«Citation: Dalkin, Sonia (2014) The Realist Evaluation of a Palliative Integrated Care Pathway in Primary Care: What Works, For Whom and in What ...»
GP2 (FG2): “Definition of what is palliative care, I think that lots of us, (laughs) are not sure when people enter palliative care, some people think it’s the last 3 days of life and some people think it’s the last 3 years of life. I think what we mean by it is a constraint to some practices.” Community Matron (FG1): “It is a new term, it’s a new term and we’re all used to, you know, the palliative being end-of-life and now it’s separated and it is confusing.” Palliative care is described by the community matron as a new term that causes confusion, this is unexpected given that the recent political agendas focus on improving palliative care services (Department of Health 2008, The National Gold Standards Framework Centre 2009, NHS North East 2012). In order to appropriately register patients, health care professionals need to understand the definitions of curative care, palliative care and end-of-life care. However, as the community matron highlights, this is difficult. In the locality, generally health care professionals refer to palliative care patients as those with palliative care needs (holistic, supportive care for those with a terminal diagnosis), and end-of-life care is referred to as the last 3 days of life when the LCP is implemented. The differences between these types of care (curative, palliative and end-of-life) can seem very distinct and therefore the cessation of curative treatment is often viewed as a discrete event.
GP1 (FG1): “I think it’s very difficult (definitions of palliative care) and it’s something we’ve endlessly discussed in our practice, and the nomenclature is very confusing, and I think cancer treatment makes it even more confusing, for example breast cancer, is often palliative but it doesn’t mean that, they may live a long life, it’s like a chronic illness isn’t it, the definition of palliative has to start to change, it’s very difficult, you know, what do you call people? I use the palliative care register and then people get confused that I mean the very end of life, the Liver pool pathway, it is very confusing.” Despite health care professionals’ discussions about the difficulties they had encountered with nomenclature, consensus in FG2 was that curative care precedes palliative care which precedes end-of-life care, as demonstrated by the quotes below.
GP1 (FG1): “you may be able to surmise that you should turn this palliative because they’ve got dementia etc, but they might not be the right person for full end-of-life care and all that, who needs hospital avoidance and all of that.” GP3 (FG1): “Once you have that label of palliative care, we’re meaning would you be surprised if this patient was to die in the next 6 to 12 months, yes and therefore let’s put them on the palliative care register, to hospital teams that might be like right ok, very different meaning.” GP1 (FG2): “Because yes they’re palliative, but they might still want specific active treatment, and to understand and give active treatment and then discharge.” GP1: “Just to tell you the, mix, the kind of, erm, confusion out there about what a palliative care register is, I put him on the palliative care register and there was merry hell on with the learning disability team. Merry hell. It was hilarious actually. It was a very good learning experience actually. They all thought I had said he was about to die and that’s the degree of confusion about what a palliative care register is.” The traffic light system can be used to facilitate health care professionals’ understanding of curative, palliative and end-of-life care. Green can relate to when a curative framework is in operation, but palliative care should begin to be considered. Amber relates to a decrease in functioning, suggesting illness progression; a palliative care framework should have been adopted by this point.
Red refers to end-of-life, where the LCP would have been implemented (pre-phase out).
Community matron (FG1): “Are they green amber or red? Where do they sit?” GP1 (FG1): “Yeah that’s (traffic light system) quite well defined now and I think it’s nice that it is so well defined. And I think that will encourage people to use the palliative care register better because I do think palliative gets connected with death and dying and it’s not and we need the palliative care register to be bigger so we can then hone down and pick out the end-of-life care pathway patients.” Once health care professionals in the locality had formed an informal consensus about these definitions and had experienced using the traffic light system they could then use these concepts in practice to place patients on the palliative care register.
The palliative care register can be considered as a tool to keep a track of patients with palliative care needs. FG3 discussions highlighted that health care professionals had become sensitised to the appropriate definitions over time, from 2008 to 2012.
GP4 (FG3): “I presume something’s changed between 2008 and 2012 and I think presumably the health care professional knowledge of the patients and family hasn’t changed very much in relative terms, so I guess obviously the end of life strategy and the agreed consensus of palliative care terms are where awareness has been heightened over the past few years. And I guess that’s facilitated the registration of palliative care patients on to the list and erm, you know use of the pathway I suppose.” The GP in FG3 explains how health care professionals’ understanding and awareness of palliative care has increased since the implementation of the ICP.
Figure 14: CMOC1, palliative care registrations In summary, publication of essential policy in the form of the End of Life Care Strategy (Department of Health 2008) sensitised practices to the need to improve palliative and end-of-life care which the locality did through the development and use of palliative care registers. In order to do this they needed to become familiar with and form consensual definitions of active, palliative and end-of-life care (context). Despite on-going difficulties, discussions identified that this had been mostly achieved. The surprise question (individual resource) has increased health care professionals confidence (individual reasoning) to suggest the patient for palliative care registration at MDT meetings (team resource) which further increases confidence of the health care professionals to operate within a palliative care framework where appropriate (reasoning). These two mechanisms (individual and team) have resulted in an increase in palliative care registrations from 2008 (ICP implementation) to 2012. This CMOC is displayed diagrammatically in Figure 14.
It is also important to note here the importance of MDT meetings. High performing practices identified in Chapter 4 (Embeddedness of the ICP as a new initiative, p.117) had more frequent MDT meetings, which were facilitated by having the shared nursing team on site. MDT meetings also have a key role in this CMOC, resulting in more palliative care registrations.
CMOC2 – Registering non-cancer patients Outcome: Less non-cancer patients than cancer patients registered Analyses were conducted to investigate whether there were fewer non-cancer registrations than cancer registrations using the most recent data available at the time, from 2011. The 2012 data was made available at a later date allowing extra comparative analyses to be conducted. All data includes sudden deaths, as removing these cases would result in biases; the definition of a sudden death is variably interpreted in some practices in the locality, one practice had recorded all noncancer deaths as sudden. Removing these people would result in a greater bias than leaving them in the analyses.
Histograms were created in order to visualise the number of cancer and non-cancer patients registered in each practice (Figures, 15, 16 and 17).
Figure 15: The number of cancer deaths in total compared to the number of cancer deaths that were on the palliative care register, for 2011, using Death Audit data.
Figure 15 indicates that a majority of cancer deaths in each practice are on the palliative care register, although there are wide variations. For example, four of the practices managed to register all of the cancer deaths in 2011, whereas practice N registered very few (the differences between practices was focused upon in more detail in the previous chapter on embeddedness, p.117).
Figure 16: The number of non-cancer deaths in total compared to the number of non-cancer deaths that were on the palliative care register, for 2011, using Death Audit data.
Figure 16 shows that in every practice there are a substantial number of non-cancer deaths that occur that are not on the palliative care register. In comparison to Figure 15 a visual discrepancy becomes apparent; there are more cancer deaths registered than non-cancer deaths registered, even in practices that clearly use the palliative care register extensively for cancer patients.
Figure 17: Ratios of palliative care registered to not palliative care registered deaths for those with cancer and non-cancer for each practice in 2011, using Death Audit data.
The use of ratios in Figure 17 allows for comparisons of cancer and non-cancer palliative care registrations irrespective of practice population size. Using ratios, a score of 1 would mean that all of the cancer or non-cancer patients who died in 2011 were on the register. A score of 0.5 would mean half of the patients were registered. Figure 16 shows that a lot more cancer patients are put on to the palliative care register than non-cancer patients, in all practices. However, there is significant variability, some practices do have better outcomes than others, for example, Practice A registered no non-cancer patients, whereas practice D registered around half of all non-cancer deaths.
A Pearson product moment correlation co-efficient showed that there was a weak non-significant correlation between cancer and non-cancer palliative care registrations (r = 0.26, n = 14, p.05), suggesting that in each practice the number of non-cancer registrations does not increase as the number of cancer registrations does. This finding is unexpected as cancer deaths only account for around 30% of overall deaths (Office for National Statistics 2013).
A paired samples t-test was conducted to investigate this further and evaluate the impact of the ICP on cancer and non-cancer registrations in 2011, using the ratio of cancer deaths on the palliative care register to total cancer deaths and the ratio of non-cancer deaths on the palliative care register to total non-cancer deaths. A significant difference between cancer and non-cancer registrations was identified (t(13) = 8.78, p.001). There were significantly more cancer palliative care registrations (M = 0.78, SD = 0.06) than non-cancer palliative care registrations (M = 0.25, SD = 0.16) in 2011 across all fourteen practices. This highlights that there are significantly more cancer patients than non-cancer patients being put onto the palliative care register.
Mechanism: Stress when registering a non-cancer patient FG3 highlighted that health care professionals find registering non-cancer patients difficult. This is due to the non-predictable trajectory of non-cancer illnesses, the difficulty in considering non-cancer patients as appropriate for palliative care and the lack of health care professionals’ involvement in treating progressive symptoms.
GP4 (FG3): I think, erm the other things as well, historically with cancer patients rather than with non-cancer patient’s we’ve been more actively involved in proactively managing symptoms as they’re reaching end-of-life. Whereas in a lot of the non-cancer deaths we might have been possibly, for example, more frequently treating infections but often it’s a gradual slide with a lot more intervention on our part necessarily, and I think the same goes for the nurses as well.” GP5 (FG3): “You’re acknowledging that they’re declining but you’re not necessarily actively doing a lot more for the symptomology… You don’t necessarily understand the pain that they’re feeling like you would with bone mets in the spine (in cancer).” GPs 4 and 5 explained how they felt that this lack of experience in treating the symptoms of non-cancer patients as they approach end-of-life, and the context of non-cancer patients’ unpredictable illness trajectory led to stress when deciding whether to put non-cancer patients on the palliative care register.
GP5 (FG3): Err, so how do you make choices? So I make choices and then they come off the board.” GP4 (FG3): “There is stress with non-cancer registrations. With a cancer diagnosis you have a fixed underlying illness I think haven’t you? That the family are aware of and recognise that their relatives generally going downhill. But with a non-cancer diagnosis you know they might have had heart failure for years and years and years and yes they’re getting a bit worse but they’ve had this diagnosis for years whereas with cancer, quite often, there’s the downhill trend.
A stressful decision like this would mean that health care professionals’ confidence about the decision is low. Low confidence in a decision would mean that they are less likely to register a non-cancer patient.
GP5 (FG3): “But I still find it difficult in my head because there’s a blip, you’re just not as confident in putting someone on the board when they’re non cancer than you are if they’re cancer.