«Citation: Dalkin, Sonia (2014) The Realist Evaluation of a Palliative Integrated Care Pathway in Primary Care: What Works, For Whom and in What ...»
Although the study identifies positive outcomes, it does not explain why they might have occurred. The aim of the intervention was to increase coordination between services, but this is highlighted as one of the challenges, thus this may not be the explanation for the outcomes observed. The resources used in this ICP need to be identified in order to understand why the health care professionals involved changed their behaviour, resulting in the outcome of improved access to palliative care services and increased home deaths.
Reymond, Israel et al. (2011) implemented the residential aged care end-of-life care pathway (RAC EoLCP). This care pathway involved the promotion of: ACP, proactive care, multidisciplinary management of residents (patients) and their family members, death in current residence (care home), optimal symptom management and increased competence in managing palliative care cases. Those working in the care home were also given palliative care training to enhance their care capacity. The implementation of the RAC EoLCP resulted in 98.3% of residents dying in their place of choice (the care home) and significantly less hospital admissions in comparison to those not on the RAC EoLCP. Staff perceived an improvement in their satisfaction with, and quality of, palliative and end-of-life care provided by the care home after the introduction of the RAC EoLCP. One staff member reported that, “before (RAC EoLCP implementation) staff had little training in palliative care and sometimes were very scared of caring for the dying” (Reymond, Israel et al. 2011). Bereaved relatives evaluations of palliative and end-of-life care did not increase from the pre-implementation to postimplementation phase; satisfaction remained consistently high. Thus, Reymond, Israel et al. (2011) highlight that training leads to increased use of ACP and LCP, however they do not explicitly identify the tools (although it could be assumed to be ACP and LCP) or thought processes of individuals implementing the new service. Furthermore, they do not refer to the context of this intervention. The same positive outcomes may not have been achieved if this were an intervention outside of a care home or if the recipients of the training had not seen the relative advantage of the new service.
However, outcomes of this study indicate that palliative care ICPs that have a focus on ACP and multidisciplinary working can facilitate death in the patients’ current place of residence.
A three year non-blinded randomised controlled trial took place to compare the effects of early palliative care integrated with a standard oncology care service in comparison to standard oncology care alone, for newly diagnosed metastatic non-small cell lung cancer patients (Temel, Greer et al. 2010). Patients assigned to the early palliative care integrated with standard oncology care service met with a member of the palliative care team within three weeks of enrolment and at least monthly thereafter in an outpatient setting till death. Additional visits were scheduled at the discretion of the patient.
Meetings focused on the assessment of physical and psychological symptoms, founding goals of care, helping with individual decision making regarding treatment and coordinating care. Patients in the palliative care integrated group had a higher quality of life in comparison to the control group (standard oncology care), as measured by the Functional Assessment of Cancer Therapy–Lung (FACT-L) scale. Of the patients who died, those in the standard oncology care group had more aggressive treatment. Despite receiving less aggressive treatment palliative care group patients survived longer.
Finally, more patients in the palliative care group had resuscitation preferences documented. Again, this palliative care ICP documents positive outcomes for those in the intervention group. However, the study does not identify what factors explain the choice of patients in the intervention group to have less aggressive treatment.
Bakitas, Lyons et al. (2009) conducted a RCT for nearly 5 years with patients who had advanced cancer. It is rare to find an RCT in this domain that is so long term; this is an asset to the study. Patients were randomly allocated to receive either standard care (n=161) or a multicomponent, psycho-educational intervention: Project ENABLE (Educate, Nurture, Advise, Before Life Ends) (n=161), implemented by advanced practice nurses. The intervention consisted of four weekly educational sessions and monthly follow-up sessions until death, which encouraged patient activation, selfmanagement, and empowerment. Patients who received the intervention had higher scores for quality of life and mood than those receiving usual care. However, they did not have improvements in symptom intensity scores or reduced days in hospital visits.
No details are provided as to why intervention group patients’ quality of life or mood improved apart from that they had educational sessions. The content of these sessions is described but no description of what elevated mood or quality of life specifically is provided.
Implementation of the Gold Standards Framework to enhance palliative care services Many studies describe implementation of the GSF; a systematic, evidence-based approach to aid health care professionals in identifying patients in the final years of life, assessing their needs, symptoms and preferences and planning care on that basis and empowering patients to live and die where they choose (The National Gold Standards Framework Centre 2009). It supports care pathways that are patient and carer centred (Thomas 2003, Pellett 2009). Hockley, Watson et al. (2010) implemented the Gold Standards Framework for Care Homes (GSFCH) and the LCP over eighteen months.
Significant in-house training was provided for staff with GSFCH facilitators visiting the care homes every ten to fourteen days. Residents’ (n=228) notes were reviewed and qualitative interviews with bereaved relatives were conducted pre and post implementation of the pathway (GSFCH and LCP). Use of DNACPR forms, ACP and LCP (which are all also used in the ICP under study) increased significantly and hospital deaths were reduced from 15% pre ICP implementation to 8% post implementation. Therefore, it can be concluded that the GSFCH was successful in improving outcomes for care home residents with palliative care needs who died.
However, similar to all of the studies described, this study does not explain why these outcomes occurred; it only provides the outcomes and tools used, and an assumption of causality between the two. Hall, Goddard et al. (2011) also implemented the GSFCH.
Nine care homes were involved and semi-structured interviews were conducted with all care home managers, eight nurses, nine care assistants, eleven residents and seven family members. Perceived benefits of the GSFCH were improved symptom control, team communication, staff confidence and adhering to patients’ preferences. However, there were some perceived barriers to the GSFCH as well. These included lack of understanding about end-of-life care and an increase in paperwork (Hall, Goddard et al.
2011). Several of the tools in the GSFCH focus on improving communication in palliative care (The National Gold Standards Framework Centre 2009). Hall, Goddard et al. (2011) found that the GSFCH improved communication within homes and with external providers including GPs and specialists in palliative care. Tools created to improve communication were similar to those used in the ICP under study and included palliative care registers, coding predicted stages of illness and ACP. These tools were perceived as beneficial in the study however some participants felt they required more experience of using these. There were also concerns about discussing death and dying (Hall, Goddard et al. 2011). Hall, Goddard et al. (2011) suggest that the use of the GSFCH tools have improved communication, but they do not explicitly state how this increase might have occurred – stronger team cohesion or improved multidisciplinary collaboration, for example. Furthermore, they highlight barriers to implementing the GSFCH but do not state explicitly if or how these barriers have affected the implementation or outcomes.
Bower, Roderick et al. (2010) evaluated the use of the GSF in two GP practices, one rural and one urban, with the aim of improving integrated team working to facilitate death in the patients’ place of choice. Despite the use of the GSF increasing health care professionals’ workload, through early identification of patients, they were able to work with patients prior to a crisis in the last few weeks of life. Additionally, community nurses, practice nurses and GPs all felt empowered to identify patients with any illness requiring palliative care who may have been in their last year of life, despite a previous focus on cancer patients. Increased and structured MDT meetings allowed patients to be monitored and individualised care plans to be altered to meet care needs. A lead GP was allocated to each patient to ensure continuity of care. Recent deaths were reflected on in MDT meetings to facilitate practical learning. Once identified at the MDT meeting, relevant information about palliative care patients (medication, next of kin, current input, potential risks) was faxed to OOH services in order to create an advanced service.
The patient’s notes (at home) provided information for all health care professionals to use and provided information for patients and their relatives about who to contact (GP or nurse) during the day and at night; this relieved feelings of anxiety that can occur due to isolation from services out of normal working hours. Health care professionals worked closely with patients to assess the need and timely implementation of anticipatory medication (a tool used in the ICP under study). An audit of twenty-one patients who died with community nurse involvement indicated that high quality MDT working with the GSF is effective in supporting patients to die in their place of choice (90% of patients died in their place of choice). However, limitations of the intervention were reported by health care professionals who stated that completing notes in the patient’s home and then duplicating these notes electronically at the practice was time consuming. There were also already strong working links between GPs and nurses;
specialist palliative care expertise in the community nursing team; and an ‘open to change’ ethos in both practices. This questions whether the same service would work as effectively in a practice where the context was not as favourable; changes in practice may have taken longer to implement and have been faced with more barriers. It must also be noted that this study had a small sample size (n = 21); larger scale implementation may have created issues. However, small-scale findings indicate that implementation of palliative care tools promoted by the GSF result in improved outcomes.
What is known so far?
Some overlaps can be seen between the palliative care ICP under study and those in the literature. They all aim to increase patient centeredness and have a focus on providing proactive care. Some use shared decision making in the form of ACP or educational sessions that promote decision making about the patient’s care. Decision aids and cause for concern trigger tools have similar properties to the surprise question and the traffic light system in the ICP under study. The GSF studies use many similar tools to the ICP under study including palliative care registers, the traffic light system, ACP and the LCP. Many of the studies have working collaboratively or enhanced communication as an aim in the implicit expectation that this will allow patients better access to palliative care services. MDT working was perceived to integrate team work and enhance patient centred care. However, it is highlighted that working collaboratively can be a challenge.
Confidence was highlighted as important in providing high quality palliative care and the perceived ability to provide palliative care may be important in ICPs. Health care professionals may also need to feel empowered in order to identify palliative care patients regardless of diagnosis (cancer or non-cancer). Many of the ICPs and GSF studies aimed to reduce hospital deaths and increase home deaths which they did successfully. Again, this highlights how health care professionals in these studies are operating within a patient centred framework. However, barriers to using palliative care ICPs were also identified in the literature including understanding about palliative and end-of-life care, increases in paper work and time constraints, and difficulties discussing death and dying.
It is acknowledged that the ICP under study is not always congruent with the ICPs reviewed in the literature. For example, the ICP under study does not use regular meetings with palliative care experts or educational sessions with patients, as Temel, Greer et al. (2010) and Bakitas, Lyons et al. (2009) did. However, consultations can be seen as meetings with experts (health care professionals) and thus some learning can still be taken from these studies.
However, in all of the studies reviewed the causality between the introduction of practice tools and improved outcomes is at present assumed rather than understood in depth. Studies evaluating palliative care ICPs state that an intervention that promotes early identification of patients in need of palliative care and the use of ACP, collaborative working and specialist palliative care services have positive effects in terms of quality of life, place of death and resuscitation orders. However, although the palliative care intervention studies described have shown improvements in care, they report these favourable outcomes but do not identify why they occur. All of the interventions described have one thing in common: they involve multiple tools and professionals. It is therefore very difficult, within the research designs employed, to decipher any detail on the causality chains which might have led to favourable outcomes. Furthermore, the studies discussed pay no attention to context such as location, population, disease type, socioeconomic factors, or understanding of palliative care. The interventions all show promising outcomes, but none go in to much depth to explain which components of the inherently complex and multifactorial interventions might have worked best for whom, and in what circumstances.