Limited conversations about constrained futures: exploring clinicians’ conversations about life after stroke in inpatient settings

Qualitative approach and research paradigm

Our study used the Interpretive Description methodology (Thorne 2016), an applied qualitative approach, with a relativist ontology, holding that multiple realities exist and are equally valid. These realities are constructed through interaction and language (Thorne 2016). Interpretive Description generates new knowledge into the ‘subjective, experiential, tacit and patterned aspects’ (Thorne 2016, p. 41) of healthcare practice, addressing practice-based questions and generating useful, meaningful, and applicable information for everyday practice (Thorne 2016).

Research context

This research occurred in three stroke services across two urban District Health Boards in Aotearoa New Zealand, including (1) an organised acute stroke unit providing hyperacute and acute stroke care, (2) an integrated hyperacute/acute/inpatient rehabilitation stroke unit where people are transferred between ‘acute’ and ‘inpatient’ beds once they were medically stable, and (3) an inpatient rehabilitation ward providing stroke and general rehabilitation services. Each service was overseen by a medical consultant (neurologist, geriatrician and/or rehabilitation specialist) and staffed by a multidisciplinary team. Before developing the research protocol, the first author met with staff in each locality to identify their priorities for the research and sought their perspectives on how it could be conducted to be meaningful but minimally obtrusive for patients and clinicians. Ethical approval for this research was granted by the Health and Disability Ethics Committee (13NTB103) and the Auckland University of Technology Ethics Committee (19/303); approval was gained from all research localities.

Participants and sampling

The research centred on interactions occurring within clinical dyads – the person with stroke and their treating clinician. We first recruited the person with stroke, then recruited staff members working with them. People with stroke were eligible for inclusion if they were over 18 years of age, English-speaking, able to communicate with the researcher (with the use of supported communication if aphasic) (Kagan 1998), and able to provide informed consent. Sampling was purposive, seeking variation in stroke severity and impairments. Potential participants were screened by stroke service staff. If they appeared eligible and were interested, the staff member provided their details to the research team. We then met the person, explained the study, and completed informed consent. We identified two people who did not meet the eligibility criteria and one person declined. All staff were made aware of the study prior to its commencement and had the option to opt-out from being approached to participate. Once people with stroke were recruited, we directly contacted staff working with them, re-orientating the clinicians to the study and seeking consent. Only one staff member opted out. Two staff members declined to be participants. All staff members working with the person with stroke were eligible, but we drew on purposive sampling to prioritise which staff were approached. In this, we sought variation in discipline. Principles of information power (Malterud et al. 2016) informed our decisions about sample size. We obtained information power by having a specific and somewhat narrow aim, participants with a depth of experience, variability among participants (patient and staff), and rich data constructed and analysed by a research team with significant clinical and research expertise.

Data construction

Data construction was multi-faceted and differed depending on the physical setting. Observations centred on interactions between the person with stroke and clinicians. These were recorded on a voice recorder and detailed field notes were taken during and after observations. In acute services, we observed every interaction between 7 am and 7 pm during their episode of care, unless any person indicated this was not appropriate or wanted. In inpatient rehabilitation, we conducted episodic observations, completing at least six 4-h observation periods during each person’s length of stay. We used event sampling to purposefully sample events (e.g. ward rounds). We conducted informal interviews with participants with stroke during observations, seeking their reflections on their stroke, its impact, their care, and their future hopes and goals. We completed short ethnographic interviews immediately after observations where possible, seeking reflections on the interaction. These lasted between 5 and 30 mins. After each patient was discharged, we conducted semi-structured interviews with purposefully selected clinicians, aiming to interview at least two clinicians who had worked with each patient participant and with whom we had completed several observations. We sought to explore their reflections on their interactions with patients (including those not observed by the researcher) and their reasoning behind their ways of talking about the future. Examples of questions are provided in Table 1. These interviews were dynamic and responsive to what was observed, and as such, most short ethnographic questions were very specific to observed features of interaction. Interviews were audio-recorded and transcribed.

Data analysis

We used reflexive thematic analysis (Braun and Clarke 2021). We familiarised ourselves with the data by listening to recordings, reviewing fieldnotes and transcripts, and creating analytic memos. Through coding, we extracted data relating to the idea of ‘the future’, or indeed, reflecting anything time-related, including references to the past. What constituted the ‘future’ was broad; we included data about any point in time beyond the immediate situation. Semantic codes stayed close to the data (e.g. ‘focusing on the immediate present’) whereas latent codes reflected a deeper interpretation (e.g. ‘waiting in a holding zone’). Initial themes were generated by grouping codes, although many reflected topic summaries (Braun and Clarke 2021), for instance, ‘context of care’. We then engaged more deeply with the data, attending to features of conversation and paying particular attention to the data from observed interactions. Through this, the temporal aspects of conversation became more apparent. Analytic questioning furthered the analysis, for example, ‘How is the notion of ‘possibilities’ evident in interactions?’ These areas of focus were influenced by our previous research and positioning (see below). We re-coded the interviews with clinicians to better understand their (self-identified) rationale for their conversational approaches. This helped us attune to different notions, such as the agendas of services. Themes were developed and refined through mapping and team discussions. Through this iterative process, three themes were constructed.

Research credibility was achieved through multiple strategies (Thorne 2016). Integrating multiple forms of data from across multiple participants and contexts, integrating raw data into analytic representation, and situating findings within the broader practice context aid the reader in interpretation and support representative credibility. Detailed descriptions of data construction and analysis processes, accompanied by thick descriptions, demonstrate analytic logic. Team discussions regarding analysis and discussions about the findings with experienced clinicians provide interpretive authority.

Researcher characteristics and positionality

The research team is experienced in conducting clinically orientated health research. We bring clinical and research experience in speech-language therapy (F.B.), health psychology (N.K.), physiotherapy (A.S.) and pharmacy (J.D.), and over many years, we have conducted research in patient-provider communication, hope, and well-being in the context of living with long-term neurological conditions. We view the relationship between the person with a stroke and their clinician as integral in care, providing a platform for engagement and helping people as they move forward after a significant diagnosis or health event. This informed our decision to focus on the communication between the dyad; our combined interests in communication hope and well-being likely influenced how we constructed the analysis.

The research was led by F.B., an experienced ethnographer. She completed all data collection, while all authors contributed to data analysis. Although she had previous clinical experience in the localities, she had not worked in them for over 10 years; through professional networks, she was known to many of the clinician participants. In this research, she was an ‘outsider’ with some familiarity with the setting. She took the role of observer-as-participant (Gold 1958), observing interactions but not taking part unless a participant directly addressed her.

Theme one: constrained temporal horizons

This theme describes the constrained nature of temporal horizons – the expanse of time constructed and made visible within people’s interactions. Interactions about the future focused on the short-term future. This reflected specific temporal horizons, which arguably shaped clinicians’ actions, informing what was prioritised. Care focused on the short-term future, with limited discussion observed about life beyond the hospital or first days at home. Disciplinary trends were evident, with nurses focusing on activities within their shift or the next 24 h. Completing required ‘cares’ (Interview, Arthur’s nurse) – monitoring for deterioration, providing medications, and not leaving work undone – was a core drive for nurses in acute care. Doctors discussed what needed to happen before discharge from their service, problem-solving issues so that people could be discharged. In acute care, there was a 'day-by-day’ focus (Interview, Arthur’s physiotherapist), which co-existed for allied health with a focus on discharge preparations and was reflected in quotes, such as ‘we’ve got the family meeting on Monday to talk about home, what we need to aim for, what needs to be done’ (Observation, Alison’s occupational therapist; OT). They also looked to the immediate time beyond discharge, discussing physical equipment needed and advising on community follow-up. These timescapes appeared to reflect what clinicians considered they needed to prioritise and what the care context required them to prioritise to ensure patient throughput and management of high, and often acute, caseloads.

As people moved toward discharge from rehabilitation, timescapes in conversation and action reflected a stronger focus on discharge and beyond, which one clinician called the ‘agenda of discharge’ (Interview, Catherine’s OT), such as ‘meal prep assessment, social supports, rehabilitation referral’ (Clinical notes, Catherine’s OT) and ‘showering, walking around her home, and getting in and out of the toilet’ (Alison’s family meeting records). Interactions centred on assessing, treating, and problem-solving to maximise people’s abilities at home. Within conversations, clinicians intimated about the future beyond discharge. For instance, people with stroke were told that ‘rehabilitation is something that happens when you come into hospital. Here (inpatient rehabilitation) is where we step it up and get more aggressive with it. But it continues once you leave hospital, at home or where you go’ (Observation, Gary’s doctor). This was reiterated in a family meeting: ‘We’re just the first part of the rehab journey. [The community rehabilitation service] picks up where we leave off once you go home. Not many people who come to us with your needs finish completely when you leave us’ (Observation, Gary’s physiotherapist). Conversations about life beyond discharge centred around services that would be provided, with few conversations appearing to attend to the person’s emotions, identity, meaningful activities or roles, or other areas literature suggests are important for life after stroke. Although longer-term rehabilitation and recovery were discussed by staff in interviews, task-focused care and discharge planning were privileged within interactions. This reinforced the focus on short-term futures.

Throughout the continuum of care, some clinicians made active decisions to not talk about a more distant and unknown future. This appeared to reflect assumptions about what inpatient clinicians considered they could and should talk about:

[We are] trying to focus on the transition home and what that would look like because the [community rehabilitation] team are in a much better place to talk about what happens after them … [The person] will always have their [general practitioner] GP and the community team will go through this with [the person], so [we’re] trying, in a way, I suppose, to not to get too far into the future. (Interview, Alison’s OT).

This quote reflects a perspective that other clinicians were better placed to discuss longer-term rehabilitation and recovery. It could also reflect that all parties experiencing uncertainty about the future, with the future outside the control of both the person with stroke and the clinician (e.g. the timing of when things might happen, such as transfer to rehabilitation, and actual outcomes, such as the degree of recovery). In the context of uncertainty, people with stroke sought and valued information, particularly from nurses as they were present around the clock. Yet nurses felt they had limited information about the short- or long-term futures, making it challenging for them to talk about the future. Across disciplines, many clinicians deferred conversations about the future. For example, in acute services, clinicians suggested that the patient and family have conversations in inpatient rehabilitation: ‘You can have more discussions up there [the rehabilitation ward] about discharge and where he will go’ (Observation, Arthur’s OT). Yet inpatient rehabilitation clinicians also deferred conversations, saying ‘[we’re] trying [to] not get too far into the future … the community team are better placed to talk about [that]’ (Interview, Alison’s OT). By deferring these conversations until others could discuss the future, uncertainty was prolonged, and even reinforced, for the person with stroke and their whānau (family, or for Māori, people within wider network who are important to them). This had the unintended consequence of constraining patients’ temporal horizons.

Theme two: limited talk under the control of clinicians

Although discussions about the future were constrained, they occurred in the context of limited communication overall. Planned interactions – ward rounds, regular neurological observations, and therapy sessions – were the primary sites of interaction, with topics shaped by information that clinicians considered they needed to know or share. Medical and nursing interactions followed an unwritten script in which physical function was reviewed before actions that needed to be completed during their shift, or before discharge, were discussed. Clinicians wove discussions about the future throughout their interactions. Examples included a nurse commenting during her regular observations: ‘Did you know the cardiologist is still reviewing you and you have a procedure tomorrow?’ (Observation, Bruce’s nurse) and an OT asking Arthur for his views of his impairments and coping during a session: ‘Do you normally shave like this … how do you think you’d get on at home?’ (Observation). However, these interactions were still initiated by, and under the control of, clinicians.

The person’s future was most overtly discussed in the context of goal-setting; however even then, the areas of life discussed and documented appeared constrained and shaped around the person’s previous function and activities. This reflected an assumptive future in which clinicians appeared to assume previous functions and activities were desired future activities: ‘What we like to do is talk to people about what they used to do and what they want to get back to’ and ‘we want to get people back to doing as much as what they used to do before.’ (Observations, Catherine’s speech-language therapist; SLT). In goal-setting, physical function, rehabilitation needs, and physical and home support needs were privileged, reinforcing the orientation toward discharge: ‘Before we send you home … [we are] making sure that the little things in life flow well. Things like medication.’ (Observation, Catherine’s SLT). When people with stroke identified goals outside of these areas, these were not often integrated into the structured goals of care. For example, Alison identified a goal to ‘cook Christmas lunch’ during a group education session on the ward (Interview), but this remained her goal; it was nowhere in the clinical notes. In an observed session, the OT told her ‘We need a goal for the upper limb’; however, goals recorded in her clinical notes focused on what was needed for her to ‘return home’: ‘to be able to walk independently, to be able to toilet independently, to be able to prepare meals, to be able to dress and shower independently’. The services’ priority of moving people toward discharge was reflected in what was prioritised in conversations and care. Unintentionally, processes designed to identify what mattered could limit what was identified and what was then focused on.

What topics were not initiated by clinicians were notable. There was little observed discussion or attention to the emotional or psychosocial impacts of stroke. This is not to say that staff did not respond to issues when raised, but these were rarely observed to be pre-emptively raised by clinicians. In one instance, a family member initiated a conversation about Alison’s emotions. The OT explained later:

[Her husband] was right, things could change down the line and there might be a realisation that things are different so we pinpointed them to like that there is a psychologist on the team … and then we just talked about the GP as a contact … and we discussed that it’s very normal to happen after stroke and it could happen in the future. Just to be aware to keep an eye on it … but it was really important because it could definitely, you know, once she gets home. (Interview).

Some clinicians reported a hesitation to ‘open the door’ and ‘start those conversations too early’ (Interview, Alison’s OT) for areas that they did not feel equipped to address or they felt could be deferred. Other topics not talked about related to social roles beyond roles within the household, relationships, and hobbies. Despite being recognised as important, these were not considered ‘essential’ in inpatient care. During a family meeting, Alison’s team said, ‘It is a good idea to think about the essential goals when you are an inpatient, and longer-term goals when you are with the community team’. Examples of ‘essential goals’ were given, such as ‘getting in and out of the toilet, bathroom and kitchen’. This was taken up by Alison’s husband, who suggested essential goals might be ‘walking within Alison’s home or perhaps 1 day, walking at [local park]’. Alison’s OT, in an interview, suggested the best place to support social roles and other domains was in the community, reflecting the process of deferral discussed above. There were explicit conversations about hobbies and emotions with only one person, Gary. This occurred in the context of a team who were highly attuned to people’s emotions, regularly considering them in team discussions, and where staff described working intentionally to build trust. That attention to areas beyond physical function is the exception rather than the rule reflects that particular functions and topics are privileged in conversation and may (not necessarily intentionally) limit what is considered in care.

Clinician control over interactions was particularly evident when communication was challenging. Two participants had significant aphasia, another was in infection control isolation, and another had significant fatigue. Interactions required greater intentionality, and sometimes skill, from the clinician. However, these people were rendered as somewhat forgotten patients, with whom staff had limited interactions. For Bruce in isolation, staff worked to do their face-to-face work ‘all at once’ (Interview, Bruce’s nurse), seeking maximum efficiency given the need for personal protective equipment (PPE), in the context of pandemic-related PPE shortages. There were few opportunities for incidental conversations. Catherine was another ‘forgotten patient’ because her communication impairments meant conversations took longer. One nurse commented:

I hope Catherine is not missing out because I’m spending time quickly with other patients to try and balance what I haven’t been able to do what I want to do with her because of the communication barrier. This morning, she just went off and had a shower because I didn’t have time to work out what she wanted. I couldn’t slow down and talk to her because my other ladies were demanding. (Interview, Catherine’s nurse).

Conversations about the future were limited; for many people, conversations in general were limited.

Theme three: opening some doors while closing others

This theme shows that although future-focused conversations were limited and focused on particular temporal horizons, clinicians did hold the door open to some possible futures, addressing the future beyond discharge.

New technologies facilitated different conversations about possible futures. Prognostic testing for upper limb recovery was routinely used in one service. Clinicians took the responsibility of determining and conveying prognoses seriously, mindful of the possible emotional ramifications, and were supported by evidence-based scripts for conveying prognosis. Alison received a ‘poor prognosis’. This was conveyed as:

We’ve had a good look and Alison, I wish we had a bit better news to give you. Unfortunately the tests have shown us the pathway isn’t working for you. What that means, if you remember, is that we would expect you to get some movement back. What that means for your rehab is two things. [We] will work on seeing how much strength you can get back in your arm and learning how to do the fine motor stuff with the left hand. (Observation, Alison’s OT).

The sense of certainty the test provided appeared attractive to Alison and her family. They sought the same information about her walking. Clinicians later commented on how Alison had taken the prognosis, considering it ‘probably skewed her perception early on … she was asking a lot ‘is my leg gonna be like this? If my leg is like this then I’m not gonna walk and then I’m definitely not going home’ (Interview, Alison’s OT). This suggested that conveying prognoses could see people make determinations of what was possible, extrapolating to other areas of recovery, but also orientated them toward their bodily function.

Another way in which clinicians opened doors to possible futures was evident when people appeared to be somewhat stuck in the present. Their temporal horizons were diminished; their sense of possibility appeared to be limited. In these instances, clinicians worked to hold the door open for hope. For Gary, they talked about the importance of keeping ‘aspirational goals’ alive (Observation, Gary’s OT). Clinicians recorded Gary’s dream of driving a specific, special car, but also recorded goals that might be possible in the shorter term, such as going for a drive in the car. Gary’s team tried to maintain an orientation to progress and possibility, keeping that particular door open for Gary and his family. They helped him see progress by focusing on taking one step at a time. Alison, another woman with significant physical impairment, actively problem-solved how she was going to manage at home based on her current level of function, not appearing to consider that this might change. However, her clinicians thought she would have more recovery, with her OT saying ‘It was part of our job to kind of instil that hope for her but still be realistic, but I guess we can see the potential in her’. The same therapist described trying to plant seeds as to what might be possible, opening up possibilities and warding off hopelessness: ‘She’s got it in her mind about getting a wheelchair for longer distances but I’m just trying to show her that with time and with the community input, she probably will get back to walking’ (Interview, Alison’s OT). In this case, the door was held conditionally open to what clinicians considered was realistic. The two teams focused on different future possibilities: Gary’s team on what brought joy and meaning, and Alison’s team on specific activities, such as walking. Through their interaction, staff adopted ‘holding the door open’ as a therapeutic strategy, employing hope and possibility as strategies to achieve the goal of patient engagement in rehabilitation.