Twenty-seven GPs took part in the focus groups. Participants in three focus groups had used ReSPECT in their own clinical practice, while participants in the other two groups had not (see Table 1).
Table 1
Focus groups participants’ experience of ReSPECT
| Focus Group (n) | Is ReSPECT used in participants’ GP practices? | Number of months ReSPECT had been used by their local hospital at the time of the focus group |
| FG1 (n = 10) | Yes | 27 |
| FG2 (n = 3) | No | 23 |
| FG3 (n = 5) | No | 23 |
| FG4 (n = 4) | Yes | 27 |
| FG5 (n = 5) | Yes | 27 |
ReSPECT is an end of life care document.
There was an implicit assumption that ReSPECT was to be used to plan end of life care, and should be used for patients who were “coming to the last two or three years of their life” (FG5); palliative care patients, frail patients, or patients in the final stages of a chronic illness. Participants found it easier to initiate RESPECT conversations with advanced cancer patients (with clear illness trajectories) than patients who were frail or were living with a chronic health condition (such as COPD or heart failure) where the prognosis was less certain and the patient themselves had little awareness of their potential trajectory.
ReSPECT-type conversations were predominantly initiated by the GPs themselves, but sometimes by other health professionals, such as Macmillan nurses. The trigger for such conversations was typically a deterioration in the patient’s health. For palliative patients, this might mean a change in Gold Standard Framework (GSF) classification, from green to amber, or amber to red. For patients with dementia the trigger for a ReSPECT conversation would be the diagnosis itself. Participants felt it was important to identify and record the patient’s wishes while they had capacity. ReSPECT was also initiated as part of a routine care home admission.
Several participants had a “hunch” or they “just knew” that it was the right time to initiate a ReSPECT conversation. Their feeling was sometimes prompted by verbal cues (such as references to recent experiences in hospital) that could be used as an opening to the conversation. An existing relationship with the patient enabled them to know when the time was right.
Less common was the patient themselves initiating a ReSPECT-type conversation. The focus tended to be resuscitation, with the patients certain they did not want to be resuscitated in the event of an emergency. Patients who wanted to formalise a desire to decline CPR were often healthy, and our participants questioned the morality of doing a ReSPECT-type form for people who were not obviously approaching end of life: “some patients will want a DNACPR in place when there’s really nothing wrong with them […] they just don’t like the idea of going through resuscitation. Then, that's a whole other minefield […] am I really doing the right thing for this person when they could have a really positive outcome potentially with treatment?” (FG2)
ReSPECT is best done in primary care.
There was a general consensus that ReSPECT-type conversations could be done well within primary care. Conversations were often planned for home visits with the patient aware that they will discuss end of life care, thus ensuring that the patient was as prepared and comfortable as possible. Some GPs had lengthy pre-existing relationships with their patients, and this made the conversation easier for both the GP and the patient: “you do get to know them, you do get to know the family and build that relationship. So it does make sort of discussions like this much easier, 'cause you’ve built that relationship with them” (FG1).
ReSPECT-type discussions were described as an ongoing process which takes time to complete; they were “not a one off conversation” (FG3). This process involved judging when the time was right to start the discussion, holding initial conversations, and ensuring family members were present when decisions were made. As patients were not usually at immediate risk of deterioration, the availability of time was seen as real “advantage” the GPs had over acute care. Because of these perceived advantages afforded by primary care participants felt that they should be the ones holding ReSPECT conversations, rather than hospital staff.
GPs described time and resource constraints as barriers to ReSPECT-type conversations. If they identified a patient approaching end of life within their usual clinic, the GPs felt constrained by the 10 minute consultation slot and aware of a busy waiting room outside: “I think when you’re reaching them, they already come to you for something else. And they’re in there within five or ten minutes, you can’t really” (FG1). Our participants also pointed out that while they knew many patients, they did not know everyone. They were reluctant to hold sensitive and emotional ReSPECT-type conversations with someone they had not met before because of the lack of rapport and knowledge about the patient. A lack of experience and confidence in end of life planning also prevented GPs from initiating ReSPECT-type conversations.
While participants reported barriers to completing ReSPECT in primary care, they ultimately thought that it was the place to hold such conversations, and were critical of some of the ReSPECT forms they had seen completed in hospitals. The GPs were sympathetic that the busy hospital environment was not well suited to in-depth conversations about a patient’s end of life care. They also thought that forms were being completed by junior doctors who were inexperienced, and who were “not necessarily savvy to what language is best at home” (FG5) where the use of medical jargon could confuse and worry patients.
ReSPECT is an emotional process.
Many GPs reported gauging how emotionally prepared a patient was to have the conversation by assessing their reaction when the topic was raised. If the patient reacted with alarm or withdrawal, the GP delayed the conversation until a later date. They described these initial contacts as “warning shots” to give the patient time to emotionally prepare themselves.
Many GPs planned a ReSPECT-type conversation for a time when the patient’s family could be included, to provide emotional support and ensure that everyone understood the plan regarding the patient’s end of life. Although inclusion of the patient’s family created potential for conflict, participants suggested that the relationship between themselves and the family could be improved by holding such conversations: “generally once you’ve had that discussion and you’ve reached a decision, it can actually be quite, quite a positive relationship going forwards. And families are very, very grateful for that input […] They’re incredibly grateful for the time that you put in with them” (FG2).
A small number of participants reported not feeling emotionally affected themselves by ReSPECT-type conversations. This was rationalised as being a consequence of experience. Additionally, the GPs typically had these conversations with older patients approaching end of life, and while they acknowledged this was sad they felt less psychological burden than when having such conversations with younger patients.
More frequently, the GPs found it hard to maintain an emotional distance, particularly when they knew the patient well. The GP’s emotional reaction could be affected by the patient’s reaction to the conversation. If the patient reacted positively GPs felt they had “done a good thing” (FG3), and felt a sense of pride in the process. However, distressed patients left the GPs feeling upset, having gone on the “journey” with them.
Conversations are driven by cultural understandings of death.
Participants held implicit understandings of what a “good death” was for their patients. Typically this was “peacefully” with no CPR or invasive treatment, either at home or in a hospice. This death was described as something that patients usually wanted for themselves; their “best case scenario”. As GPs were holding ReSPECT-type conversations with patients they expected to die fairly soon, these understandings often underpinned their conversations and the medical recommendations they recorded: “we've discussed […] whether they don’t want to be admitted to hospital, just die peacefully at home or die, want to die in the hospice” (FG4). Some GPs were aware that their understandings were culturally-bound. They described how in some cultures a ‘good death’ meant trying to maintain life for as long as possible, rather than focussing on quality of life: “the main thing that culturally they would want to, to kind of, keep that person going and, and give them the best chance possible […] as opposed to thinking, “Hang on, about, what, let’s think about the quality of their life left,” and I think that’s a big cultural, thing” (FG3).
GPs’ understandings of a ‘good death’ for their patients sometimes conflicted with the wishes of the patient themselves. Typically, disagreements were around resuscitation when patients (or their family) wanted CPR attempted and the GPs felt this was inappropriate. During such conflict our participants would try to nudge the patient/family in the specific direction that they thought appropriate. For example, participants described how patients did not understand what resuscitation meant, and they would stress that there was no guarantee that CPR would work. Others described in blunt terms what resuscitation would entail, persuading patients that it was inappropriate: “What do you want [paramedics] to do? Do you want them to push your husband away and assault you? Or do you want them to check that your heart has stopped? And then put an arm around your husband and make him a cup of tea?” (FG5). Participants observed that patients often believed that if they agreed not to be resuscitated, they would receive no treatment or care at all. They described reassuring patients that they would still receive good care but it sometimes took several consultations for the GPs to explain why resuscitation or hospital admission would not be in the patient’s best interest.
GPs noted that families often struggled to discuss end of life care. They felt “scared” of using terms such as ‘death’ or ‘dying’ and felt that by raising the topic they were “condemning” their relative. Participants suggested that they should be ones to initiate the conversations and use phrases such as “when you die” to remove that burden from families. Our participants theorised that resistance to discussing death was grounded in fear. It was acknowledged too that doctors can be fearful and subtly feed into patients’ fear by reinforcing the need to take medication to prevent death: “I think sometimes as doctors, there can be a fear about death. And sometimes we’re the biggest culprit for that, and kind of feeding into patients’ ideas that you can’t die, and you’ve just kind of gotta keep taking the tablets, keep alive” (FG2). A few participants were keen to break down taboos surrounding discussion of death in order to normalise decisions about end of life care. They suggested that holding ReSPECT-type discussions earlier, on first diagnosis or as part of a routine check-up, would help to normalise it for both themselves and for their patients.
There can be difficulties translating ReSPECT across care settings.
GPs gave examples where ReSPECT recommendations had translated into care. For example, paramedics had used ReSPECT to decide whether to transport a patient to hospital or not. However, they also described situations when the recommendations recorded on ReSPECT were not transferred into care. Generally this involved patients being admitted to hospital, despite a recorded preference for non-admission. This was sometimes because of a lack of service availability, such as home care support services or hospice beds. Sometimes, patients were transferred into hospital because their family could no longer cope with caring for them at home. Several participants mentioned that translating ReSPECT recommendations into care could be difficult in nursing homes because occasionally the staff-to-client ratio meant that if a client deteriorated then healthcare staff would be limited in the emotional and physical support they could provide: “you’re the only nurse looking after 80 clients overnight and one goes off it’s actually really difficult. Even if it says not for resuscitation, not for admission, you’re the only nurse, what does that actually mean, how do you kind of support?” (FG1).
The GPs were aware that their lack of knowledge of specialist interventions and treatments available within acute care could be a barrier to completing a ReSPECT that was meaningful in hospital. One participant suggested that ReSPECT should be initiated in secondary care and reinforced in primary care, because GPs “would struggle to have that detailed conversation” (FG3). The GPs talked about focussing in more general terms on preferences around hospital admission and resuscitation, and on treating chronic or terminal illness rather than emergencies. The ReSPECT form was seen as an important document that patients should take with them to hospital, whatever the cause of admission. Participants were keen that, having gone through a lengthy process with the patient, ReSPECT recommendations should be used to inform care. However, they acknowledged that it is difficult for any health professional not to actively treat patients: “you’ve got somebody who’s palliative care, advanced cancer, bed bound, falls out breaks their leg, and goes in and, and absolutely everything gets scanned from head to toe […] they may well have expressed that that’s not what they wanted. It’s very difficult not to just treat the bit that needs treating” (FG1).
Some GPs had not seen hospital-issued ReSPECT forms even though their patients’ discharge letters mentioned ReSPECT. These participants expected to see a copy of the ReSPECT form with the discharge letters and felt disadvantaged by this apparent lack. They were unclear whether the form was kept with the patient or in hospital records. Participants felt that an electronic version transferable between settings would be beneficial “so that all the different people providing care for a particular patient have got the […] same kind of document that they can resort to in terms of palliative care and patient’s wishes” (FG4). To gain a digital copy of the form they had completed, GPs created workarounds, such as manually transferring ReSPECT recommendations to their electronic records in the surgery. These tended to increase their workload and diverged from the intended usage (hard copy held by patients).
When our participants had seen hospital-issued ReSPECT forms, they thought that they were too focussed on specific treatments available in hospital, or were used as replacement DNACPR forms. Important but uncomfortable topics, such as where a person wants to die, were not discussed: “They might deal with […] IV, antibiotics, fluids. But they don’t properly discuss, like, hospice or, you know, things, where you want your end of life to be, and, which are a bit more challenging, I think, for us to discuss” (FG4). These comments suggest that ReSPECT could be being used for different purposes in primary and acute settings.