Article Text

Download PDFPDF

What challenges good palliative care provision out-of-hours? A qualitative interview study of out-of-hours general practitioners
  1. Mark Taubert1,2,
  2. Simon I R Noble2,3 and
  3. Annmarie Nelson4
  1. 1Marie Curie Hospice Penarth, Penarth, South Glamorgan, UK
  2. 2Palliative Care Department, Velindre Hospital NHS Trust, Cardiff University, Cardiff, UK
  3. 3Department of Palliative Medicine, Royal Gwent Hospital, Newport, UK
  4. 4Wales Cancer Trials Unit, Marie Curie Cancer Care, Neuadd Meirionnydd, University Hospital of Wales, Cardiff, UK
  1. Correspondence to Dr Mark Taubert, Marie Curie Hospice Penarth, Bridgeman Road, Penarth CF64 3YR, UK; mtaubert{at}


Background Not much is known about how general practitioners (GPs) who work regular out-of-hours shifts in the community feel about prescribing medication for symptom control in end-of-life care patients, how they gain palliative care experience and what they perceive as benefits and hindrances within service delivery to this vulnerable patient and carer group.

Objectives To determine, by interviewing GPs who provide out-of-hours care, aspects of care provision that augmented or challenged palliative care delivery.

Methods Semistructured interviews were conducted with GPs who worked out-of-hours shifts regularly. All interviews were analysed using interpretative phenomenological analysis.

Results Out-of-hours doctors were not confident about their palliative care knowledge and expressed a wish for more education. Lack of familiarity with patients requiring acute palliative care, compared with the closer bonds formed within the in-hours general practice setting, was perceived as troublesome and problematic, and lack of follow-up was felt to be a major factor. There was a clearly expressed fear of killing or harming patients with the strong drugs used in palliative care. Opiophobia, especially when faced with patients who were seen only once, affected prescribing decisions.

Conclusions This study provides further evidence of the needs of a specific subgroup of GPs. In-depth interview analysis of their experiences with palliative care provides a framework for which aspects are perceived as a hindrance in providing good palliative care, and can inform out-of-hours policy and teaching.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.



Provision of out-of-hours care by primary care doctors is a crucial element of primary healthcare in many countries, including the UK.1 The working time for ‘out-of-hours’ shifts in the UK is defined as the period between 18:30 and 8:00 on weekdays, at weekends from Friday evening to Monday morning, and on bank and public holidays.2 General practitioners (GPs) who choose to work in this setting are expected to see palliative care patients as a priority—for instance, to treat pain, nausea, vomiting or any other problems that arise in this often very unwell group of patients. Patients with cancer are estimated to spend 90% of the last year of their lives in their homes and at least two-thirds of this falls within time that is defined as out-of-hours.3 4 In Wales, the 2008 ‘Sugar report’ on palliative care provision concluded that ‘The current way of working means much of the out-of-hours services are delivered by staff who lack the relevant skills. Out-of-hours support should be delivered by GP or district nursing staff who have been sufficiently trained in palliative care and have access to 24/7 specialist support and advice’.5 Based on these findings, this study set out to explore what specifically GPs felt worked well and what worked less well, and by using semistructured interviews we facilitated a free flow of ideas and information from out-of-hours doctors.

What is already known on this topic

  • Since 2004, following contractual changes, a large number of general practitioners (GPs) stopped working out-of-hours shifts.

  • Out-of-hours organisations across the UK see patients in the community for a substantial time-period outside regular working hours in a typical 7-day week and this includes palliative care patients and their carers.

  • Out-of-hours support should be delivered by GP or district nursing staff who have been sufficiently trained in palliative care and have access to 24/7 specialist support and advice.

What this study adds

  • Out-of-hours doctors expressed a wish for palliative care knowledge and felt they needed educational support regularly.

  • Lack of familiarity with these patients and their carers made care provision more difficult for doctors.

  • Out-of-hours doctors worried about killing or harming their patients with the strong medications commonly used in palliative care.

Structure of UK out-of-hours provision and changes in 2004

Out-of-hours services in the UK are provided by a range of organisations, including GP co-operatives, mutual organisations, commercial deputising services or ambulance services. Consultation options include advice via the telephone, face-to-face clinical assessments at designated out-of-hours clinics and home visits. These services are provided by a range of professionals, including doctors, nurses and paramedics. Doctors working for out-of-hours services are GPs and they are required by law to be registered on an NHS board's performer list in order to work within that NHS board area. This ‘performers list system’ gives health boards across the UK responsibility for monitoring the conduct of GPs. They have admission, suspension and removal procedures intended to parallel those for doctors in secondary care.6

Organisation of out-of-hours care has recently changed in many countries in Europe.7,,9 When the new out-of-hours provisions under the General Practitioner Contract was introduced in the UK in October 2004,10 a large percentage of the overall GP workforce opted out of working in this specific setting.11 Those GPs who still work for out-of-hours services are paid hourly rates by the organisation that provides care within the locality, and remuneration can be high depending on how antisocial the shifts are deemed to be, including night, weekend and public holiday shifts.

Local structure of out-of-hours services and access to specialist palliative care

Cardiff and its immediate surroundings has a population of approximately 330 000 people and out-of-hours services in the city in 2008 were tendered to a private multinational company called Serco.12 Serco is a FTSE100-listed company with extensive healthcare provision experience across the world, and has a track record of organising out-of-hours services within other health trusts across the UK. At the time these interviews were conducted, Serco provided out-of-hours community services to more than one million people in Cardiff, Oxford and Cornwall.

GPs requiring advice both in-hours and out-of-hours on palliative care issues have access to a specialist palliative care service that operates at all times and is accessible via a local hospice in Cardiff.


A literature review revealed a lack of qualitative data on identifying the main issues surrounding palliative care provision in the period outside normal working hours, especially since the advent of the new service contract for GPs in 2004. The aim of this study was to interview GPs who work out-of-hours shifts regularly and to explore problems relating to palliative care and symptom control in this particular setting. We devised a semistructured interview to enable participants to derive their own themes and establish whether there were particular needs and concerns they had pertaining to this patient group, and if so what these might be. Our aims were therefore very general, and we narrowed down emerging themes as they arose in interviews.


Participant selection

Sampling was purposive, seeking out only GPs who worked regular out-of-hours shifts and were likely to have had palliative care concerns in this setting. In total, 60 doctors employed by the out-of-hours service in Cardiff were contacted, and were sent a letter and participant information sheet. GPs who had worked out-of-hours shifts locally for at least 12 months were chosen, in order to ensure they had, at the point of interview, sufficient experience with palliative care patients in this specific setting. A total of nine GPs came forward. No further recruitment drive was undertaken at this stage, when it became clear that the nine interviews had yielded abundantly rich data. No new data were obtained after interview number seven once all nine interviews had been analysed for themes.

Data collection

For the purpose of data collection, interviews were audio-taped and field notes were taken by the principal investigator (MT). Interviews were semistructured. There was a loose interview structure for the first interview, which had been informed by a trial ‘dry-run’ interview with another GP who was not one of the nine participants. These interview prompts evolved during subsequent interviews to include topics that had arisen in each individual interview and we have included our final interview schedule in figure 1, to give readers an idea of the process (figure 1). The shortest interview lasted for 39 min, while the longest interview was 81 min, with an average duration of 58 min. Interviews were subsequently transcribed verbatim within a week and anonymised.

Data analysis

Data from transcripts were analysed by MT and AN using an approach called interpretative phenomenological analysis (IPA). This qualitative research method seeks to ‘make sense of the participant making sense of their own world’ at the same time allowing the researcher to interpret the participant's views and perspectives.13 14 It is based on an idiographic approach starting with a single case and then developing more general categories through a case-by-case analysis.

After each interview had been conducted, a transcript was written within a week of the interview. Two investigators (MT, AN) read and re-read the transcripts in turn independently. Important comments and reflections, deemed relevant to the overall topic, were highlighted. Both researchers compared their findings regularly and agreed on relevant headings for themes. Following this, final titles for all themes were chosen and agreed upon by AN and MT. If a new topic came up in the sixth consecutive interview, for instance, then previous interview transcripts were analysed with this specific new item in mind, as is typical for this research methodology. A third researcher (SIRN) was asked to review the quotes and to select those he felt were most representative of the theme emerging.


The nine GPs interviewed had a broad range of experience, and some of the salient participant data are summarised in table 1. Interviews generated a large volume of data on different topics and two other studies have been published with findings from this project, one focusing on information exchange between services and the other seeking a clearer picture of processes within the out-of-hours setting—for instance, note keeping and time pressures.15 16

Table 1

Participant data

Three key areas that were seen as major challenges by GPs emerged from the interviews:

  • learning and knowledge base;

  • doctor–patient–carer barriers;

  • fear of prescribing and altering doses.

Learning and knowledge base

The interviewees commented on various aspects of learning and knowledge acquisition in symptom control and palliative care. There was a common thread of perceived learning needs in palliative care and lack of basic background knowledge among these doctors, as exemplified by the following statement. GPD: … you forget sometimes, I think you sit at your desk in GP with the internet in front of you and the books on the shelf, that, you are, you have big gaps in your experience and knowledge when it comes to things like that. Like symptom control in palliative care and I mean it's in the out-of-hours and the evening that you find those gaps and it's not just in palliative care, it's like, more likely to be a fissure.

GPG expressed the view that he felt at times inadequate because of a lack of background palliative care knowledge. GPJ expressed insecurity when the topic of knowledge was broached in the interview, using strong negatives such as ‘not totally au-fait’ with a strong emphasis on the word ‘totally’ and then later emphasising the word dangerous in ‘I don't think I'm dangerous’, implying that lack of knowledge has degrees of consequence and that she was at least functional. The above interpretations relied on the interviewer's ability to observe the interviewees and there levels of discomfort when talking about these matters and interpreting their words. The statements below and the interpretations that have been made will give the reader a better understanding of how IPA works.

GPJ: I guess you could, I guess that could be better, if I had more experience. More knowledge on symptom management in specific palliative care situations, really.

GPJ: You know, it's not an area where, errm, where I feel totally au-fait.(laughs).

MT: OK. And do you feel confident then when you deal with difficult symptoms out-of-hours. We've touched on parts of that [but]?

GPJ: [yeah] errm, I'd like to feel more confident. I don't think I'm dangerous.

GPC felt that all GPs (including in-hours) had learning needs in palliative care, especially when it came to the UK integrated care pathway for the last days of life.17 18 She had brought up the subject of the pathway in the initial stages of the interview, indicating a high level of familiarity with it and frequently using the recommended medication list in the 2007 version of this document.

GPC: … a lot of GPs won't know the pathway, which is a little unfortunate.

GPC: … a lot of in-hours GPs (…) out-of-hours GPs don't know about it or don't think it's relevant to them.

Her solution was that GPs, including those working ‘in-hours’ required teaching about end-of-life problems, reflecting her lack of confidence in most GPs' ability to manage symptom control well (as was shown in the two sentences above).

GPC: I think all GPs should have end-of-life teaching, should know what the pathway looks like.

GPG felt similarly.

GPG: But we need some more of sessions specifically for out-of-hours palliative care, that one we had was very useful, it was very relevant to out-of-hours practice, focus on prescribing and doses, converting doses over 24 hours, but we need it on a regular basis.

There was a general frustration among GPs about not being able to ‘learn on the job’ in out-of-hours palliative care, because nearly always the outcomes of their own actions/interventions were not known to them. This reflects the shift system in out-of-hours and the lack of regular doctors working in the system. GPA for instance found this quite upsetting, using an uncharacteristically (compared with the rest of the interview) evocative adjective, ‘horrid’.

GPA: And then you've got a patient and you never know what you've done ever helped them I suppose, so that's quite horrid really.

GPH: … you just, hope that they're going to respond to the changes that you make, rather than, kind of know that they are …

GPG: There's no follow-up, you know, there, that is a problem. You learn, you don't learn from it …

GPJ used humour in the interview, which was interpreted as her mitigating a serious point, that of being worried about making a mistake and another doctor having to pick up the pieces.

GPJ: Yeah. Definitely it stops you from learning. Particularly of what I was saying before in palliative care, you might have made some changes to their drug regime overnight and you don't know if it's worked or whether the normal GP's gone in the next working day and gone ‘Oh my god what have they done!’(says this in high-pitch) (laughs).

The above makes for interesting reading, given that GPJ seems to see little learning experience from one-off out-of-hours encounters. Also, GPH feels that the main palliative care training out-of-hours doctors get is from their in-hours workload, dismissing potential learning experiences within out-of-hours. This possibly reflects a preference for a problem-based learning approach involving follow-up and learning from mistakes or non-success. On a similar point, all of the GPs interviewed also did in-hours work; however, this leaves questions, not explored in detail here, about those GPs who do only out-of-hours work.

MT: Right. Do out-of-hours doctors get much training in palliative care?

GPH: No, not especially, I think it depends what you pick up along the way of your in-hours GP work, that's when you learn most really.

Interestingly GPJ was the only interviewee to point out potential de-skilling in palliative care.

GPJ: Sometimes you see a lot of palliative patients in a row, then you don't see that many for quite a while and you can forget a lot of your previous knowledge in the symptom-control field, you know.

The general self-perceived lack of palliative care knowledge among out-of-hours doctors had potential for more direct patient care consequences according to one interviewee. He does not express it directly in this exchange, but the lack of knowledge appears to have an effect on his prescribing habits, perhaps making him less inclined to increase a patient's drug dose, or certainly making him more conservative than he might feel appropriate. There is a subtle sub-theme of fear of recrimination here, and this may well be directly linked to lack of knowledge in the field of palliative care.

MT: Would you say, that makes you more conservative in your management?

GPF: (Sighs) If they are already on a high dose of opiates it's probably alright. I usually don't have a problem with something that's already been prescribed, which you know, sounds like an awful thing to say. But if, if, if, if, an adjustment is required, then it can become difficult.

MT: An adjustment?

GPF: Yeah, for example an adjustment on syringe driver or an adjustment, that errm, you know the Oramorph isn't working, so, you know, what can we do now? And then, if they're on different medication, then you don't really know what, which ones to increase which one's to decrease and these things are common, yeah.

MT: So from that, would you say you underestimate doses or increases, or stay very conservative to stay safe?

GPF: You mean that one prescribes less rather than more? (sighs) Errm, (silence 3 seconds) I don't know. I can't really say (silence 2 seconds). Maybe sometimes, maybe subconsciously you do that (silence 2 seconds). But I can't really give you a straight answer on that one.

He appears to think very carefully about his answers at the end, as indicated by uncertainty and several long silences and perhaps does not feel confident to elaborate on how his symptom management is affected out-of-hours. This, during the interview, felt like a significant exchange, perhaps because of the honesty and the ‘thinking aloud’ aspect which is evident here.

In general, it was felt that teaching sessions for out-of-hours doctors were not well attended. Interviewees spoke of non-attendees in the third person usually, thereby disassociating themselves from this culture of avoiding teaching.

GPA: … they might find it difficult enough to attend teaching, they all do extra work as well.

GPE: … whether you'd get many out-of-hours doctors bothering to attend, I can't see it really,

One doctor felt that to feel comfortable with palliative care a previous job in the field was beneficial.

GPF: … other doctors may be more comfortable in palliative care than me, they may have done junior doctor jobs in palliative care, again that is something I never did.

Doctor–patient–carer barriers

It was clear that there were problems that did not help the transaction between doctor and patient, and doctor and carer in the out-of-hours setting. Interviewees were at pains to point out the lack of a patient–carer–doctor relationship, and that they frequently entered and exited the patient's world in one short encounter.

This was frequently compared with in-hours general practice, were the patient–carer–doctor relationship is probably one of the most enduring compared with other medical specialties. GPA compares this to a journey. This familiarity factor, as GPB puts it, creates an ease that is not easily replicable in out-of-hours.

GPA: … you've taken them through the whole palliative care journey (in-hours).

MT: You mention as well that you don't have much information on the patient.

GPB: None, I think that makes a big difference, again I mentioned the familiarity factor which puts the doctor and the patient and family at ease, you know, errm.

This GP therefore appears to feel that he can be more honest and perhaps a bit bolder with his patients, when he knows them well. On a similar line, GPE feels that a lot of the relational restrictions fall away when she knows the patients and carers.

GPE: It makes it much, much easier if you know who they are, it cuts out a lot of the, the you know, formalities, a lot of the barriers.

The fact that this lack of familiarity causes increased barriers between doctor, patient and carer is exemplified by the following extracts.

Difficult, hard and uncomfortable are some of the negative words suffusing the following statements, concerning these encounters between strangers, and lack of trust is mentioned.

GPA: You don't know anything about the patient, it's a lot more difficult really.

GPA: It's a knowledge of the patient, thinking back on situations that makes it very hard really.

GPA: You don't know that family and you can feel quite uncomfortable.

GPF: Bloody great! I don't even know him or the family. I'm not their GP. You know, two hours later etc., etc.

GPH: Like, who is this out-of-hours doctor, I've never met her before, why should I trust her anyway?

Lack of familiarity with patients and their carers was therefore felt by interviewees to be important contributing factors to lowering confidence when seeing palliative care patients out-of-hours.

Fear of prescribing and altering doses

Some GPs were less confident than others when it came to prescribing drugs for palliative care patients. There was cross-over here with the theme of ‘learning and knowledge base’ described above, but it was felt that this deserved to be a further sub-theme in its own right. From nearly every interview a fear about harming or even killing patients with medication was discerned.

GPA: … am I going to kill them?

GPA: … don't want to give them too much (medication)

GPC who appears to be more confident about prescribing these drugs for her patients, acknowledges that she believed other out-of-hours doctors have a fear of this aspect of work.

GPC: I know I write them up precisely and I know not every GP would do that, as they wait until they absolutely have to …

GPC: … or they're (other out-of-hours doctors) worried because of the nature of the meds …

GPF: You want to help the patient's pain, but on the other hand you don't want to knock them out completely or even kill them, so you worry of what dose to go to with the analgesia.

GPF: … you write them up and hope for the best …

GPF in the above describes a loss of control, he feels he is exceeding the prescribing limits of controlled drugs, and while he is uncomfortable with this, he writes them up and ‘hopes for the best’ rather than being assured in his actions and happy with his decisions.

MT: Is that a fear-factor, the high doses?

GPF: It's, well, (silence 5 seconds) Yes. If you are in unknown terrain as a doctor, then you try to stick to the rules as, as closely as possible.

GPF: … It's a fear-factor that you're going to damage the patient, but, errm, injure or hurt a patient or do, do, do anything detrimental to him.

He gives an extreme example to show why he is fearful, and this gives a good description of the fear that the patient may die soon after.

GPF: And you don't want to, and I guess you're always quite aware of this as a locum or out-of-hours, you don't really want to go in the patient's home as the unknown out-of-hours doctor, make a big change and leave and then the patient dies half an hour later.

The fears he expresses appear to be grounded in medicolegal recriminations.

Syringe drivers were a frequent topic in interviews and nearly all out-of-hours doctors had something negative to say about them, either difficulty of access or not knowing how to set them up.

GPB: I don't know, there's always the thing with syringe drivers, you're having to make sure that the drugs you know, I've never set up a syringe driver set …

GPF: … how often as a GP do you write up a syringe driver?

GPF: … coming up to something like ‘It's five mls over 24 hours' and you're like ‘What?’ …

GPF: … no training in syringe drivers, it's usually organised by the nurses …


This study demonstrates some of the areas that out-of-hours GPs perceive as burdensome in the provision of palliative care. Several barriers are suggested, including the fact that doctors, patients and relatives do not know each other. Furthermore, difficulties in gaining additional knowledge and education in palliative care were highlighted. Concerns were raised about prescribing, in particular, relating to opioids and syringe-driver pumps. ‘Opiophobia’ has been described in a general doctor population and the comments made by this group of GPs appear to compound this phenomenon, especially when they are faced with high-dose opioids in the out-of-hours setting and when having to potentially increase doses further.19

For some, low confidence was also associated with never having been trained in palliative care specifically—for instance, having done a palliative care hospital job during training. Higginson's scoping exercise in 200720 concludes that there is a gap in research in the area of what is required for ‘improving education and support to out-of-hours practitioners’. The interviews presented here suggest that out-of-hours GPs feel enthusiastic about such education but are not sure how good the general uptake among the out-of-hours doctor population would be. Topics that GPs were keen to be taught included the last days of life care pathway and prescribing and dosing in palliative care. GPs acquired their palliative care knowledge mainly from their in-hours work.

The study had nine participants and no further recruitment drive was undertaken. IPA research focuses on a smaller number of often ‘complex’ or unusual participants, offering a detailed account of their experience. Historically, researchers seeking to make an impact in research settings are more familiar with quantitative paradigms and this has meant using larger sample size numbers in IPA studies to appease reviewers.21 Smith now suggests that just three to six participants are a reasonable size for an IPA study. This number allows the identification of similarities and differences in each group without overwhelming the rich description.

While generalisability is not an aim of IPA studies, the geographical focus may be a concern to some readers wishing to extrapolate data to their own areas of work. However, the needs and concerns expressed by this group of doctors may serve as an important lesson to what makes or does not make for good palliative care provision within out-of-hours general practice, and their concerns seem sufficiently generic (eg, opiophobia) to suggest that they may be a problem for out-of-hours care as a whole, not just in one location and not just for these doctors.

This may also have been a highly motivated, self-selecting group of out-of-hours GPs who were enthusiastic about palliative care, but it is hard to comment about this and the language they used suggested that this was an area of huge uncertainty. The fact that they came forward suggests a need to share information about good and not so good aspects of this highly focused area of care. IPA is less concerned about reproducing repeat statements to highlight an issue, but is rather more interested in unusual, novel, outlying information, that perhaps a survey might not detect.


This study provides further evidence of the needs of a specific subgroup of doctors who work community out-of-hours shifts regularly. Individual quotes captured during the interview processes were systematically analysed and interpreted. The resulting in-depth analyses of out-of-hours experiences of palliative care provide us with a framework and a needs assessment of which aspects are perceived as a hindrance in providing good palliative care, and should inform future out-of-hours policy, funding and teaching. Specific areas of concern highlighted by these GPs were lack of palliative care knowledge base, fear of prescribing strong pain medication, especially in high doses, the use of syringe-driver pumps and a lack of any relationship with the patients or family carers.


The authors would like to thank the out-of-hours doctors they interviewed for giving them their valuable time.



  • Competing interests None.

  • Provenance and peer review Not commissioned; externally peer reviewed.