‘It’s tough. It is hard’: A qualitative interview study of staff and volunteers caring for hospice in-patients with delirium

Background: Delirium is a distressing condition often experienced by hospice in-patients. Increased understanding of current multidisciplinary care of delirium is needed to develop interventions in this setting. Aim(s): To explore hospice staff and volunteers’ practice, its influences and what may need to change to improve hospice delirium care. Design: Qualitative interview study using behaviour change theory from a critical realist stance. Setting/participants: Thirty-seven staff, including different professional groups and roles, and volunteers were purposively sampled from two in-patient hospices. Results: We found that participants’ practice focus was on managing hyperactive symptoms of delirium, through medication use and non-pharmacological strategies. Delirium prevention, early recognition and hypoactive delirium received less attention. Our theoretically-informed analysis identified this focus was influenced by staff and volunteers’ emotional responses to the distress associated with hyperactive symptoms of delirium as well as understanding of delirium prevention, recognition and care, which varied between staff groups. Non-pharmacological delirium management was supported by adequate staffing levels, supportive team working and a culture of person-centred and family-centred care, although behaviours that disrupted the calm hospice environment challenged this. Conclusions: Our findings can inform hospice-tailored behaviour change interventions that develop a shared team understanding and engage staff’s emotional responses to improve delirium care. Reflective learning opportunities are needed that increase understanding of the potential to reduce patient distress through prevention and early recognition of delirium, as well as person-centred management. Organisational support for adequate, flexible staffing levels and supportive team working is required to support person-centred delirium care.


What is already known about the topic?
• • Delirium is a distressing condition often experienced by hospice in-patients.
• • Tailored evidence-based interventions for delirium in hospices need to be developed.

What this paper adds
• • Hospice staff and volunteers' practice predominantly focused on managing hyperactive symptoms of delirium, rather than delirium prevention or early recognition. • • Our theoretically informed analysis identified this focus was influenced by staff and volunteers' emotional responses to the distress associated with hyperactive symptoms of delirium underpinned by the lack of a shared team understanding of delirium care. • • Non-pharmacological delirium management was supported by adequate staffing levels, supportive team working and a culture of person-centred care, although behaviours that disrupted the calm hospice environment challenged this.

Implications for practice, theory or policy
• • Organisational support for adequate, flexible staffing levels and supportive team working is required to support personcentred delirium care.

Introduction
Delirium is a distressing condition often experienced by hospice in-patients. 1,2 It is characterised by acute, fluctuating disturbances in attention, awareness and cognition, stemming from underlying physiological causes. 3 Palliative care patients most commonly experience the under-recognised hypoactive subtype, and report as much distress as those with hyperactive symptoms. 1,[4][5][6][7] There is little research into how to prevent and manage delirium in hospice in-patients. [8][9][10][11][12] In the UK context, in contrast to other countries, 13,14 hospices are independent, charitably funded organisations which provide inpatient symptom management and end-of-life care, amongst other palliative care services. In hospital settings multicomponent interventions can reduce delirium incidence by a third. 15 Clinical guidelines (not palliative carespecific) recommend delirium screening, interventions targetting delirium risk factors, assessment and treatment of underlying causes, non-pharmacological strategies to support patients and family involvement in care. 16,17 Systematic reviews demonstrate little high quality evidence supporting routine use of medication for delirium. 18,19 Evidence and guidelines from other settings are useful, but differences in hospice patients, organisation and culture need to be taken into account.
Qualitative studies provide useful insights into how delirium care practice in palliative care settings aligns with, or differs from, evidence and guidelines, and the influences on this. Our qualitative review found most research explored nurses' perspectives and focused on delirium management rather than prevention or early identification. 2 Interview studies with nurses identified they lacked knowledge and skills in delirium recognition, assessment and management. [20][21][22] Use of medication was triggered by patients', families' and clinicians' distress, and time and staffing pressures. 20,[23][24][25] Personcentred approaches and family involvement were important enablers for good delirium care. [22][23][24][26][27][28][29][30][31] This study addresses UK in-patient hospice care. A more comprehensive understanding is needed of the delirium practice behaviours of multidisciplinary hospice teams operating within organisational cultures distinct from the NHS and other palliative care contexts. Structured analysis of the influences on these behaviours is necessary to inform the development of tailored interventions to improve delirium care in this setting. Our study objectives were to explore:

Methods
We report this study according to the Standards for Reporting Qualitative Research (SRQR). 32

Design
This qualitative interview study is part of a mixed-methods research programme to design an intervention to improve hospice delirium practice. 33,34 We used thematic analysis, informed by behaviour change theory; the COM-B (Capability, Opportunity, Motivation-Behaviour) framework and Behaviour Change Wheel. 35,36 COM-B theorises that Capability (physical and psychological), Opportunity (physical and social) and Motivation (automatic and reflective) interact to generate behaviour (See Table 1). For the purpose of the study this enabled structured analysis of the influences on staff and volunteers' delirium care behaviours to target for change, and techniques useful for interventions. We used a critical realist approach, based on the assumption of a shared reality, our understanding of which is mediated by our cultural contexts. 37

Setting
Two independent third-sector UK hospices with in-patient units (Hospice 1 = 21 beds; Hospice 2 = 18 beds) were included. Neither had a delirium guideline, delirium screening tool or patient/family information in use at the time of data collection.

Participants
Eligible participants were consenting hospice staff members (doctors, nurses, allied health professionals, health care assistants, domestic workers, managers, fund-raising staff, board members) and volunteers. There were no specific exclusion criteria.

Sampling and recruitment
Participants were purposively sampled to include the perspectives of these different groups. Participants with nonpatient facing roles were included to gain understanding of organisational level factors, hospice culture, staff support systems and resources.
Potential participants were informed about the study through emails, posters and researchers (IF, LJ) attending staff meetings. Those interested contacted the research team directly.
'Information power' was evaluated during the study to inform sample size, guided by the criteria: breadth of study aim; sample specificity; use of theory; quality of interview dialogue; analysis strategy. 39 This approach is congruent with our use of theoretically-informed thematic analysis.

Data collection
Semi-structured individual interviews were conducted inperson and by telephone (IF, LJ, October 2017-April 2018) using an interview guide and case vignettes of patients with hypoactive and hyperactive delirium developed by the research team (IF, LJ, NS, MJ, AH) with public involvement group input (Supplemental File 1). Pilot interviews were conducted at another hospice site. Interviews were audio-recorded and transcribed verbatim.
Quantitative demographic information was collected on participants' age, sex, role and years of palliative care experience.

Data analysis
Thematic analysis was used to generate, analyse and interpret themes from the interview data, 36 supported by Nvivo software. 40 We used elements of both 'theorydriven' and 'data-driven' approaches. 36,41 Coding was informed by delirium guidelines but a pre-defined coding framework was not used. The development of descriptive themes was based upon participants' experiences of delirium care, and the COM-B framework then used to support development of analytical themes.
Two researchers (IF, EC) initially carried out independent line by line coding of the same interviews, before developing a shared coding scheme. They then coded separate interviews, regularly reviewing the developing analysis together with a third reviewer, LJ. Descriptive themes and subthemes were developed by sorting and collating codes, reviewing their content and exploring relationships between them using memos and mindmaps.
COM-B was then used to support structured analysis of the influences on staff and volunteers' delirium care behaviours, strengths and what may need to change (written by IF, reviewed by NS, MJ, TS, LJ and the public involvement group). 35 Trustworthiness and credibility was enhanced by the use of 'constant comparison' and the involvement of several researchers. 42

Findings
Participants: Thirty-seven participants were interviewed. Mean interview duration was 35 min (Range 15-63 min). See Table 2 for participant characteristics.
Thematic analysis: An over-arching theme was, 'A reactive approach to hyperactive symptoms of delirium'. This expresses how participants' practice focused on managing hyperactive symptoms. Other aspects of care, particularly prevention, early recognition and hypoactive delirium, received much less attention. We used COM-B to explore the influences driving this reactive approach to delirium care, as well as the influences on how hyperactive symptoms were managed. (See Table 3). Due to the breadth of scope of the COM-B concepts, all of our findings fitted with this theoretical framework. Many participants described how distressing the changes in patient's behaviour can be for family members, Because things aren't as normal and they're not as used to. . . dad turning round and f'ing and blinding and, and taking all his clothes off. . .and it's very, very distressing for. . . relatives. (P06, Nurse) Their empathy for the patient and family's distress could be a powerful motivation to act, Obviously that person, that family, are your absolute focus. . . I think to relate to a situation and put yourself in that poor wife's situation is the way that you, you're really fighting for them. It sounds quite dramatic (laughs) but it is at the time. (P22, Nurse) b. Stress in managing the patient's delirium and its impact on others Participants, particularly those who spent most time with patients, described experiencing intense stress due to  Social/physical opportunity it is a severe hyperactive delirium, and are trying to climb out of bed, they're pulling their catheters and, yeah, fighting and whatever, and, yes, they expect you to do something to remove that as quickly as possible. (P11, Doctor) I think they really struggle to see somebody restless and, and agitated and immediately want to settle that down, usually through giving medication. (P03, Doctor) In contrast, patients with hypoactive delirium were likely not to be recognised by clinical staff, I think that we're bad at picking up the hypoactive ones; so, when people are actually quite quiet we probably don't pick up those things, we might be asking them how they're feeling but, as I say, I think they're the ones that tend to get missed. (P15, Doctor) Patients with hypoactive symptoms did not disrupt the calm environment or provoke emotional urgency to act, The hypoactive patients, there may be people who are sitting very quietly and placidly and not 'causing any trouble' and therefore they don't necessarily get the same degree of attention as somebody who is, you know, being very loud and confused and wandering. (P20, Doctor) There's less pressure. . .she's quite quiet, isn't she, so she's not causing a problem; and that's where I think it's not always acted on. (P25, Doctor) In terms of COM-B, participants' emotional responses, part of 'automatic motivation', powerfully influenced their focus on managing hyperactive symptoms, including medication use. This was compounded by a cultural necessity to maintain a calm environment, an influence related to 'social opportunity'.

Varied understanding of delirium influences the reactive focus on managing hyperactive symptoms (Psychological capability/reflective motivation)
When staff and volunteers responded to patients' distressing behaviours, many were not acting in the context of an understanding of delirium as a medical condition with a structured process of care. Understanding of delirium recognition, prevention and management, varied between individuals and staff groups. This strongly influenced both the reactive focus on managing hyperactive symptoms, and the management itself. Some nurses and therapy staff expressed limited knowledge of ways of preventing delirium, but these were not systematically implemented. Some doctors clearly articulated a range of strategies but it is likely that HCAs and nurses would primarily be responsible for many of these tasks. It was highlighted that some of these tasks may be done during usual care, without awareness of their relevance for delirium prevention, Specific prevention, I don't think we're particularly aware of. I think obviously in the hospice it's easier to make sure patients are getting plenty of drinks, we review patients' medications most days; so, all those side of things I think we're good at. So, I think probably without being aware of it we're preventing it but we're just not saying we're preventing it. (P03, Doctor).
The lack of a shared team understanding of delirium as a medical condition, with hypoactive as well as hyperactive symptoms, which may be prevented, contributed to staff's reactive focus on hyperactive symptom management.
The influence of participants' understanding on how they managed delirium was also explored.

c. Varied understanding of modifiable causes of delirium
The extent of clinicians' understanding of causes of delirium, and their modification potential, varied. This influenced their management; most HCAs and volunteers, and some nurses and therapy staff expressed understanding of limited possible causes of 'confusion', including medication and infections. Some nurses and therapy staff described a range of modifiable causes, and the doctors articulated more detailed knowledge of predisposing and precipitating factors, A shared team understanding would support a structured approach to delirium care, including prevention and early recognition, as well as management.

Hospice culture and environment influence person-centred delirium care (Social and Physical opportunity)
Person-centred and family-centred approaches to delirium care were both supported and challenged by hospice cultural norms (social opportunity) and staffing levels (physical opportunity).
a. 'Patients and their families are at the centre': Person-centred and family-centred approach to delirium care (Social opportunity) Participants involved in direct care reported making efforts to engage with, and calm, patients who were confused or agitated through learning about their lives and interests and person-centred activities, Participants providing direct care described a range of reassuring interpersonal strategies including a calm tone of voice, non-threatening body language and sensitive use of touch. They encouraged families to bring in familiar items to help orientate and reassure patients, . . to bring in some photographs and put them on a memory board and make him familiar that this is his family. . . and then I suggested that his family maybe do a tape of their voices so that on a night. . . it was a ritual where he would get in his bed and. . . he was in and out, in and out, he couldn't settle, and then once we put the headphones on and. . . his tape and the voices of his family, he seemed to settle and relax. (P26, HCA) Many participants identified that the presence of a family member could help calm patients, although they recognised this could be stressful for the family, We. . .appreciate that (family) carers, they use this as their respite as well when the patients are in, but particularly if people know they're unsettled overnight sometimes they do stay if they know that makes them more feel at home and more comfortable and less agitated. However, participants identified challenges in providing person-centred care with people with hyperactive delirium, including staff stress (theme 1) and the intensive staffing levels required (theme 3). When patient behaviours conflicted with needs of other patients and their families, the needs of others were sometimes prioritised. We

Main findings of study
Our study found that staff and volunteers focused on managing hyperactive symptoms. Delirium prevention, early recognition and hypoactive delirium, received much less attention. A powerful motivating influence for this reactive approach, including the use of medication, was participants' emotional responses (automatic motivation) in managing the impact of hyperactive symptoms on patients and their families. Knowledge and understanding of delirium care, along with training opportunities (psychological capability), varied between staff groups. Under-recognition of delirium as a medical condition, and poor knowledge of the potential for delirium prevention, strongly influenced this reactive focus. When patients had delirium close to death, some clinicians assumed only symptom control was appropriate, while others maintained active clinical reasoning (reflective motivation) regarding the benefits and burdens of investigating modifiable causes, for example, treatment of constipation or medication review.
Staff and volunteers' management of hyperactive symptoms was supported by the cultural norm of valuing person and family-centred approaches. However, hyperactive symptoms which disrupted the norm of a calm hospice environment, challenged person-centred care (social opportunity). Adequate time and staffing levels (physical opportunity) and supportive team working (social opportunity) supported this approach.

Strengths and limitations
Strengths of this study are our diverse participants; robust data collection and analysis. Our use of COM-B enabled us to carry out structured analysis of the influences on delirium practice which can inform the design of behaviour change interventions. 35 Despite including only two UK hospices, transferability is increased as our findings can be compared with those of other delirium studies using the COM-B framework. 43 A limitation of using qualitative interviews was that participants may not have identified preventative behaviours carried out during usual care, as they were unaware of their relevance to delirium.

What this study adds
Our findings suggest that to improve delirium care in inpatient hospices, a shift 'upstream' in practice is needed, with increased focus on delirium prevention and early recognition. There has been only preliminary research into delirium prevention in palliative care settings. 8,9 A pilot study, reporting low adherence to a delirium prevention intervention, found that clinicians were more strongly motivated to enact responsive care than anticipatory. 44 Our theoretically informed analysis of the influences on this 'reactive approach' can be used to inform further research to develop effective interventions.
Our findings regarding the influence of psychological capability, align with and build upon previous interview studies with palliative care nurses which found limited understanding of delirium recognition, assessment and management. [20][21][22] Hosie et al. 21 found variation in nurses' use of delirium terminology, capability to frame symptoms as delirium and conduct comprehensive assessment. Importantly, we also identified a need for increased understanding of the potential for delirium prevention, particularly for staff most likely to carry out the relevant strategies.
Due to our range of participants, our study highlighted the variation in knowledge between different staff and volunteer roles, with doctors having more learning opportunities and understanding than others involved in direct patient care. A previous systematic review recommended inter-professional education for complex delirium care. 44 There is a need to develop a shared team understanding to enable a structured approach to delirium care; reflected by recent calls for effective and strong interdisciplinary collaboration in this field. 45 Our finding that staff's emotional responses (automatic motivation) are important in driving their focus on managing hyperactive symptoms, including the use of medication, is consistent with the results of an Australian survey study, which found that 82% of palliative care clinicians reported emotional influences, including patient, family and staff distress, on their delirium treatment behaviours. 43 Delirium educational approaches need to address the emotional impact of delirium, as well as clinical learning or more technical needs. 46 Behaviour change theory can inform intervention design to both increase delirium knowledge and engage staff's emotional responses to motivate practice change.
Behaviour change techniques could be used to increase staff and volunteers' understanding of the potential health and emotional consequences of increased practice focus on delirium prevention and early recognition: reducing delirium incidence, severity and distress. Other emotionally engaging behaviour change techniques could include: the use of 'credible sources' to deliver delirium education (e.g. senior staff; external experts; people with lived experience), and 'modelling', such as reflective case study-based learning. 38 As guidelines primarily recommend non-pharmacological management, 16,17 our findings regarding supportive social and physical opportunities are important to inform clinical practice and service provision. Enablers included a cultural norm of valuing person-centred and family-centred care throughout the organisation. However, this was challenged when clinical decisionmaking, including medication use, was influenced by the disruption caused to others by hyperactive symptoms. In the UK context, the high value placed upon maintaining the cultural norm and perception of hospices as calm and welcoming, may also partly be influenced by their reliance on charitable fundraising. Our previous qualitative literature synthesis found that, although some patients and families may wish for sedation to be used to achieve comfort and a 'peaceful' death, others prefer to be able to communicate, despite the delirium. 2 Further, sedation may mask hyperactive symptoms and 'convert' patients to hypoactive delirium, 47 experiencing as much distress as those with hyperactivity. 4,7 Interventions combining education on medication use with reflective learning on complex treatment decision-making, including whose needs are primarily being addressed, would build upon the person-centred approach to delirium care.
The delirium prevention pilot study in palliative care units, found that a culture of compassionate care and interdisciplinary collaboration supported the intervention. 44 We found that supportive team working and organisational support were important and may need further strengthening. Lack of time or manpower inhibits the use of non-pharmacological strategies in favour of medication use. 25 We found favourable staffing levels compared to hospital settings but an increase was needed with highly dependent patients, especially at night-time. The cost implications of this may be challenging in the context of the cost-of-living crisis and COVID-related financial pressures on hospices. Harnessing the skills of suitably trained and supported volunteers may become increasingly important.
In conclusion, our theoretically informed study identified influences on delirium practice behaviours in hospices, including strengths and areas for change. Further research, using ethnographic observation methods, would be valuable to increase understanding of practice relevant to delirium prevention during usual care. Our study findings can be used to inform clinical practice and research to develop behaviour change interventions, tailored to hospice settings, which aim to improve delirium care and reduce the distress that it causes.