«Citation: Dalkin, Sonia (2014) The Realist Evaluation of a Palliative Integrated Care Pathway in Primary Care: What Works, For Whom and in What ...»
You’re just not confident about where the end point is going to happen, how it’s going to happen, what it’s going to be like, so there is kind of a confidence thing about it.” Self-efficacy is the perceived confidence to carry out desired tasks (Bandura 1977) and can be influenced by experience, persuasion, or physiological and psychological states such as distress. High self-efficacy correlates with low perceived stress in professional caregivers (Hulbert and Morrison 2006), thus suggesting that when health care professionals have more confidence (self-efficacy) then they have less stress about the decision they have to make. The results indicated that health care professionals were confident in registering cancer patients but less confident in registering non-cancer patients. This suggests they have a lower self-efficacy (reasoning) when registering non-cancer patients with unpredictable prognoses (context), giving them more perceived stress about the decision (reasoning) which results in less non-cancer registrations. Furthermore, effective co-worker communication and work place peer support have been identified as important variables for occupational stress management (Alexander and Ritchie 1990, Searle, Bright et al. 2001, Bradley and Cartwright 2002). The three practices which were identified as high performers (in Chapter 4: Embeddedness of the ICP as a new initiative, p.117) demonstrated strong peer support for palliative care.
In addition to low self-efficacy in dealing with non-cancer patients, the traffic light system (resource) is not as useful when dealing with non-cancer trajectories. Noncancer patients often move between traffic light stages non-linearly meaning that if a patient is in ‘green’ they could very quickly deteriorate to ‘red’ and potentially be put on to the LCP, but their condition could improve and they can become ‘green’ again. This is unlikely to happen in cancer diagnoses. Thus, the traffic light system (resource) and the increased self-efficacy it gives health care professionals (reasoning) to register palliative care patients cannot be used as efficiently in the context of the unpredictable trajectory of non-cancer diagnoses.
To summarise this mechanism, health care professionals have little experience of treating the symptoms of non-cancer palliative illnesses and thus have low selfefficacy (reasoning) in registering this patient group. The traffic light system, that is one of the resources used to help register cancer patients, is not as useful in the context of unpredictable trajectories of those patients with non-cancer illnesses.
Context: Unpredictable illness trajectories As discussed in the opening of this chapter, non-cancer diagnoses typically have an unpredictable trajectory. This makes decisions about when palliative care is appropriate particularly difficult and may help to explain why health care professionals’ self-efficacy in treating this patient group is low. Therefore registering a non-cancer patient is challenging for health care professionals, as a period of significant decline can be followed by substantial improvement, despite a downward trend in wellness (Murtagh, Preston et al. 2004, Murray, Boyd et al.
2005). A GP in FG3 explained this.
GP5 (FG3): “I think the, the definition of palliative care is still difficult, not so much with cancer deaths, but still with non-cancer.
I still put people on the board and then take them off it, they’ve rallied. I’m like “err, he was dying last week?” and now he’s rallied, you know (inaudible).” The same GP went on to elaborate on the difficulty of the unpredictable prognosis in older people with non-cancer. He felt that all older people are at risk of entering end-of-life care rapidly and this is an unpredictable transition from palliative care.
The care home population frequently have non-cancer illnesses which can exacerbate quickly and unpredictably. Health care professionals cannot predict this but also do not wish to over populate their palliative care registers by registering all non-cancer elderly patients. Once a decline in health begins in elderly people with non-cancer illnesses, it can be very rapid and thus end-of-life care is implemented.
The GP felt that this left no time for putting people on to the palliative care register.
Comparatively, this is not the case with cancer diagnoses as often there is a specific diagnosis and steady illness trajectory.
GP5 (FG3): “I think the confidence comes from the potential for flood gates, really, if you say well basically everyone in a care home, I wouldn’t be surprised if they all had an acute illness and died from whatever frailty the suffer, COPD, dementia, whatever, you feel that you could then just open the flood gates to say that everyone could be on the palliative care register. You know, they’re all in their 80s and 90s, they’re all getting to the end of their life, and so it’s very difficult to kind of be selective about them, and the illness that they get that ends their life is only a matter of weeks which isn’t a lot of time to engage palliative care in your brain and go to the board and write it down, you’re chasing your tail really. At least in cancer you’ve got this progression and diagnosis and it’ll end in one way, but with this you’re chasing your tail, it’s just the timing and the... I could put basically everyone in care homes on that list.
The inequity between cancer and non-cancer palliative care registrations may also be exacerbated due to the emergence of the palliative care register from the cancer register. In the past, practices were required to identify and register people with cancer and then undertake an annual review. Following this, a requirement to have a palliative care register emerged. However, these registers were under populated and those on the list were often cancer patients (identified from the aforementioned cancer register). Furthermore, typically no action was taken following registration.
This history of palliative care registers being populated almost solely by cancer patients exacerbates the notion that palliative care only relates to terminal cancer illnesses and thus may cause health care professionals to refrain from registering non-cancer patients. The evidence highlights that a health care professional who has a palliative care definition inclusive of non-cancer will be more likely to register such a palliative care patient.
Figure 18: CMOC2, registering non-cancer patients The unpredictable trajectory of non-cancer illnesses and difficulties in considering non-cancer patients as appropriate for palliative care (context), mean that health care professionals have less experience of treating palliative and end-of-life care symptoms in those with non-cancer (in comparison to cancer), meaning that health care professionals are less confident in registering non cancer patients (reasoning).
This results in significantly less non-cancer patients than cancer patients being registered in 2011. This CMOC is displayed in Figure 18.
CMOC3 – Care home residents Outcome: Less non-cancer patients than cancer patients registered As described in CMOC2, in 2011 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) across all fourteen practices.
Mechanism: Anxiety about registering care home residents In FG3 health care professionals discussed the difficulties in the decision (reasoning) to put elderly patients who are in care homes and have a non-cancer diagnosis on the palliative care register (resource), due to their rapid declines and unexpected recoveries (context). This led to concerns regarding the over population of the palliative care register, as described by GP5.
GP5 (FG3): “There is the potential for flood gates, really, if you say well basically everyone in a care home, I wouldn’t be surprised if they all had an acute illness and died from whatever frailty they suffer, COPD, dementia, whatever, you feel that you could then just open the flood gates to say that everyone could be on the palliative care register. You know, they’re all in their 80s and 90s, they’re all getting to the end of their life, and so it’s very difficult to kind of be selective about them, and the illness that they get that ends their life is only a matter of weeks which isn’t a lot of time to engage palliative care in your brain and go to the board and write it down, you’re chasing your tail really. At least in cancer you’ve got this progression and diagnosis and it’ll end in one way, but with this you’re chasing your tail, it’s just the timing and the... I could put basically everyone in care homes on that list.
Err, so how do you make choices? So I make choices and then they come off the board.” Although the ICP and surprise question (The National Gold Standards Framework Centre 2009) premise would suggest that all patients who are likely to die in the next 6 months should be registered, this can result in over population of the palliative care register. This can lead to a greater workload for health care professionals who already have stringent time constraints and laboured discussions about palliative care patients at MDT meetings. The social care team lead and a GP in FG3 described how it is unhelpful to over populate the register, especially with frail elderly people in care homes, as this results in those who need high care input not receiving it.
Social care team lead (FG3): “You want to be giving care to those who most need it” GP6 (FG3): “And that’s a burden administratively, the dangers are that we take our eye of the ball because we have so little time and focus on people who really need the medical input.” GP5 suggested that the palliative care register is inappropriate for those with noncancer palliative care needs and that another tool that is more flexible may be more useful. However, it seems that this would still result in the same issue of register over population.
GP5 (FG3): I think defining it, fear of flooding the board, and erm, perhaps, perhaps we should be using the palliative care register less as a black and white thing in our mind and more of a kind of flexible, ah here’s someone getting a bit worse, put them on the board. A palliative care, sort of declining, register (laughs). A bit softer, it kind of helps you to see it as some kind of, ‘going through a blip register’. You know, may not recover register, I dunno.
Something that’s a bit more flexible but palliative for non-cancer is really the wrong word because they go through this dip and you’re not quite sure if they’re going to recover. The definition doesn’t quite fit; it’s difficult to put them in that box. It’s difficult.
The GP referred to ‘palliative care’ as inappropriate for those with non-cancer.
Other health care professionals in the locality did not express the same belief but did agree that non-cancer patients are a difficult population to provide palliative care to.
Context: Elderly non-cancer patients’ prognosis is difficult to predict The literature highlighted that there is an ageing population in the UK (Forder and Fernandez 2011, Gomes, Calanzani et al. 2011) and the level of frailty, impairment and need of people admitted to care homes is now higher than it was 10-15 years ago (Forder and Fernandez 2011). Non-cancer elderly patients’ health can decline rapidly, which makes palliative care registration difficult for health care professionals.
GP5 (FG3): They’re insidious aren’t they? The lethargy that comes with being ninety kind of, you just kind of accept the fact that someone starts using a Zimmer frame because of their arthritis” Accompanying this rapid decline can be an unexpected recovery.
GP5 (FG3): “They become Lazarus and rise from the dead and start having breakfast the next day.” The unpredictable prognosis of elderly patients in care homes with non-cancer diagnoses poses particular issues for effective health care management. Elderly people in care homes form a key context within which the decision to register is difficult to trigger (reasoning) even with ICP resources (surprise question, traffic light system, MDT team meetings).
Figure 19: CMOC3, care home residents.
The literature has indicated for some time that failure of health care professionals to consider patients as palliative may contribute to the inappropriate care of patients (Graham and Livesley 1983), thus it is essential that appropriate patients are put on to the palliative care register, regardless of diagnosis. However, elderly people in care homes constitute a key context, in which the decision to register is difficult. In 2011, fewer non-cancer patients who died were on the palliative care register (outcome). This could be due to health care professionals’ decisions not to register elderly care home residents with non-cancer diagnoses (reasoning) due to anxiety about flooding the register and not providing needs based care to patients requiring palliative or end-of-life care (Figure 19). Contributing to this is CMOC2, in that most care home patients have non-cancer illnesses which can be very difficult to prognosticate.
CMOC4 – Recent increase in non-cancer registrations Outcome: Significant increase in non-cancer registrations from 2011 to 2012 Although significantly less non-cancer patients than cancer patients were put on to the palliative care register in 2011, this does not indicate whether non-cancer palliative care registered deaths are increasing. To identify if practices using the ICP were increasing their registration of non-cancer patients 2011 data was compared to 2012 data.