Improving communication with palliative care cancer patients at home - A pilot study of SAGE & THYME communication skills model.

European Journal of Oncology Nursing xxx (2015) 1e8

Contents lists available at ScienceDirect

European Journal of Oncology Nursing journal homepage: www.elsevier.com/locate/ejon

Improving communication with palliative care cancer patients at home e A pilot study of SAGE & THYME communication skills model Jane Griffiths a, *, Charlotte Wilson b, Gail Ewing c, Michael Connolly d, Gunn Grande a a

University of Manchester, School of Nursing Midwifery and Social Work, Jean McFarlane Building, Manchester, United Kingdom Barts & The London School of Medicine, Centre for Primary Care and Public Health, Yvonne Carter Building, 58 Turnser Street, London, United Kingdom c University of Cambridge, Centre for Family Research, Free School Lane, Cambridge, United Kingdom d University Hospital of South Manchester NHS Foundation Trust, Southmoor Lane, Wythenshawe, Manchester, United Kingdom b

a b s t r a c t Keywords: District Nurse Palliative care Communication skills

Purpose: To pilot an evidence-based communication skills model (SAGE & THYME) with UK District Nurses (DNs) who visit patients with advanced cancer early in the dying trajectory. Evidence suggests that DNs lack confidence in communication skills and in assessing cancer patients' psycho-social needs; also that they lack time. SAGE & THYME is a highly structured model for teaching patient centred interactions. It addresses concerns about confidence and time. Method: Mixed methods. 33 DNs were trained in SAGE & THYME in a three hour workshop and interviewed in focus groups on three occasions: pre-training, immediately post-training and two months post-training. Questionnaires measuring perceived outcomes of communication, confidence in communication and motivation to use SAGE & THYME were administered at the focus groups. Results: SAGE & THYME provided a structure for conversations and facilitated opening and closing of interactions. The main principle of patient centeredness was reportedly used by all. Knowledge about communication behaviours helpful to patients improved and was sustained two months after training. Increased confidence in communication skills was also sustained. Motivation to use SAGE & THYME was high and remained so at two months, and some said the model saved them time. Challenges with using the model included controlling the home environment and a change in style of communication which was so marked some DNs preferred to use it with new patients. Conclusion: Training DNs in SAGE & THYME in a three hour workshop appears to be a promising model for improving communication skills when working with cancer patients. Crown Copyright © 2015 Published by Elsevier Ltd. All rights reserved.

Introduction Patients are living longer with incurable cancer. In the UK, the majority spend their last year at home managed by generalist primary care practitioners, predominantly District Nurses. These patients are known to have high psychological morbidity. Overall, depression and anxiety is around 20e30% (Hotopf et al., 2002) but there are various stages in the disease trajectory when patients are particularly vulnerable (Morse, 1992; White and Mcleod, 2002; Thekkumpurath et al., 2008). Incidence of depression and anxiety

* Corresponding author. E-mail address: jane.griffi[email protected] (J. Griffiths).

is high at diagnosis and in the following six months, and increases again with advanced and metastatic disease and when the prognosis is poor. Assessing psycho-social needs is required when disease progresses, active treatment stops and until death (Thekkumpurath et al., 2008). District Nurses (DNs) visit patients with advanced cancer early in the dying trajectory to support them and their families, and build a relationship before hands-on, intimate care is needed at the end of life (Seale, 1992; Austin et al., 2000; Mok and Chui, 2004; Ohman and Sodreberg, 2004; Coffey, 2006; Stolz et al., 2006: FingfeldConnett, 2007). DNs are confident and skilled in providing practical support and physical care (Griffiths et al., 2012) and are ideally placed to identify and respond to psychological concerns (Luker et al., 2000; Pateman, 2003; Kennedy, 2005; Griffiths et al., 2010).

http://dx.doi.org/10.1016/j.ejon.2015.02.005 1462-3889/Crown Copyright © 2015 Published by Elsevier Ltd. All rights reserved.

Please cite this article in press as: Griffiths, J., et al., Improving communication with palliative care cancer patients at home e A pilot study of SAGE & THYME communication skills model, European Journal of Oncology Nursing (2015), http://dx.doi.org/10.1016/j.ejon.2015.02.005

2

J. Griffiths et al. / European Journal of Oncology Nursing xxx (2015) 1e8

In some studies, DNs report that they have good communication skills and provide skilled psycho-social assessment to cancer patients (Austin et al., 2000; Wright, 2002; Kennedy, 2005). Other evidence suggests however that DNs lack confidence in communication skills and in assessing cancer patients' psycho-social needs; also that they lack time for this work (Dunne et al., 2005; Aitken, 2006; Addington-Hall et al., 2006; Griffiths et al., 2007). For example in our observation study of palliative home care we found that DNs tend to avoid cancer patients' cues of distress and block exploration of their concerns (Griffiths et al., 2010). Connolly and colleagues have developed an evidence based communication skills model (SAGE & THYME), taught in a three hour workshop, that addresses concerns about confidence and time (Connolly et al., 2009). SAGE & THYME is a mnemonic that guides the nurse through exploring concerns, patients' own strategies for dealing with them, when to offer advice and when to refer for specialist help (Bandura, 1982; Maguire et al., 1996; McCormack and McCance, 2006). It encourages person-centered interactions by helping the nurse to hold back with premature advice that can close down interactions. SAGE guides practitioners through active listening, THYME through simple patient centred problem solving (Fig. 1). SAGE prompts the nurse to organize the Setting; Ask about concerns; Gather all concerns and Empathise. THYME prompts them to ask the patient who they have to Talk to; whether this Helps; what You (the patient) think would help; what the patient would like Me (the nurse) to do; and finally to End/summarise the discussion. Nationally, 26,000 health care practitioners have been trained in SAGE & THYME in the UK. It has been evaluated in hospital settings and found to increase nurses' knowledge and confidence (Connolly et al., 2009, 2014). The aim of this study was to explore whether SAGE & THYME has similar utility with District Nurses in the palliative cancer home care and which factors facilitate and hinder its effective use. Methods Design Mixed methods study using focus groups and questionnaire survey.

Training During the study, District Nursing services in four Trusts in North West England took part in SAGE and THYME training. The workshop lasted three hours and included brief taught components, group work and role play. Post workshop, the DNs were given a prompt card to take to patient visits to remind them of the sequence of the model (Fig. 1). Recruitment and sample DNs were recruited by the DN clinical leads in each Trust. Inclusion criteria were: RGN qualified, working as a DN, and more than one patient with cancer on the caseload. Ethical approval Ethical approval was granted by the Local Research Ethics Committee (NRES:Reference 11/NW/0525). Focus groups The DNs were interviewed in focus groups on three occasions: pre-training, immediately post-training and two months posttraining. The pre-training and two months post-training focus groups were held at local health centres and the immediately posttraining focus group at the training venue. Each group had between 5 and 12 participants. All focus groups were audio-recorded and were 30e90 min in duration. 40 DNs took part in the study, but there was attrition at the different stages. 40 DNs attended the pre-training focus groups, 33 attended the training itself and focus groups held immediately after, and 26 attended the two months post-training focus groups. Five DNs who could not attend the two months post-training focus groups were interviewed individually on the phone. The other two were unavailable. This paper reports findings at each stage. Focus groups allowed in depth exploration of experiences, and promoted sharing and discussion of ideas (Kitzinger, 1995; Kaufman, 1996; Kreuger, 2008). They were facilitated by two researchers using a topic guide. One of the researchers (JG) led the discussion, and the other researcher (CW) took notes, checked equipment, probed for fuller responses from participants, and ensured that all of the topics had been covered. The guide was flexible to allow participants to introduce other topics of relevance to the discussion. The purpose of the pre-training focus groups was to explore DNs' experiences and challenges of assessing the psycho-social needs of patients (reported in Griffiths et al., 2014), and to ascertain whether DNs perceived a need for communication skills training. The immediately post-training focus groups asked DNs to consider whether SAGE & THYME would be useful in their practice. The two months post training focus groups captured DNs' reflections on the training, whether it needed to be modified in any way, the use and utility of SAGE & THYME in practice, and barriers and facilitators to its use. Questionnaires

Fig. 1. SAGE & THYME model.

Three questionnaires were used that had been developed by the Maguire Communication Skills Training Group for use in a previous, hospital based evaluation of SAGE & THYME. The first was an ‘outcomes’ questionnaire (Table 1, column 1) comprising 19 items about the participant's perceptions of the likely outcomes of their interactions with patients e.g. that asking about concerns or emotions can damage them/be of benefit. Perceptions were rated on a scale of 1e9 from very likely to very unlikely. The second was a



3

Table 1 Perceptions of outcomes of consultations questionnaire. Comparison of pre-training and two months follow up. 1: very likely e 9: very unlikely. The bold values significance p < 0.05. Question

N

Before Mean (SD)

After Mean (SD)

t-value

Df

p-value

1. Asking your patient to talk about their concerns &emotions can be of benefit to them 2. Your patient will become uncontrollably upset if you ask about their feelings 3. Asking patients to talk about concerns will damage the way they cope 4. You will get too close to your patient, if you ask about their feelings/concerns 5. Your workload will become unmanageable if asking about emotions 6. Asking about concerns/emotions will give the patient false expectations 7.You will not ask about feelings/concerns as it is not my role 8. You would be criticised by colleagues if the patient became upset 9. If you talk to patients about their concerns you will feel down 10. The patient may become upset by feelings and you will lose control of the interview 11. You could say something which would harm your patient 12.You would have been helpful, exploring your patients concerns 13. Your patient will only cope if you make them feel everything will be ok 14. Asking your patient about concerns will help you identify what may be helpful 15. Talking about concerns/feelings will distress the patient and make them cry 16.Your patient will show strong emotions which will be overwhelming for you 17. The patient will raise concerns/feelings without my asking 18. By giving information/advice initially this will stop the patient from being upset 19. Asking patients about their concerns/emotions will benefit your management of them

27 27 27 27 27 27 27 27 26 27 27 27 27 27 27 26 27 27 27

2.67 5.11 6.63 6.37 6.00 7.30 8.56 8.15 6.31 6.81 6.96 2.96 5.93 2.00 4.74 6.08 4.19 5.26 2.41

1.96 5.81 7.30 7.19 6.89 7.70 8.52 8.30 6.62 7.52 7.59 2.19 7.00 1.93 5.93 7.31 4.48 5.67 2.37

2.27 1.61 1.65 2.35 2.13 1.44 .19 .42 .67 1.75 1.19 2.16 2.04 .21 2.73 2.70 .84 1.07 .146

26 26 26 26 26 26 26 26 25 26 26 26 26 26 26 25 26 26 26

.032 .119 .110 .026 .043 .163 .852 .681 .507 .092 .245 .040 .051 .839 .011 .012 .409 .296 .885

‘confidence’ questionnaire (Table 2, column 1) that comprised 16 items about participants' confidence in communication skills which were rated 0e100. These two questionnaires were administered at the pre-training focus groups, immediately post training focus groups and 2 months post training focus groups. A third ‘motivation’ questionnaire (Table 3, column 1) was administered at the immediately post-training focus groups and two months posttraining focus groups only. It comprised seven items on motivation to use SAGE & THYME and perceived usefulness of the model, rated 0e100. Data analysis The focus group and individual interview data were transcribed verbatim. Transcripts were read, anonymised and checked for accuracy, then uploaded to NViVo software to facilitate coding and thematic content analysis. Transcripts were analysed concurrently with data collection using a constant comparative analysis approach (Strauss and Corbin, 1998). Data were read independently by JG and CW and the main themes coded. Written summaries of the main themes were presented to the steering group which included the project team, DN team leaders and two service user representatives. The group discussed the main themes, considered rival explanations and agreed on the interpretation of the study findings.

(1.240) (1.783) (1.822) (1.984) (1.881) (1.235) (.80) (1.460) (1.828) (1.388) (1.971) (1.581) (2.286) (1.593) (1.559) (1.742) (1.902) (1.745) (1.575)

(1.315) (1.594) (1.265) (1.798) (1.805) (1.295) (.893) (1.137) (1.899) (1.503) (1.824) (1.039) (1.754) (1.999) (2.018) (1.436) (1.740) (2.287) (1.445)

Paired sample t-tests were used to compare scores for questionnaires 1 and 2 (outcomes of communication, confidence in communication) before and two months after training, and for questionnaire 3 (motivation to use SAGE & THYME) immediately after training and at two months. Statistical analysis was on a on a sub-sample of participants who completed questionnaires on both occasions. Results Forty DNs took part. All were RGN qualified and five held the DN qualification. Years of experience working as a DN ranged from 1 to 18 years (median 12.5). With the exception of one nurse, none had received any form of communication skills training. Team caseloads varied from 150 to 600 depending on the size of the team. Mean daily visits was 12. Patients on the caseload with cancer varied from 75 to 200, which again reflected the size of the team. Qualitative findings In the pre-training focus groups, DNs (n ¼ 40) confirmed that they provided support visits to cancer patients early in the dying trajectory and verified the findings of previous studies that assessing psychological needs can be challenging. Challenges included the complexity of the home environment, family

Table 2 Confidence in communication skills questionnaire. Comparison of pre-training and two months follow up (a higher score means greater confidence). The bold values significance p < 0.05. Question

N

Before Mean (SD)

After Mean (SD)

t-value

Df

p-value

1. Create a setting in which patients can speak openly about their concerns 2. Initiate a discussion with a patient about their illness and concerns 3. Ask a patient directly how they are feeling 4. Ask questions that will increase a patient's disclosure of their feelings and concerns 5. Avoid being distracted by the first concern 6. Listen to and respond in a way that encourages a patient to disclose other concerns and feelings 7. Use empathic supportive comments with patients (not sympathy) 8. Summarise the concern/s you have discussed to check that you fully understand it 9. Check for other concerns that the patient might have 10. Give information, reassurance and advice only at the end of the conversation 11. Make a statement that moves the patient on in the interview 12. Ask about support structures the patient has and whether they are useful 13. Ask the patient what they think might be useful themselves (given their circumstance) 14. Find out whether the patient feels you might offer some support 15. Close a conversation with a patient who has concerns/worries 16. Have a structure for a concerns based interaction with a patient

25 25 25 25 24 25 24 25 25 25 25 25 25 25 25 25

74.00 77.12 84.92 71.40 71.25 77.80 76.25 74.68 78.60 69.20 66.20 72.60 71.40 71.20 66.40 63.80

84.80 88.88 94.0 84.4 81.0 86.4 88.75 89.8 91.2 84.2 82.8 92.0 91.6 88.8 85.2 86.2

2.89 5.55 3.52 3.42 2.47 2.94 4.73 5.99 5.99 5.33 5.82 6.15 6.03 5.56 5.66 5.81

24 24 24 24 23 24 23 24 24 24 24 24 24 24 24 24

.008 .001 .002 .002 .020 .007 .001 .001 .001 .001 .001 .001 .001 .001 .001 .001

(13.463) (12.969) (11.839) (16.678) (14.529) (14.237) (11.136) (11.495) (12.543) (15.591) (14.382) (15.147) (18.344) (14.669) (14.967) (18.044)

(14.396) (9.194) (8.22) (16.762) (14.295) (14.084) (9.918) (10.847) (8.201) (13.895) (13.469) (8.0364) (9.0967) (11.019) (14.964) (11.482)


4


Table 3 Questionnaire C: Motivation and Usefulness of SAGE & THYME. Comparison of immediately post-training and two months follow up (a higher score means higher motivation). The bold values significance p < 0.05.

1a. How strong is your intention to use SAGE and THYME™ in your practice? 1b. How motivated do you feel to use all of the SAGE and THYME™ structure? 1c. How motivated do you feel to use elements of SAGE and THYME™? 1b. How much do you feel that SAGE and THYME™ will be useful for enabling patients? 2b. How much do you feel that SAGE and THYME™ will be useful for you in working with patients? 3b. How strongly do you feel that SAGE and THYME™ will leave you more satisfied in the work you do with patients? 4. Has this training made you more or less likely to have further communication skills training in the future?

N

After Mean (SD)

Two months after Mean (SD)

t-value

Df

p-value

26 26 26 26

93.19 95.12 96.00 93.58

90.54 83.04 96.08 88.04

(12.488) (19.048) (6.330) (13.601)

.902 3.478 .042 1.946

25 25 25 25

.375 .002 .967 .063

26

93.38 (7.915)

88.5 (14.621)

1.599

25

.122

26

90.00 (16.613)

85.8 (19.541)

1.390

25

.177

26

1.92 (.272)

.570

25

.574

(19.722) (7.896) (6.759) (6.801)

1.89 (.326)

dynamics, breaking bad news to children, lack of acceptance of the diagnosis, denial and anger (Wilson et al., 2014). They reported that they had learned how to communicate with patients through experience, were unclear whether they were taking the right approach and had received no communication skills training since beginning work as a DN.

would be that they were dying. She was unsure how the model could help with a concern as seemingly insurmountable as this:

‘I think you develop your communication skills with experience … I've never had any training in, this is how you sit down with a family … this is how you discuss death and dying, I've never had that [agreement from group].’ (FG3A)

Other focus group members countered this by suggesting that the model could be used with dying patients to explore which aspect of the dying process was of concern and what could be done to alleviate the patient's distress.

‘I‘m struggling with terminal patients … saying to them, well have you got any concerns..they‘d say, well I‘m going to die‘ (FG2A)

Perceptions two months post-training All expressed a strong interest in basic communication skills training to build on their existing experience. Immediate perceptions of SAGE & THYME In the focus groups held immediately post-training, DNs (n ¼ 33) were very positive about the model and reported that they were confident about using it in practice. Several benefits of the model were identified. DNs reported that SAGE & THYME ensured that the patient was at the centre of the consultation: it encouraged them to listen to patients, to gather all concerns rather than just the presenting one or two, and to hold back with advice until appropriate. Several DNs commented on the potential for empowering patients through the model: ‘Turning it around..thinking what can they do. For me that's the important thing about it, you allow them to find their own solutions (FG2A)

In the 2 months post training focus groups (n ¼ 26) and individual interviews (n ¼ 5), most DNs reported that they had used SAGE & THYME on several occasions, with only two focus group participants reporting that they had not used it at all. All however reported that their communication skills had improved and that there would have been a noticeable difference had they been observed before and after the training. The two who had not used the model said that this was because a situation had not arisen in which its use would have been appropriate. Interestingly, at the end of the focus group, both nurses stated that the preceding discussion had clarified the scope of the model and situations in which, in retrospect, they could have used it. All who had used SAGE & THYME found it beneficial and none made any recommendations for modification of the structure of the model for the home care setting: ‘I wouldn't change or modify any of that tool because I know it will work’ (Individual Interview 1)

‘You‘re listening not fixing (FG2C) Adaptation of the model They also reported that the model provided a useful structure for consultations, and could potentially save them time. ‘It gives you a structure..leads you through the questions you need to ask: it gives you a closing as well..We‘ve all experienced that..where we just can‘t get out the door’ (FG2A) Whilst most comments about the model were positive, a number of questions were raised about the utility of SAGE & THYME in practice. One of the DNs questioned whether the model was sufficiently advanced because the emphasis was on gathering information and helping patients to find their solutions rather than exploring concerns in any depth:

Although they would not modify the structure of the model, the view of many of the DNs was that the way SAGE & THYME was used in practice could be adapted, using language that DNs were comfortable with. They felt that flexibility was important. The principles however were followed by all i.e. listening to patients, gathering all concerns and keeping advice to the end of the consultation: ‘I've had the card in my pocket, so I've looked at it, but, I've not religiously gone through every stage on it; ‘it's not a natural style, I used my own style to carry on with it’ (FG3 C) ‘I just tend to use the general idea that I'm trying to facilitate them being able to talk, and to come up with their own ideas' (Individual Interview 2)

‘You‘re not really exploring, you‘re just getting the information’ Another of the DNs was concerned about using the model with patients at the end of life, predicting that the patient's main concern

Other adaptations that some of the DNs made were at the gathering stage (‘G’ of SAGE). Some nurses were more comfortable gathering a few concerns, working with those, then asking if there



was anything/something else they were concerned about. Also, not all DNs used note taking when gathering concerns e recommended during the training e because some found it difficult to maintain eye contact: DN1:I would not have felt happy about getting my notebook out and start writing things down … in a family home we need to keep things more relaxed DN2:I tried but couldn't do it DN3: I found it unnatural as well, and I thought I was looking at the patient, but they thought I wasn't listening to them’ (FG3 D)

Experiences of using the model The DNs reported that they were already using some elements of the model, but other parts were new. Asking about concerns (the ‘A’ of SAGE) and empathizing (the ‘E’ of SAGE) were part of current practice, but gathering all concerns (the ‘G’ of SAGE) was new to most of them: ‘I thought I was quite a good listener … now it's made me realise that I was sort of listening, but half sort of planning what the resolutions to the issues would be … I'm trying to hold back on that now and let them finish everything they've got to say’ (Individual Interview 2) The second part of the model (THYME) was mostly new. Asking the patient who they had to talk to, whether it helped and what they thought would help (the ‘T’ and ‘H’ and ‘Y’ of THYME) was new and considered to be very effective and empowering: ‘I would never have asked that question, who else have you got to support you, I've never thought of asking that question before SAGE and THYME, but I always ask that now’ (FG3 D) ‘The question where you say … who have you got to help you’ … That was brilliant..that brought a lot more information’ (FG3 A) Leaving advice to the end of the model (the ‘M’ of THYME) was also considered to be very empowering: ‘It's when they get to that is there something you'd like me to do, that's when you see the anger dissipate, and that's when they seem to calm down’ (FG3 B) When using SAGE & THYME the DNs felt they were able to close conversations more easily (the ‘E’ of THYME) knowing that patients felt listened to. In some instances this was thought to save time: ‘I'm doing the same thing..but I'm getting in and out faster than I've ever done using this … I do find if you rush in and rush out with a patient with the old set up they're upset whatever time you leave them, but with (SAGE & THYME) they always feel like you've helped them and thank you. So it does work’ (FG3 B) ‘ … rounding up allows you a way out, as well, because they've been listened to’ (FG3 C)

Challenges of using the model in practice There were challenges when using the model in practice, for example controlling the setting for the conversation (the ‘S’ of SAGE). The home care environment sometimes presented difficulties:

5

‘I tried using it, the patient had died and the son was there and son's girlfriend and another colleague and I just couldn't do it there … but maybe if I could take him out of the situation’ (FG3 A) ‘Often if you've got family members there … it may be difficult for a patient to open up’ (Individual interview 3) Other nurses were confident about controlling the setting and mentioned fairly minor distractions such as television as the greatest challenge. The home setting was still seen as distinctive but controlling it achievable. Some of the DNs felt that the model was most appropriate for new rather than existing patients, and that the initial assessment provided an ideal opportunity for using SAGE & THYME: ‘With all the new patients..that you don't know as much about..you're trying to get a good background, then it's really useful’ (Individual interview 3) ‘I think they're more open to that probing and that questioning at that first visit..when you're new to them kind of thing’ (FG 3 C) The main reason for preferring to use it with new patients was that the DNs wanted to avoid being seen to change their communication style: If you were to use with existing patients ‘why are you suddenly putting it back to me … it's that thing of them thinking well why is she being different’ (FG3 B) ‘I think the first contact just in terms of how you are with somebody initially begets a pattern doesn't it of how they're going to respond to you’ (FG3 A). However, one DN emphasised that existing patients had not noticed a change in her communication style: ‘I tried it and I was quite surprised that they didn't notice, or certainly haven't told me they've noticed’ (FG3 B) A few of the DNs expressed that elements of SAGE and THYME were less appropriate for patients towards the end of their palliative phase: ‘With the patient that's newly diagnosed it's excellent.. But on the terminal stage or when the patient's died I think you can only use part of it, because obviously they're too ill to sort some of the problems out themselves or the family's too upset … there's still a place for it but only part of it.’ Individual interview 1 An alternative view however was that SAGE & THYME had been useful with a dying patient: ‘She was biting at the bit as I walked in through the door, but when we'd finished she actually seemed, even though she was dying, that I'd empowered her really at the end of life..and she allowed us to go back again’ (FG3 B) Overall therefore, the nurses felt that their communication skills had improved since training, which was supported by the findings from the questionnaires which will be discussed next.


6


Quantitative results This section presents data from two time points to establish whether reported changes in behaviour were maintained over time. Paired results are presented from the three questionnaires completed pre training and at two months follow up. As noted, the first questionnaire asked the nurses to rate their perceptions of outcomes of consultations, the second to rate their confidence in communication skills and the third the usefulness of SAGE & THYME. 27 DNs completed the outcomes questionnaire on both occasions, 25 the confidence questionnaire and 26 the usefulness questionnaire. The paired results of the outcomes questionnaire are presented in Table 1. Statistically significant changes at p < 0.05 were all in the hypothesized direction i.e. towards better perceived outcomes: asking about concerns and emotions benefits patients, they will not get ‘too close’ to their patients nor their work become unmanageable, exploring concerns is helpful and will not distress patients, and the nurse will not become overwhelmed by the patient's emotions. Table 2 presents the paired results of Questionnaire B (Confidence in communication skills). All results were highly significant in the direction of increased confidence in communication skills even though confidence was high at baseline. Table 3 presents the paired results of the motivation questionnaire which was completed immediately post training and at two months follow up. Overall, motivation to use SAGE & THYME and perceptions of its usefulness did not go down significantly from immediately post training to two months later. Motivation to take further communication skills training was high immediately post training and remained so two months post training. There was only one significant finding from this questionnaire which was that the nurses were less inclined to use the entire SAGE & THYME structure when they had been using it in practice for two months, which was also shown in our qualitative findings. Discussion This mixed methods pilot of SAGE & THYME communication skills model with DNs demonstrated its utility in palliative home care, an aspect of district nursing work where distress is common and conversations difficult. SAGE & THYME provided a structure for conversations and facilitated opening and closing of interactions. Knowledge about communication behaviours that are helpful to patients improved and was sustained two months after the training. Increased confidence in communication skills was also sustained. Motivation to use SAGE & THYME was high and remained so at two months, and some said that the model saved them time. Although some of the DNs reported that they had adapted SAGE & THYME to their own style, the main principle of patient centeredness was reportedly used by all i.e. listening more, helping patients to find their own solutions and holding back with premature advice. Challenges with using the model included controlling the home environment and reluctance by some to take notes while gathering concerns. For others, the change in style of communication when using SAGE & THYME was so marked that they preferred to use it with new patients than those already on the caseload. The implications of these findings will now be discussed. There has been a lot of interest and research in training health care professionals in communication skills in recent years, particularly in oncology and palliative care. It is widely acknowledged that communication skills do not simply improve through experience, but need to be taught (Moore et al., 2013). Training courses are broadly similar and typically comprise taught components, group discussion and role play. The main difference between courses seems to be their length (typically two days or more), and

whether they cover more complex skills such as breaking bad news. SAGE & THYME teaches basic/‘foundation level’ communication skills in one three hour workshop which is potentially highly cost effective. If, as our findings suggest, the patient centredness of interactions both improves and is sustained after a three hour workshop, then SAGE & THYME is a promising model. The NHS seems to be similarly confident in the model because they have trained 26,000 health care professionals in its use, although the majority of these are hospital rather than community based. Improving the patient-centredness of interactions has long been recognized as crucial to both assessing patients' needs and providing care (McCormack and McCance, 2006). It is defined as exploring patients' main concerns, seeking complete understanding of their world, finding common ground and mutually agreeing management, enhancing prevention and health promotion and nurturing a continuing relationship with the health care provider (Stewart, 2001). These are important goals of all aspects of district nursing work, not just palliative care, which suggests that training in the model had wider application. This is supported by other evaluations of SAGE & THYME in acute settings (Connolly et al., 2009, 2014). Our findings need to be set in the context of the wider literature on communication skills however. A recent systematic review of the efficacy of communication skills training on staff behaviour in oncology settings (Barth and Lannen, 2011) found that interventions lasting less than three days showed consistently small effects. As our data were reports of behaviour change, rather than observations of practice, it is possible that our findings are overly optimistic. For example, a systematic review of pre-registration training in communication skills (Chant et al., 2002) found that self perceived improvements consistently over estimated actual behaviour change. However Connolly et al.'s recent evaluation of SAGE & THYME with nurses from acute settings (Connolly et al., 2014) found statistically significant increases in the use of helpful communication behaviours when interactions with nine simulated patients were assessed before and immediately after the workshop. Further investigation of behaviour change on a larger scale and over time is needed. Changing practice is recognised to be highly complex (Azjen, 1991) and influenced by a number of factors that need to be addressed for the change to be both initiated and sustained (Rycroft-Malone et al., 2004; May and Finch, 2009). Our preliminary findings on challenges to using SAGE & THYME suggest a number of difficulties with implementing the model at a professional/patient level. Some of these are perhaps easier to overcome than others. Concern about changing the style of communication was a short term difficulty that occurred with patients already known to the DN. If SAGE & THYME becomes imbedded in practice, this should be less of a problem. Reluctance to take notes by some of the DNs is however an issue that needs to be overcome because it is very difficult to remember all patients' concerns if there are many, as is often the case when someone is distressed (Connolly et al., 2009). Mastering this skill while interacting with patients requires practice so that note taking becomes more natural and comfortable. Further rehearsal through role play may help with this. Controlling the home environment is a more complex issue however as we have reported elsewhere (Wilson et al., 2014). The physical context of the home can be difficult to control and is compounded by family dynamics which cannot be controlled. Strategies for dealing with this unique environment need to be developed and built into the training of community nurses in SAGE & THYME. There may be other challenges with using SAGE & THYME that were not captured by this study. These might also be at an



individual level, but could be at other levels within the organization. For example, the literature suggests that both immediate colleagues and senior leaders/managers need to be supportive of the change if it is to be sustained, and the change needs to fit with the organisation's strategic aims for it to receive support at the highest level (NHS Institute for Innovation and Improvement, 2010). It is also recognized that there needs to be an infrastructure for sustainability, which may be difficult in these straitened times. Further exploration of the full range of potential challenges is needed, including how these can be overcome. Limitations of current communication skills research in oncology are that while many studies demonstrate that health care professionals' behaviour seems to improve with training, there is almost no evidence of improvement in patient outcomes (Moore et al., 2013). This is mostly because research hasn't been conducted, which is perhaps because designing trials in this field is notoriously difficult. Interactions are unpredictable and outcomes hard to measure. A powered trial of SAGE & THYME is arguably necessary to evaluate its impact on patient outcomes, or it may be difficult to justify continuing to train health care professionals. The trial may need to be in an area of oncology where interactions are more predictable, and outcomes easier to measure. This might be in specialist clinic settings rather than palliative home care. Limitations The qualitative component of this study was inevitably small scale, and focus group methods can be criticized for the dominance of some voices over others, with potential misrepresentation of the views of the group. The facilitator attempted to overcome this by encouraging participation from all of the group members, and checking out responses from more vocal participants with the rest of the group. The quantitative study was also small scale (25e27 questionnaires completed on both occasions). It was however sufficiently large to allow some exploratory statistical testing. Conclusion SAGE & THYME appears to aid patient-centred communication in palliative home care by encouraging DNs to hold back with offering pre-mature advice which can close down conversations, and allowing patients to find their own solutions. It improves DNs' confidence in their communication skills and their knowledge of helpful communication behaviours. Training DNs in SAGE & THYME in a three hour workshop appears to be a promising model for improving communication skills when working with cancer patients. Further research is however needed to explore in a wider population and over time, whether actual behaviour changes and is sustained, barriers and facilitators to this and ultimately, the benefits to patients of health care professionals using SAGE & THYME in their interactions. Conflict of interest statement We have no conflict of interest. Acknowledgements We would like to thank Dimbleby Cancer Care for funding this project, our steering group of clinicians and service users for their valuable advice throughout, and the District Nurses who took part in the project and generously gave their time.

7

References Addington-Hall, Shipman, C., Burt, J., Ream, E., Beynon, T., Richardson, A., 2006. Evaluation of the Education and Support Programme for District and Community Nurses in the Principles and Practice of Palliative Care. DOH, London. Aitken, A., 2006. District nurses' triggers for referral of patients to the Macmillan Nurse. British Journal of Community Nursing 11 (3), 100e136. Austin, L., Luker, K., Caress, A., Hallett, C., 2000. Palliative care: community nurses perceptions of quality. Quality in Health Care 9, 151e158. Azjen, I., 1991. The theory of planned behaviour. Organizational Behavior and Human Decision Processes 50, 179e211. Bandura, A., 1982. Self-efficacy theory in human agency. American Psychologist 37 (2), 122e147. Barth, J., Lannen, P., 2011. Efficacy of communication skills training courses in oncology: a systematic review and meta-analysis. Annals of Oncology 22 (5), 1030e1040. Chant, S., Jenkinson, T., Randle, Russell, 2002. Communication skills: some problems in education and practice. Journal of Clinical Nursing 11 (1), 12e21. Coffey, S., 2006. The nurse-patient relationship in cancer care as a shared covenant. A concept analysis. Advances in Nursing Science 29 (4), 308e323. Connolly, M., Perryman, J., McKenna, Y., Orford, J., Thomson, L., Shuttleworth, J., et al., 2009. SAGE & THYME™: a model for training health and social care professionals in patient-focussed support. Patient Education and Counseling 79, 87e93. Connolly, M., Thomas, J., Orford, J., Schofield, N., Whiteside, S., Morris, J., et al., 2014. The impact of the SAGE & THYME foundation level workshop on factors influencing communication skills in health care professionals. Journal of Continuing Education in Health Professions 34 (1), 37e46. Dunne, K., Sullivan, K., Kernohan, G., 2005. Palliative care for patients with cancer: district nurses' experiences. Journal of Advanced Nursing 50 (40), 372e380. Fingfeld-Connett, D., 2007. Meta-synthesis of caring in nursing. Journal of Clinical Nursing 17 (2), 196e204. Griffiths, J., Ewing, G., Rogers, M., 2007. Supporting cancer patients with palliative care needs. District Nurses' role perceptions. Cancer Nursing 30 (3), 156e162. Griffiths, J., Ewing, G., Rogers, M., 2010. “Moving Swiftly On.” psychological support provided by district nurses to patients with palliative care needs. Cancer Nursing 33 (5), 390e397. Griffiths, J., Ewing, G., Rogers, M., 2012. Early support visits by district nurses to cancer patients at home: a multi-perspective qualitative study. Palliative Medicine 27 (4), 349e357. Griffiths, J., Ewing, G., Wilson, C., Connolly, M., Grande, G., 2014. Breaking bad news about transitions to dying: a qualitative exploration of the role of the District Nurse. Palliative Medicine. http://dx.doi.org/10.1177/0269216314551813. Hotopf, M., Chidgey, J., Addington-Hall, J., Lan Ly, K., 2002. Depression in advanced disease: a systematic review part 1. Prevalence and case finding. Palliative Medicine 16, 81e97. Kaufman, J.E., 1996. Personal definitions of health among elderly people: a link to effective health promotion. Family and Community Health 19 (2), 58. Kennedy, C., 2005. District nursing support for patients with cancer requiring palliative care. British Journal of Community Nursing 10 (12), 566e574. Kitzinger, J., 1995. .The methodology of focus groups: the importance of interactions between research participants. Sociology of Health and Illness 16, 103e121. Kreuger, R.A., 2008. Focus Groups: a Practical Guide for Applied Research, third ed. Sage, London. Luker, K., Austin, L., Caress, A., Hallett, C., 2000. The importance of “knowing the patient”: community nurses constructions of quality in providing palliative care. Journal of Advanced Nursing 31 (4), 775e782. McCormack, B., McCance, T., 2006. Development of a framework for person-centred nursing. Journal of Advanced Nursing 56 (5), 472e479. Maguire, P., Faulkner, A., Booth, K., Elliott, C., Hillier, V., 1996. Helping cancer patients to disclose their concerns. European Journal of Cancer 32a, 78e81. May, C., Finch, T., 2009. Implementing, embedding, and integrating practices: an outline of normalization process, theory. Sociology 43, 535e554. Mok, E., Chui, P.C., 2004. Nurse-patient relationships in palliative care. Journal of Advanced Nursing 48 (5), 475e483. Moore, P., Mercardo, S., Attigues, M., Lawrie, T., 2013. Communication skills training for healthcare professionals working with people who have cancer. Cochrane Database of Systematic Reviews. http://dx.doi.org/10.1002/4651858.CD0037 51.pub3. Issue 3, Art No.: CD003751. Morse, J.M., 1992. Negotiating commitment and involvement in the nurse-patient relationship. In: Morse, J.M. (Ed.), Qualitative Health Research. Sage, London, pp. 333e359. NHS Institute for Innovation and Improvement, 2010. Sustainability Model. Available at: http://www.institute.nhs.uk/sustainability_model/general/welcome_ to_sustainability.html (accessed 03.09.14.). Ohman, M., Sodreberg, S., 2004. District nursing: sharing an understanding of being present. Experiences of encounters with people with serious chronic illness and their close relatives in their homes. Journal of Clinical Nursing 13, 858e866. Pateman, B., 2003. Continuing care after cancer treatment. Journal of Advanced Nursing 44 (2), 192e199.


8


Rycroft-Malone, J., Harvey, G., Seers, K., Kitson, A., McCormack, B., Titchen, A., 2004. An exploration of the factors that influence the implementation of evidence into Practice. Journal of Clinical Nursing 13, 913e924. Seale, C., 1992. Community nurses and care of the dying. Social Science and Medicine 34 (4), 375e382. Stewart, M., 2001. Towards a global definition of patient centred care. British Medical Journal 322, 444e445. Stolz, P., Lindholm, M., Uden, G., Willman, A., 2006. The meaning of being supportive for family care givers as narrated by registered nurses working in palliative home care. Nursing Science Quarterly 19 (2), 163e173. Strauss, A., Corbin, J., 1998. Basics of Qualitative Research: Techniques and Procedures for Developing Grounded Theory. Sage, London.

Thekkumpurath, P., Venkateswaran, C., Kumar, M., Bennett, M., 2008. Screening for psychological distress in palliative care: a systematic review. Journal of Pain and Symptom Management 36 (5), 520e528. White, C., Mcleod, U., 2002. ABC of psychological medicine: cancer. British Medical Journal 325, 377e380. NICE 2004. Wilson, C., Griffiths, J., Ewing, G., Connolly, M., Grande, G., 2014. A qualitative exploration of district nurses' care of patients with advanced Cancer: the challenges of supporting families within the home. Cancer Nursing 37 (4), 310e315. Wright, K., 2002. Caring for the terminally ill: the district nurse perspective. British Journal of Nursing 11 (18), 1180e1185.


The impact of the SAGE & THYME foundation level workshop on factors influencing communication skills in health care professionals.

The Critical Care Communication project: improving fellows' communication skills.

Palliative care communication.

Improving communication skills in the Southeast Asian health care context.

Improving communication in cancer pain management nursing: a randomized controlled study assessing the efficacy of a communication skills training program.

Letters to patients: improving communication in cancer care.

Improving medical communication with patients and families: Skills for a complex (and multilingual) clinical world.

Is home-based palliative care cost-effective? An economic evaluation of the Palliative Care Extended Packages at Home (PEACH) pilot.

Improving communication with multiple sclerosis patients.

Improving Palliative Cancer Care.

Communication and Quality of Care on Palliative Care Units: A Qualitative Study.

Improving communication on hope in palliative care. A qualitative study of palliative care professionals' metaphors of hope: grip, source, tune, and vision.

A Curriculum to Improve Residents' End-of-Life Communication and Pain Management Skills During Pediatrics Intensive Care Rotation: Pilot Study.

Assessment of dental students' communication skills with patients.

Teaching communication skills.

Improving health care communication for persons with mental retardation.

Communication skills and anesthesiologists.

Training students with patient actors improves communication: a pilot study.

P076. Improving patient communication and management by the use of the "Headache Digest", a pilot study.

The Use of Standardized Patients to Evaluate Interpersonal and Communication Skills of Anesthesiology Residents: A Pilot Study.

Intercultural doctor-patient communication in daily outpatient care: relevant communication skills.

Design of, and enrollment in, the palliative care communication research initiative: a direct-observation cohort study.

Improving clinician-carer communication for safer hospital care: a study of the 'TOP 5' strategy in patients with dementia.

Pediatric palliative care: using miniature chairs to facilitate communication.