Hospital transfers from care homes: conceptualising staff decision-making as a form of risk work

Abstract

When making decisions about whether to transfer residents to hospital, care home staff consider the possible benefits and risks of different courses of action. However, to date, an in-depth and theoretically informed engagement with these decision-making processes and their associated behaviours has been lacking. We conducted an ethnographic study of care home staff’s decision-making about resident hospital transfers in England between May 2018 and November 2019. We combined staff interviews at six care home sites, with 30 members of staff, with 113 hours of ethnographic observation at three care homes sites. ‘Risk’ and risk management emerged as important overarching themes. In this article we conceptualise staff decision-making about potential hospital transfers for residents as a form of risk work. In doing so, we identify the different forms of risk knowledge that staff used to conceptualise risk and explore the ways staff navigated tensions between different forms of risk knowledge. We highlight the ways individual understandings of risk were influenced by social interactions with others, both at an interpersonal and organisational level, before identifying strategies that staff use to manage risk. By understanding transfer decisions explicitly in terms of the different forms of risk that care home staff manage, our analysis provides new insights into hospital transfers from care homes and contributes to the wider literature around risk work, demonstrating the utility of this concept in researching organisations that fall under the umbrella of social care, which have been previously neglected in academic research.

Keywords:

• Risk perception
• care home
• transfer
• hospital
• risk work

Introduction

In England, there are approximately 15,000 care homes, providing care for approximately 400,000 people. The main regulator of care homes in England, the Care Quality Commission (CQC), differentiates between ‘care homes with nursing services’ (which account for approximately one third of all care homes in England) and ‘care homes without nursing services’ (which account for the remaining two thirds). A small number of homes are ‘dual registered’ to provide support to both residents who require nursing services and residents who do not. Throughout this article, the term ‘care home’ is used to encompass all three types of facility.

People who live in care homes often have complex health and social care needs (Gordon et al., 2014). Although care home residents are a diverse group of individuals (British Geriatrics Society, 2011), they are usually the ‘oldest old’ in society, with over half of all residents aged 85 years and above (Smith et al., 2015). Care home residents are more likely to be transferred to and admitted to hospital than older people living in their own homes (Smith et al., 2015). However, for some care home residents – particularly those who are living with significant frailty and/or cognitive impairment – hospital transfers and admissions may be associated with a subsequent decline in physical health, cognition and psychological well-being (Ashcraft & Owen, 2014; Calnan et al., 2013; Fogg et al., 2018). Furthermore, if admitted to hospital, care home residents face a higher risk of inpatient mortality than older people admitted to hospital from their own homes (Dwyer et al., 2014).

Care home resident transfers to hospital have received increasing research attention. Researchers have sought to identify clinical conditions and symptoms that precede a hospital transfer (Ashcraft & Owen, 2014; Smith et al., 2015), and examined interventions to reduce hospital transfers from care homes (Graverholt et al., 2014; Steventon et al., 2018). Others have explored the ‘patient’ factors (related to individual residents) and ‘organisational’ factors (related to features of care homes and healthcare delivery) which influence rates of emergency care use (Graverholt et al., 2013). Some scholars have sought to identify emergency care use which could be labelled ‘inappropriate’ or ‘avoidable’ (Parkinson et al., 2021; Steventon et al., 2018). Such labels are increasingly seen as problematic, due the lack of consensus about what constitutes an (in)appropriate transfer, both across academic research (Lemoyne et al., 2019) and amongst healthcare professionals that support care home residents (Harrison et al., 2016), and due to the way the use of such labels may lead to care home residents being portrayed as a ‘problem’ for healthcare services. However, they reflect an understanding that hospitalisation carries risks (including increased treatment burden, poor experience and outcomes of hospitalisation). Despite risk being frequently mentioned in academic and policy discourses, research that explicitly draws on sociological theories of risk is lacking. In this article we draw on the literature around risk work in order to further explicate and understand staff decision-making about potential hospital transfers for residents.

Managing risk as ‘risk work’

Horlick-Jones (2005) introduced the term ‘risk work’ to refer to the everyday practices that individuals use to manage risk. Through the use of direct observation, Horlick-Jones (2005) explored the ways risks were (de)prioritised in particular social contexts. Gale et al. (2016) further developed the concept of risk work, to foreground the practices that healthcare workers in patient-facing roles use to assess and manage risk. For Gale and colleagues, risk work is comprised of three interconnected components: ‘risk knowledge’ concerns the ways individuals assess and conceptualise risk; ‘risk interventions’ are the practices individuals use to mitigate against and manage risk; and ‘social relations’ are the ways in which risk practices are influenced by, and embedded within, interpersonal relationships (Brown & Gale, 2018a; 2018b; Gale et al., 2016). At the centre of their conceptualisation, borne out of each of the three components, is the worker’s ‘lifeworld’: this represents the ‘common sense’ and often ‘taken for granted’ social world as experienced by the individual.

Brown and Gale (2018a; 2018b) called for a better understanding of the lived experiences of healthcare workers undertaking risk work in client-facing roles, who apply population-level risk knowledge to individual people. Doing so creates a paradox for workers: Often whilst it is possible to determine the prevalence of risk across a population, it is more difficult to ascertain the likelihood of a particular occurrence for an individual. This notion has applicability to hospital transfers from care homes. Care home staff often consider the potential benefits and risks when deciding whether to initiate a resident hospital transfer (Laging et al., 2015; Trahan et al., 2016). However, although the likely biomedical outcomes of some procedures are understood (to some extent) at a population level, these are more difficult to predict in individual residents. The trajectory of (often multiple and interacting) long term conditions may limit the potential benefits of hospital-based treatment; and the person’s own values, preferences and priorities fundamentally shape what can be considered a ‘benefit’.

Despite the emphasis on social context in many sociological theories of risk, existing research tends to focus on individual understandings of risk. However, a growing body of research has begun to illustrate the ways that individual understandings of risk may be shaped by social interactions: Seppola-Edvardsen et al. (2016) described the ways in which people living with cancer developed their understanding of risk through interactions with others; and Rodrigues (2016) suggested that risk related to everyday medication management was collectively performed through a process of negotiation with others. As a result Brown (2015; 2016) has called for shift towards a greater emphasis on the social relations that shape individual understandings of risk.

When deciding whether or not to initiate a hospital transfer for a resident, care home staff often liaise with others including residents and their families,, care home colleagues and external healthcare professionals. Therefore, understanding decision-making as a form of risk work may be particularly useful, providing a lens to explore the broader social relations that influence individual staff members’ conceptualisations of risk. Therefore, in the analysis presented below we draw on the concept of ‘risk work’ to further explicate and understand staff decision-making about potential resident hospital transfers.

Methodology

In this article we use data collected as part of a broader research study to identify factors which influence care home staff decision-making about potential hospital transfers and develop a ‘model of escalation’ to describe the actions undertaken within care homes when a resident’s health deteriorates (Harrad, 2021; Harrad-Hyde et al., 2022).

Our study was conducted under the philosophical assumptions of critical realism, which is well-suited to examine complex phenomena (Danermark et al., 2002). In line with the critical realism paradigm, which combines a realist ontological belief with an interpretivist epistemology and suggests that all knowledge is socially constructed (Rosenberg, 2012), interactive research methods were chosen.

Phase 1: interviews with care home staff

We conducted face-to-face, semi-structured interviews with care home staff between May 2018 and February 2019. The interview schedule included questions about personal experiences of being involved in hospital transfers and a number of vignettes, which reflected situations that could occur in care homes. Both the interview schedule and the accompanying vignettes were developed based on a review of existing literature and through engagement with four care home managers, and piloted prior to commencement of data collection. Data collected during piloting were not included in the analysis.

Care home sites were sampled purposively to reflect factors that influence transfer rates. A summary of each home’s key characteristics and the number of participants recruited there is provided in Table 1. Individual participants were also sampled purposively to ensure a range of staff voices were heard. Participants included care home managers, deputy managers, registered nurses, senior care workers, and care workers. Each interview was audio recorded and transcribed verbatim. All identifiable data were removed, and each transcript was allocated a transcript number. In total, we conducted 28 interviews with 30 members of care home staff (four participants preferred to be interviewed as two pairs) across six care homes in the East and West Midlands of England. Each care home was provided with a £50 multistore gift voucher as a thank you for their participation. Although the research team did not stipulate what the care home should do with the voucher, it was suggested that it could be used to purchase an item for the care home, the staff team or residents. Ethical approval was obtained from the University of Leicester’s Research Ethics Committee for Medicine and Biological Sciences (15,304).

Table 1. A table to describe the characteristics and number of participants recruited at each care home.

						Participants recruited
Number	Type of service	Type	Provider size	Number of beds	CQC rating	Manager	Deputy manager	Registered nurse	Senior carer	Carer
1	Care home with nursing	Private for profit	Small chain (5)	35	Good	1	1	1	1	1
2†	Care home without nursing	Charitable not for profit	Small chain (5)	45	Good	1	1	*	3	2
3†	Care home without nursing	Private for profit	Independent (1)	15	Good	2	*	*	*	2
4†	Dual registered care home	Private for profit	Large chain (120)	60	Good	1	0	3	2	0
5	Care home with nursing	Private for profit	Independent (1)	40	Outstanding	1	0	1	1	0
6	Care home with nursing	Private for profit	Large chain (300)	35	Requires Improvement	1	3	2	*	1

† Took part in both phases of data collection

* Indicates this job role did not exist within specific home

Phase 2: ethnographic observations

We conducted ethnographic work at three care homes where staff had taken part in interviews. We used data from the first phase of data collection to determine the sampling of care home sites. We identified two features in interviews which influenced staff experiences of decision-making: (i) whether the home provided nursing services, and (ii) the extent to which the staffing structure was ‘hierarchical’. Three sites were invited and agreed to take part in the ethnographic observation stage. Information about these homes is shown in Table 2.

Table 2. A table to describe the staffing structure and roles at each care home.

Number	Type of service	Staffing structure
2†	Care home without nursing	Hierarchical: Manager, Deputy Manager Senior Carers, Carers and a large team of auxiliary staff consisting of an administrator, catering staff, cleaning staff, laundry staff, a full-time maintenance person and regular volunteers
3†	Care home without nursing	Non-hierarchical: Business Manager, Care Manager, Care Workers and a small number of auxiliary staff employed as cleaners and one part-time maintenance person
4†	Dual registered care home (with and without nursing)	Hierarchical: Manager, Deputy Manager/Clinical Lead, Unit Managers (Nurses), Registered Nurses, Senior Carers, Carers and a large team of auxiliary staff consisting of an administrator, receptionist, catering staff, cleaners, laundry staff, several maintenance people, activity workers and regular volunteers

The first author (FHH) carried out 113 hours of observations over 26 visits across the three sites between April 2019 to November 2019. Our approach was consistent with the ‘short-term, theoretically informed’ approach to ethnography, (Pink & Morgan, 2013). This approach emphasises the need for deliberate engagement with the research field. As such, our observations focused on staff activity interactions and documentation that occurred within the care home. In particular, we focussed on the ways staff managed residents’ health conditions, responded to deteriorations in residents’ health, and the work that care home staff undertook to maintain residents’ health and prevent hospital transfers. This form of ‘guided observation’ (Leverton et al., 2019) has been identified as being particularly useful for exploratory work. In addition, we engaged in deliberate ethnographic-theory dialogue – spending time iteratively collecting data, searching for and re-examining data in light of relevant theories and concepts.

Observations occurred at different times of the day across all days of the week. During each visit to a care home, hand-written notes were made. After each visit, initial thoughts were audio-recorded. Hand-written notes and audio-recordings were later transcribed into a single electronic Word document (one per home) which was uploaded to NVivo for data analysis. Observations were supplemented by documentary analysis (for example, of policies and procedures relevant to hospital transfers) and informal, unstructured conversations with care home staff, residents, visiting family members, and visiting healthcare professionals. Care homes were provided with a £100 multistore gift voucher as a thank you for their participation. Ethical approval for this phase was obtained from the Social Care Research Ethics Committee (18/IEC08/0033).

Data analysis

We collected and analysed data concurrently, combining data from both phases. In line with the Straussian approach to grounded theory (Singh & Estefan, 2018), we analysed data in three steps using the constant comparative method (Bryant & Charmaz, 2007). First, we coded each section of each data source according to the phenomenon or concept that was being discussed, thus identifying a number of ‘open codes’. Second, we explored the relationships between codes, producing ‘axial’ codes, before finally identifying ‘selective codes’ to which all data related. We initially carried out coding by hand before using NVivo12. We continually reviewed, validated and refined these as a team throughout data analysis until theoretical saturation was achieved.

Early in data collection, we identified risk as an emergent theme. We then explored this theme through further interviews and ethnographic observation. Forms of risk were identified first as ‘open’ codes (early in analysis) and then as selective codes. Staff identified several forms of risk which influenced their decision-making: risks to the resident; to staff (as decision-makers) and their social relationships (for example, with colleagues, residents, families, or external healthcare professionals); to the care home as an organisation (for example, to the care home’s reputation (Rothstein, 2006)); and to the wider health and social care systems (for further information see, Harrad-Hyde et al. (2022)). Given the emergence of risk as a key issue in staff decision-making, in this article, we uses risk work as a theoretical lens through which we analyse ethnographic and interview data.

Findings

Conceptualising risk: ‘knowing’ residents and other forms of risk knowledge

Staff generally described an intuitive approach to identifying risk, using phrases such as ‘you just know!’ (Carer 2, Care Home 2). This was underpinned by a significant body of experiential knowledge about what was considered ‘usual’ for each resident. Staff used this knowledge to identify subtle changes that signified a potential deterioration in residents’ health. Examples included but were not limited to changes in mobility, appetite, consciousness, continence, behaviour and physical appearance.

Although staff identified some situations in which a hospital transfer was always felt to be necessary (for example, when residents had suspected head injuries or other trauma), more often – in situations that did not trigger an immediate transfer – staff drew upon their experiential knowledge to make resident-specific judgements about the significance of the change that they observed.

“As a carer you are there every day. You are on the floor. You see it, you hear it, you know what’s what. You are getting this person up every single day and know something is not quite right.” (Carer 2, Care Home 2)

“We have one resident here who always eats breakfast, always eats lunch, again at tea-time and at supper. Say he went off his food, we would know there was something off.” (Carer, Care Home 6)

Besides ‘resident-specific’ knowledge, staff also used experiential knowledge, obtained through caring for other residents, to identify patterns that might indicate a potential deterioration. For example, staff suggested that ‘dark, foul-smelling’ urine could be indicative of a urinary tract infection. Although staff learnt some patterns of change through formal training, this appeared to be gained mainly through their own experiences or from informal learning from other carers. Experiential knowledge therefore shaped both their understanding of what was (un)usual for a particular resident and what should be more generally considered concerning.

“Experience will help them … Nurses and care workers … time can give them more examples of different situations. I have worked for hospitals, in different countries as well and in different situations.” (Nurse 1, Care Home 4)

Staff were likely to interpret a situation as more serious (and therefore were more likely to initiate a hospital transfer) when faced with something that was particularly unusual for the resident. This was noted both in interviews and ethnographic observations: during handover meetings at each shift change, the person leading the meeting often discussed each resident in reference to what was usual for them. For example, two phrases commonly used to signal that a resident had been their usual self was ‘no changes and no concerns’ or ‘[John] has been [John]’. Frequently when this was the case, the person leading the handover would not spend further time discussing the resident and instead utilised the time to discuss more unusual occurrences.

Care home staff took a more cautious approach to decision-making when their experiential knowledge was limited. This might occur when a resident had only recently moved into the home or when new or temporary staff were working in the home. In these situations, staff described processes by which they would seek out colleagues’ experiential knowledge or call an external healthcare service for support. For example, in response to a vignette in which a new resident exhibited a poor appetite, reduced fluid intake and appeared to be drowsy and confused, one participant responded:

“It depends how well they know her. She had only been [living at the care home] for three weeks. If she had been there three years and staff felt that they knew her … They might speak with the family, because she has only been there three weeks, we don’t really know her.” (Deputy Manager, Care Home 1)

Having detected a change, staff then combined, or ‘bricolaged’ (Horlick-Jones et al., 2007), this experiential knowledge with other forms of risk knowledge to decide what actions they should take. This included: written advance care plans and emergency healthcare planning documents; their understanding of the wishes, preferences and reactions of the resident’s family (who had a powerful role in transfer decisions); and structured clinical observations (typically measuring residents’ respiration rate, oxygen saturation level, blood pressure, pulse rate, level of consciousness and temperature).

There were some differences in how structured observations were incorporated as risk knowledge. Some staff reported ‘doing the obs’ routinely in their care home for residents who were well, but it was more common for observations to be done to corroborate or further investigate a concern that had arisen due to a change that had been observed. During interviews, structured observations were most often mentioned by registered nurses and management staff with a background in nursing, demonstrating how individual risk logics can be bound up with an individual’s professional identity, expertise and prior training.

Navigating tensions between different forms of risk knowledge

Although care home staff positioned their experiential knowledge as a legitimate form of risk knowledge that could be used to inform decision-making within the care home, staff felt less comfortable with the perceived legitimacy of this approach beyond the home. By contrast, ‘doing the obs’ – drawing on risk knowledge that was perceived to be more objective and acceptable to external colleagues – provided staff with knowledge that could be codified and more easily articulated. Although staff often combined structured clinical observations with their experiential knowledge in discussions with external professionals, staff sensed that external professionals tended to privilege ‘objective’ observations over more subjective reports of ‘change from usual’.

[Interviewer: “What happens once the ambulance staff arrive?”]“They want to see all the information, the [medication] sheet, the past history of the patient and the [diagnoses] of the patient. They want to know all of this, when they last saw the GP and things like that.” (Nurse 2, Care Home 4)

At times, different sources of risk knowledge contradicted one another and led to conflicting ideas about the most appropriate way to respond to a situation. The most frequently mentioned example, discussed across all six care homes, related to resident falls in which there was no obvious injury. Several care homes had a policy advocating sending residents to hospital following all falls and staff understood the importance of ruling out a possible break or fracture. However, staff were also aware that initiating a transfer could result in a resident spending several hours in hospital, often unaccompanied. Therefore, when their experiential knowledge suggested that a resident was unlikely to be injured (and therefore less likely to benefit from a transfer) they were reluctant to initiate a hospital transfer.

Care home staff responded to this tension in different ways: sometimes they would initiate steps towards a hospital transfer, even if this did not align with their experiential knowledge, as the decision to work outside of the guidance provided in written policies was perceived to be too ‘risky’ for the resident and for themselves as a decision-maker. In other instances, when faced with such dissonance, care home staff suggested that they might choose to privilege their own subjective, experiential knowledge.

“Sometimes I know they’re gonna be fine but … there’s always a part of me that thinks ‘what if they are not?’ and then I think ‘well I have to [call for help] because it says in the policy’.” (Carer 2, Care Home 3)

“If someone falls we will call 999.”[Interviewer: “Do you call 999 every time someone falls?”]“It all depends, because we were told by a GP if somebody has a fall to call 999 - but not really, no. We check them over. If they are in pain, if it is the hip or anything like that then we do.” (Senior Carer 2, Care Home 2)

By ‘bending the rules’ (drawing on experiential risk knowledge), staff were able to reduce the risk that a resident might spend unnecessary time in hospital. However, this introduced new risks. For example, if staff decided not to transfer a resident to hospital who was later found to have an injury, the resident may experience avoidable pain and distress and receive less timely treatment. Importantly, this could also introduce risks to staff members as decision-makers: including potential professional risks (being reprimanded), emotional risks (feeling as if they had made the ‘wrong’ decision) and social risks (through damaged relationships and trust with colleagues, the resident, their families and external healthcare professionals).

Social interactions influencing risk knowledge

When making decisions about potential hospital transfers, the processes of coming to know what was a ‘change from usual’ and weighing-up potential benefits and risks, were influenced by a range of social interactions. At an interpersonal level, staff perceptions of what represented a change for each resident – and, by extension, to what extent the changes were a cause for concern and what an appropriate course of action might be – were negotiated through interaction with others, including residents’ families, colleagues within the care home, healthcare professionals from external services and, to a lesser extent, residents. Reflecting on a transfer that had occurred recently, one participant stated:

“His wife said he wasn’t well. He was twitching but I was thinking ‘this is normal’ but she said ‘this isn’t normal’. So I thought ‘OK maybe she knows him more than I do’.” (Carer, Care Home 6)

Individual perceptions of risk could also be influenced by social interactions that were structured and formally organised. Two of the six care homes held regular sessions in which staff reflected – as a team – on recent events such as hospital transfers or deaths. Staff were encouraged to reflect on whether the hospital transfer was necessary and if a different course of action may have been ‘better’. These scheduled sessions provided a means through which individual understandings of risk could be shaped, through formally organised interactions with colleagues.

The prioritisation of particular forms of risk knowledge could also be shaped by organisational processes. For example, staff used various forms of documentation to evidence the work and decision-making processes they undertook. However, these varied significantly across homes. In one of the three care homes where ethnographic data collection took place, staff made hand-written notes in residents’ diaries at specific times of the day. At another site, staff used hand-held electronic devices that connected to an online recording system to document their work in real-time. At the final site, staff recorded their work on an online system at the end of each shift. Both electronic systems were able to flag tasks or documentation that had not been completed – for example, if a resident’s blood pressure had not been recorded or a care plan had not been updated – thus prompting staff to prioritise this task.

Furthermore, whilst all homes had established practices for staff to document the work and decision-making processes staff undertook, some risks were more prominently discussed in particular homes. At one site, there was a pronounced emphasis on avoiding instances in which staff could be seen as potentially responsible for deteriorations in residents’ health, for example, through not identifying deteriorations in a timely manner or failing to respond appropriately. Signage in staff areas reiterated the importance of undertaking different tasks (and the consequences to staff of not doing so). In this home, staff who took part in formal interviews were more likely to mention the need to be ‘covered’ than in other homes; and during ethnographic data collection, staff appeared to complete much more documentation than at other sites, with more of the day-to-day work tied to staff perceptions of professional risk. By contrast, in other homes, whilst staff were aware of professional risks, these were less ‘visible’ in the home environment and discussed less frequently in interviews.

Managing risk: strategies to mitigate risks

Documentation as a risk intervention

As mentioned in the previous section, staff placed importance on documenting their actions when responding to a potential deterioration in residents’ health. Documentation was often discussed as a means of ensuring staff were ‘covered’, demonstrating the actions they had taken to detect and respond to observed changes in residents’ health. As well as documenting their actions and the justification for these, staff documented who had been involved in decision-making, including members of the staff team, residents’ family members and external healthcare professionals. It appeared that documenting others’ involvement was used as a method of sharing risk.

“As long as I’ve documented it I’m covered. I have told younger staff, if you have involved a senior put who you have informed, put the initials. You know, just to cover your own back basically and to support the service user the best way you’re supposed to.” (Carer 2, Care Home 2)

Involving colleagues and healthcare professionals in decision-making

Staff valued being able to consult their colleagues and to draw on the knowledge of others within the home, particularly when they perceived uncertainty. However, in order to provide an effective means of minimising perceived risk, staff needed to trust that their colleagues could identify and interpret changes in residents’ health accurately. When working within a team that had not had opportunities to develop trust, individuals were likely to choose to implement more risk averse practices.

“I will ask the other nurses … I trust their judgment. That is important - to have trust … the longer they are here, the more you have trust … The new ones I find it very difficult because I don’t know if what they are telling me - if they observe it the correct way … I will ask staff who have worked here for years … I say ‘do you think there is really a deterioration in Mrs X’ and she might say ‘no sometimes she has episodes like that’ and I will believe her.” (Nurse, Care Home 5)

One consequence of this approach to managing risk within the home was that staff in senior positions were expected by junior staff to take on more responsibilities and professional risk associated with decision-making. Seniority could be determined by formal staffing hierarchies. In addition, in homes with a less pronounced staffing hierarchy, such as Care Home 3, seniority was often associated with an individuals length of employment in the home and the number of hours worked each week. Staff who had worked in the home for a greater length of time and staff who worked a greater number of hours each week were perceived to be more ‘senior’ than newer members of staff and staff who worked fewer hours.

“They want reassurance. If we have an accident here, you hear the emergency bell. They don’t switch it off until they have a member of management or a nurse … Even though their colleagues are there … They want some reassurance … an element of ‘well it’s your responsibility at such a high level’” (Manager, Care Home 4)

Similarly, when staff felt unable to assess or manage potential risks amongst the staff team, they would contact external healthcare professionals. Staff suggested external healthcare professionals could offer ‘professional advice’ (Deputy Manager, Care Home 1) that would enable them to better assess and manage a situation. However, staff also discussed the involvement of external healthcare professionals as a means through which they could shift responsibility away from themselves (as an individual) and the care home (as an organisation).

“It’s the GP’s judgment call, or the out-of-hours, as to whether they should be coming to see that person and whether they should advise us to send for emergency services. It takes away quite a lot of responsibility, especially from the senior carers … it is spreading the load as such, erm, it’s not a way out, it’s not a cop out … It’s just giving people an opportunity to discuss it with a medical practitioner.” (Manager, Care Home 4)

Both of these strategies – involving care home colleagues and external healthcare professionals – were described as ‘erring on the side of caution’ and as a way of shielding staff from some of the potential personal, professional and social risks associated with decision-making. However, these strategies became less effective during nights and weekends. Although all homes had an emergency ‘on call’ system that enabled staff to contact a member of the management team, they were rarely present on site at these times. Staff were also unlikely to be able to contact their usual general practitioner (GP). As a result, staff described decision-making as especially difficult during these periods.

“In the evening there is only one nurse … It makes a difference because you haven’t got anyone … In the evening, when management has gone, you are on your own, so all the pressure is on you.” (Nurse, Care Home 6)

At these times, staff were able to contact an out-of-hours GP or ambulance services. However, staff across all care homes reported experiences of out-of-hours services ‘refusing to come out’ and recognised that healthcare professionals from these services were less likely to have an existing relationship with the resident and therefore would lack experiential knowledge of what was (un)usual for that resident. Therefore, whilst the support of out-of-hours services and ambulance professionals were used in instances when staff felt they could not manage a situation and the associated risks amongst the staff team, staff generally preferred support from a regular GP when possible.

Involving residents’ families in decision-making

Although staff spoke fondly of residents and almost always spoke of making decisions that would improve or retain their health and/or quality of life, residents were less frequently described as being involved in decision-making. More often, staff discussed the involvement of family members in decision-making. This was seen as especially important when a resident lacked capacity to make decisions, but even when decision-making centred on a resident who had capacity, staff referred to the involvement of family members.

“I think some of it falls down to capacity. Do they have capacity in that specific moment to make that decision? I think if it was something serious then you would have to get the family involved as well.” (Carer 1, Care Home 3)

Often, involving family members was described as a ‘courtesy’. However this understated the observed importance attached to family member’s involvement in decision-making, This involvement also served risk-related functions: staff anticipated the likely response of family members when weighing up the most appropriate course of action following a deterioration.

[Interviewer: “You have said that you inform families when residents fall, regardless of whether the resident is injured. Why is that?”]“We just inform them, it’s courtesy. It’s courtesy. [Interviewer: “What might happen if you didn’t call them?”]“You can get funny families – I mean, if my mum or my dad was here and someone turned around and said ‘oh dad had a fall’ I’d be like ‘well why weren’t we informed’ it’s just courtesy.”(Carer 1, Care Home 4)

Staff felt that most interactions regarding the appropriateness of hospital care were characterised by agreement. However, disagreements between staff and family members were a significant source of difficulty – particularly if staff and family members held different views about the ‘right’ thing to do. Staff described experiences of residents being transferred to hospital due to pressure from family members, even when staff felt a transfer was not appropriate. Staff raised concerns that acting contrary to the wishes of family members’ presented a risk of legal and/or disciplinary procedures that could have significant consequences for them as an individual. In these situations, staff described having to balance risks to the resident, individual staff, the care home as an organisation and their relationship with families. In these situations, the relationship with families was often privileged due to the potential consequences for staff and the care home if families complained.

“It is a very sobering thought that you and your career, especially as a nurse, you can be suspended and scrutinised, you can face legal ramifications, families can sue.” (Manager, Care Home 4)

Responding to disagreements amongst decision-makers

Although involving others in decision-making provided a means for staff to minimise several forms of risks, it also introduced the potential for disagreement amongst decision-makers. Some participants suggested that they would be happy to follow the advice of others who were perceived to be more senior to themselves, even if the advice was at odds with their own perception of what was ‘best’ for the resident. This could also include ‘senior’ colleagues within the care home and external healthcare professionals (for example, GPs and ambulance staff). This appeared to be related to a desire to be ‘covered’ and to minimise individual risks to themselves. When there were disagreements, perceptions of authority of the decision-maker (within and outside the care home) were important to how these were resolved.

A carer was particularly uncertain about the cause of a rash-like mark on the inside arm of a resident with advanced dementia. She contacted the GP, who visited the resident, prescribed antibiotics (for the possible skin infection) and scheduled an x-ray at the local hospital for two days later (to rule out possible injury). After the GP left the care home, the nurse expressed concerns that an x-ray might not be in the ‘best interest’ of the resident, who would likely spend a long period of time unaccompanied in hospital. However, she stated that once the GP became involved, the decision was “taken out of [her] hands”. When I asked why, she said, ‘because I need to be covered’. Since she had requested support, she would find it difficult to justify acting in ways that contradicted the GP’s recommendation.(Field Note, Care Home 4)

There were also instances where care home staff were willing and able to challenge the decisions of others. However, this generally occurred when staff felt that greater ‘escalation’ was required. Staff felt less able to ‘push’ for the resident to remain in the care home when another person deemed a transfer necessary. This seemed to be related to difficulties in providing justification for ‘not acting’. However, the overall effect was that, in the absence of a ‘senior’ decision-maker individuals felt more comfortable to advocate for more but not less treatment, which moved decisions in a more invasive direction.

“[Ambulance staff] know more about stuff than I do so I trust their decision. But I feel like, if I did [disagree], I would say. I would really try and make sure they went [to hospital].” (Carer 2, Care Home 3)

“If the family are saying that they don’t want him admitted to hospital, I would overrule that if I think there is a fracture there and I would call the paramedics anyway and get them to check him over and potentially take him in.” (Senior Carer 3, Care Home 2)

Using advance care planning documents as a risk intervention

Although described above as a source of ‘risk knowledge’, advance care plans could also be used as a risk intervention, providing staff with an opportunity to manage risk and justify their actions, particularly when staff were choosing not to initiate further action, for example, by deciding not to transfer a resident to hospital. Without an advance care plan in place, staff suggested they could feel obliged to initiate care and/or treatments that they felt would not be in the best interest of a resident (for example, initiating a hospital transfer or performing cardio-pulmonary resuscitation) in order to avoid being accused of not fulfilling a duty of care. Following an advance care plan therefore enabled staff to frame their decision as a deliberate and considered act of commission (thus actively deciding to ‘do nothing’), rather than a passive act of omission (failing to act or respond). However, when advance care plans were perceived to be unclear and ambiguous, care home staff, and sometimes other healthcare professionals too, did not feel comfortable following the plan, which increased the likelihood that a resident might be transferred to the hospital.

“Most recently we had a gentleman who was breathless, and he was unresponsive and the care plan wasn’t clear. It didn’t tell us if he was for hospital admission or not … In the end I called the ambulance … it’s just tricky sometimes to understand what they wish.” (Nurse 3, Care Home 4)

Discussion

When faced with situations in which a resident may require a hospital transfer, staff make complex decisions in which they weigh up a number of benefits and risks. In this article, we position staff decision-making about potential resident hospital transfers as a form of risk work (Gale et al., 2016). Gale and colleagues suggest that in order to understand the experiences of frontline staff who work with risk, we should attend to: the ways individuals assess and conceptualise risk (sources of risk knowledge); the practices individuals use to mitigate against and manage risk (risk interventions); and the ways in which risk practices are influenced by, and embedded within, interpersonal relationships (social relations; Brown & Gale, 2018a; 2018b; Gale et al., 2016).

By explicitly using this framework to explore the data collected, the findings we present in this article demonstrate the relevance of this concept to care home staff decision-making regarding hospital transfers from care homes. In doing so, we demonstrate the utility of the concept of risk work to organisations that fall under the umbrella of social care, which have largely been neglected by academic research in comparison to healthcare.

Conceptualising care home staff decision-making about potential resident hospital transfers as a form of risk work provides novel insights into the phenomenon. For example, using the language of ‘risk knowledge’ enables an exploration of the different forms of knowledge that staff draw upon to assess risk. It also provides a framework through which to explore the tensions between different forms of knowledge, to examine how these are ‘bricolaged’ (Horlick-Jones et al., 2007), and the conditions under which some are privileged over others.

Our findings suggest that care home staff placed importance on the experiential knowledge they held, which they felt they could use to assess the likely risks of transferring a resident to hospital (or not). This form of risk knowledge was developed a posteriori and encompassed ideas of what was ‘usual’ for residents (both for individual residents and across residents). However, staff often combined this form of risk knowledge with several other disparate forms of risk knowledge. As such, our findings provide support for the work of Horlick-Jones et al. (2007) and others who have suggested that individuals ‘bricolage’ several sources of risk knowledge.

Despite valuing their experiential risk knowledge,, in times of uncertainty, care home staff often privileged more formal and objective forms of risk knowledge (for example, by referring to advance care plans or completing structured nursing observations), as these felt easier to defend and more likely to be valued by healthcare professionals. Therefore the findings of this study could point towards the use of standardised tools that are able to accommodate (and therefore legitimise) the experiential knowledge of care home staff. One example of this is the RESTORE2 tool, developed by the Royal College of Physicians, which combines ‘soft signs’ of deterioration (for example, changes in what is usual for the resident) alongside structured nursing observations.

The component ‘social relations’, described as one of the three central components of risk work by Gale et al. (2016), encourages a focus on the wider social context surrounding individual experiences of risk, enabling the identification of forces that can promote or limit the provision of care. By focussing on the social relations surrounding transfer decisions, we provide insight into the ways risk practices are influenced by other people and socially constructed. The findings that we present provide evidence to suggest that sources of risk knowledge and individual understandings of risk can be influenced by social interactions with others. Existing work has focussed on interactions between healthcare professionals and patients. However, we demonstrate the importance of interactions that occur amongst professionals/colleagues. We highlight the ways individual understandings of what is ‘usual’ for residents can be negotiated through interactions with others and the ways care home staff anticipate the likely responses of others to the actions they take. In doing so, we identify and offer plausible explanations for occasions where staff may feel unable to act on their own perception of what is ‘best’ for a resident – for example, when staff anticipate that their own perception conflicts with the view of families and/or healthcare professionals.

Viewing hospital transfers through the lens of risk work also provides new insights into trends that staff perceive to occur in practice. For example, in this article we frame the support of senior staff, managers and a regular GP as a risk intervention. Conceptualising support from senior staff, managers and regular GPs in this way provides a plausible explanation as to why staff report that decision-making can feel particularly ‘risky’ during periods of time that these people are not available to be consulted (particularly evenings and weekends). Furthermore, framing escalations to external healthcare professionals as a risk intervention also provides a plausible explanation for variations in transfer rates between different care homes. It is likely that, in care homes where staff feel unable to obtain support from a trusted GP, the only way to stay ‘on the safe side’ is to call for an ambulance, which may in turn increase transfer rates and healthcare costs.

Drawing on the concept of risk work provides a deeper understanding of behaviours that may, at face value, appear illogical. For example, existing research has suggested that care home staff sometimes initiate a transfer to the hospital without the expectation of better clinical outcomes or quality of life for the resident (Arendts et al., 2013, McCloskey, 2011). By thinking about transfer decisions explicitly in terms of the different forms of risk which staff manage, we can start to understand why, at times, care home staff may feel as if it is safer or less risky to do more: to call an external healthcare professional, to seek additional support via the NHS telephone (111) service or initiate a hospital transfer, rather than to care for the resident within the home. In these circumstances, a transfer may be considered clinically ‘inappropriate’ (in that it will not be beneficial for the resident), yet still a logical response situated within the broader social contexts of risk and accountability relations that shape staff decision-making.

Encouraging resident-focussed decision-making by explicitly acknowledging risk

At times, staff may decide to act against their own perception of what is ‘best’ for residents in order to reduce other forms of risk – for example, risks to themselves and the care home in which they work. Future work should explore ways of reducing non-resident related risks during transfer decisions (without compromising on the quality of care offered) to enable care home staff to prioritise risks to residents. For example, the findings we present suggest that individual ways of assessing and managing risk can be shaped by the wider social context. The notion that care homes, as organisations, can shape individual understandings of risk provides promise for the future. If care home organisations are able to shape individual perceptions of risk, and in turn to shape the steps staff take to manage risk, then they can do so in ways that encourage resident-centric decision-making. In two of the six care homes, staff took part in regular sessions in which they reflected on events that had occurred within the home. This included, but was not limited to, resident hospital transfers. Introducing discussions about the risks associated with hospital transfers could provide an opportunity for staff to reflect critically on their thought processes and the factors that influence them and to identify strategies and interventions that could lessen some of the risks that staff perceive. Even in the absence of immediate resolutions, explicitly confronting the ways risks are understood could provide a means for allaying staff fears, which in turn, could encourage practices that prioritise risks to residents.

Strengths and limitations

Strengths of this work include the use of ethnography and interviews as distinct methods of data collection across a purposively sampled range of care homes and participants. Although interview data can be subjective and prone to desirability bias (with participants reporting the more positive aspects of their behaviour), conducting ethnographic fieldwork provided us with an opportunity to triangulate data and to compare what was reported in the interviews with what was observed in practice. This project did not have an explicit a priori focus on risk or risk work. Instead, ‘risk’ was identified as an important overarching theme within our work. That risk was identified in an inductive way is advantageous, ensuring that the application of the concept of risk work is grounded in care home staff’s lived experience of transfer decisions.

Data were collected prior to the COVID-19 pandemic, that had a profound impact on care homes worldwide. Now and in the future, when making decisions about potential resident hospital transfers, care home staff are likely to consider the possibility that a resident could have COVID-19 whilst in the care home or be exposed to COVID-19 as the result of a hospital transfer. However, the central argument of this paper – that care home staff decision-making about resident transfers can be conceptualised as a form of risk work and that conceptualising decision-making in this way could be advantageous – remains relevant in light of COVID-19.

In previous work we have outlined several forms of risk staff feel responsible for balancing when deciding whether or not to initiate a resident hospital transfer. This includes risks to residents, staff and their social relationships, care home as an organisation; and wider health and social care systems (Harrad-Hyde et al., 2022). Although conceptualising hospital transfer decisions as a form of risk work appears to be helpful for understanding the work of care home staff, further research is needed to explore the application of this concept to developing interventions aimed at ensuring residents receive appropriate care in an appropriate setting.

Conclusion

In this paper, we conceptualise staff decision-making about potential hospital transfers for residents as a form of risk work. Our results suggest that staff draw on a range of sources of risk knowledge to assess potential risks, including their experiential knowledge of what is usual for individual residents and across residents, structured clinical observations and advance care plans. In addition, individual understandings of risk were influenced by social interactions with others, both at an interpersonal and organisational level. Our paper highlights a range of strategies that care home staff use to manage risk. This includes the use of documentation, involving others in decision-making and anticipating and seeking to resolve potential disagreements in decision-making.

By understanding transfer decisions explicitly in terms of the different forms of risks staff manage, this paper provides new insights into hospital transfers from care homes. For example, although we have discussed the pre-requisites for advanced care planning documents to be effective in the context of transfer decisions in a previous paper (Harrad-Hyde et al., 2021), framing these documents explicitly as a risk intervention provides further support for the important role they play in reassuring staff, providing them with greater ability to justify their decision-making to others.

In presenting our findings, we provide further support for the concept of risk work as outlined by (Gale et al., 2016), by applying this concept to the everyday risk practices of staff working in care homes. Future policy and research into hospital transfers from care homes could benefit from drawing on the concept of risk work, as this would provide a more robust theoretical model for the development of interventions designed to ensure residents receive appropriate care in appropriate settings.

Acknowledgments

The ENRICH (Enabling Research in Care Homes) teams at Clinical Research Network West Midlands and Clinical Research Network East Midlands supported the identification of potential care homes. Care home managers allowed us to conduct research at each home and care home staff, residents, relatives and visiting healthcare professionals generously gave their time.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

Fawn Harrad-Hyde's doctoral work was funded by a University of Leicester College of Life Sciences student ship.