policy 68, 359-372.

Aggarwal G., Glare P., Clarke S. & Chapuis P.H. (2006) Palliative and shared care concepts in patients with advanced colorectal cancer. ANZ Journal of Surgery 76, 175-180.

Baxter J. & Eyles J. (1997) Evaluating qualitative research in social geography: establishing ''rigour'' in interview analysis. Transactions of the Institute of British Geographers 22, 505-525.

Billings J.A. (2002) Vicissitudes of the clinician-patient relationship in end-of-life care: recognizing the role for teams. Journal of Palliative Medicine 5, 295-299.

Bruera E., Neumann C., Gagnon B. et al. (1999) Impact of a regional palliative care program on the patterns of terminal cancer care in the Edmonton region. Canadian Medical Association Journal 161, 290-293.

Bruera E., Neumann C.M., Gagnon B., Brenneis C., Quan H. & Hanson J. (2000) The impact of a regional palliative care program on the cost of palliative care delivery. Journal of Palliative Medicine 3, 181-186.

Brumley R., Enguidanos S., Jamison P. et al. (2007) Increased satisfaction with care and lower costs: results of a randomized trial of in-home palliative care. Journal of the American Geriatrics Society 55, 993-1000.

Burge F.I., Lawson B., Critchley P. & Maxwell D. (2005) Transitions in care during the end of life: changes experienced following enrolment in a comprehensive palliative care program. BMC Palliative Care 4, 3.

Chinna Samy D., Hall P., Rounsevell J. & Carr R. (2007) 'Shared care - shared dream': model of shared care in rural Australia between mental health services and general practitioners. Australian Journal of Rural Health 15, 35-40.

Cope M. (2010) Coding qualitative data. In: I. Hay (Ed.) Qualitative Research Methods in Human Geography. 3rd edn. pp. 281-294. Oxford University Press, Don Mills, Canada.

Costantini M., Higginson I.J., Orengo M.A., Garrone E., Henriquet F. & Bruzzi P. (2003) Effect of a palliative home care team on hospital admissions among patients with advanced cancer. Palliative Medicine 17, 315-321.

Crawford G.B. & Price S.D. (2003) Team working: palliative care as a model of interdisciplinary practice. The Medical Journal of Australia 179, S32-S34.

Dudgeon D., Vaitonis V., Seow H., King S., Angus H. & Sawka C. (2007) Ontario, Canada: using networks to integrate palliative care province-wide. Journal of Pain and Symptom Management 33, 640-644.

Fainsinger R.L., Brenneis C. & Fassbender K. (2007) Edmonton, Canada: a regional model of palliative care development. Journal of Pain and Symptom Management 33, 634-639.

Ferris F., Balfour H., Bowen K. et al. (2002) A Model to Guide Hospice Palliative Care: Based on National Principles and Norms of Practice. Canadian Hospice Palliative Care Association, Ottawa.

Garrett G. & Weingast B.R. (1993) Idea, interests and institutions: constructing the European community's internal market. In: J. Goldstein & R.O. Keohane (Eds.) Ideas and Foreign Policy: Beliefs, and Political Change, pp. 173-206 Cornell University Press, Ithaca, NY.

Hall P.A. (1993) Policy paradigms, social learning, and the state: the case of economic policymaking in Britain. Comparative Politics 25, 275-296.

Hearn J. & Higginson I.J. (1998) Do specialist palliative care teams improve outcomes for cancer patients? A systematic literature review. Palliative Medicine 12, 317-332.

Herrera E., Rocafort J., De Lima L., Bruera E., Garcia-Pena F. & Fernandez-Vara G. (2007) Regional palliative care program in Extremadura: an effective public health care model in a sparsely populated region. Journal of Pain and Symptom Management 33, 591-598.

Higginson I.J., Finlay I.G., Goodwin D.M. et al. (2003) Is there evidence that palliative care teams alter end-of-life experiences of patients and their caregivers? Journal of Pain and Symptom Management 25, 150-168.

Hobbs H., Hamilton Wilson J. & Archie S. (2004) Evaluation of the Alumni Program: a shared-care model for psychosis. Journal of Psychosocial Nursing & Mental Health Services 42, 28-36.

Howell D., Marshall D., Brazil K. et al. (2011) A shared care model pilot for palliative home care in a rural area: impact on symptoms, distress, and place of death. Journal of Pain and Symptom Management 42, 60-75.

Hutchison B. (2008) A long time coming: primary healthcare renewal in Canada. Healthcare Papers 8, 10-24.

Hutchison B., Abelson J. & Lavis J. (2001) Primary care in Canada: so much innovation, so little change. Health Affairs 20, 116-131.

Lawson B.J., Burge F.I., McIntyre P., Field S. & Maxwell D.

(2009) Can the introduction of an integrated service model to an existing comprehensive palliative care service impact emergency department visits among enrolled patients? Journal of Palliative Medicine 12, 245-252.

Lombardo M. & Golding G. (2003) Shared antenatal care: a regional perspective. Australian Family Physician 32, 133139.

Marchildon G.P. (2005) Health Systems in Transition: Canada, WHO Regional Office for Europe on behalf of the European Observatory on Health Systems and Policies, Copenhagen, Denmark.

Marshall D., Howell D., Brazil K., Howard M. & Taniguchi A. (2008) Enhancing family physician capacity to deliver quality palliative home care: an end-of-life, shared-care model. Canadian Family Physician 54, 1703.e1701-1703.e1707.

Mazowita G. (2004) Shared care: the core concept of primary care renewal. Canadian Psychiatric Association Bulletin 36, 16-18.

McWhinney I.R., Bass M.J. & Donner A. (1994) Evaluation of a palliative care services: problems and pitfalls. British Medical Journal 309, 1340-1342.

Meier D.E. & Beresford L. (2008) The palliative care team. Journal of Palliative Medicine 11, 677-681.

Moorhead R. (1995) Sharing care between allied health professionals and general practitioners. Australian Family Physician 24, 1985.

Neergaard M.A., Olesen F., Jensen A.B. & Sondergaard J. (2008) Palliative care for cancer patients in a primary health care setting: bereaved relatives' experience, a qualitative group interview study. BMC Palliative Care 7, 1.

Neergaard M.A., Olesen F., Jensen A.B. & Sondergaard J.

(2010) Shared care in basic level palliative home care: organizational and interpersonal challenges. Journal of Palliative Medicine 13, 1071-1077.

OACCAC (Ontario Association of Community Care Access Centres) (n.d.) Ontario Association of Community Care Access Centres [WWW document]. URL http:// www.ccac-ont.ca (accessed on 31 May 2011).

O'Connor M. & Fisher C. (2011) Exploring the dynamics of interdisciplinary palliative care teams in providing psychosocial care: ''Everybody thinks that everybody can do it and they can't''. Journal of Palliative Medicine 14, 1-6.

Oliver A. & Mossialos E. (2005) European health systems reforms: looking backward to see forward? Journal of Health Politics, Policy and Law 30, 7-28.

Patton M. (2002) Qualitative Research and Evaluation Methods. Sage, Thousand Oaks, CA.

Saunders C. (2000) The evolution of palliative care. Patient Education and Counseling 41, 7-13.

Seow H., King S. & Vaitonis V. (2008) The impact of Ontario's end-of-life care strategy on end-of-life care in the community. Healthcare Quarterly 11, 56-62.

Statistics Canada (2010) Deaths: 2007. [WWW document]. URL http://www.statcan.gc.ca/pub/84f0211x/84f0211x 2007000-eng.pdf (accessed on 31 May 2011).

Tuohy C.H. (2004) Health care reform strategies in cross-national context: implications for primary care in Ontario. In: S. Shortt, R. Wilson & J. Dorland (Eds.) Implementing Primary Care Reform: Barriers and Facilitators. pp. 73-96 McGill-Queen's University Press, Kingston, UK.

Vrijhoef H., Diederiks J., Spreeuwenberg C. et al. (2002) The nurse specialist as main care-provider for patients with type 2 diabetes in a primary care setting: effects on patient outcomes. International Journal of Nursing Studies 39,441-451.

Waddell C., Lavis J.N., Abelson J. et al. (2005) Research use in children's mental health policy in Canada: maintaining vigilance amid ambiguity. Social Science & Medicine 61, 1649-1657.

Wiart L., Church J., Darrah J. et al. (2010) Cross-ministerial collaboration related to paediatric rehabilitation for children with disabilities and their families in one Canadian province. Health and Social Care in the Community 18,378-388.

Williams A.M., Crooks V.A., Whitfield K. et al. (2010) Tracking the evolution of hospice palliative care in Canada: a comparative case study analysis of seven provinces. BMC Health Services 10, 147.

Wilson D.M., Truman C.D., Thomas R. et al. (2009) The rapidly changing location of death in Canada, 1994-2004. Social Science & Medicine 68, 1752-1758.

Witmer E. (2000) Bringing healthcare closer to home: one province's approach to home care. Healthcare Papers 1, 98-102.

Woodward C.A., Abelson J., Tedford S. & Hutchison B. (2004) What is important to continuity in home care? Perspectives of key stakeholders. Social Science & Medicine, 58, 177-192.

Yin R.K. (2009) Case Study Research: Design and Methods, Sage Publications, Los Angeles, CA.

Zimmermann C., Riechelmann R., Krzyzanowska M., Rodin G. & Tannock I. (2008) Effectiveness of specialized palliative care: a systematic review. The Journal of the American Medical Association 299, 1698-1709.