Research design and methods

This study was informed by an epistemological position which recognises that illness experiences and, relatedly, treatment decision-making are socially and contextually informed. [34] This position informed our approach to both data collection and analysis. For instance, we developed topic guides (Table 1) which allowed us to explore what patients’ lives were like before suffering a stroke. We also explored their experiences soon after their admission in order to set the context for understanding the treatment decisions which were subsequently made.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Guide to the topics covered in patient interviews.

https://doi.org/10.1371/journal.pone.0222500.t001

Initial interviews took place within a week of a patient’s admission to the stroke unit. This time point was chosen to capture their early experiences after a disabling stroke. Also, we recognised that most treatment decisions that we were interested in exploring (such as initiation of enteral feeding) should have been taken within a week of admission.

Where possible, we undertook follow-up interviews at six months following first admission to the stroke unit. We chose this time frame because we recognised that, by six months, most patients would have plateaued with respect to their functional recovery.[35–37] Had we interviewed them at an earlier point in time (for example, at three months following first admission to the stroke unit), their recovery might have still been ongoing. By contrast, had we interviewed at a later point in time (for example, at one or two years following first admission to the stroke unit), this might have meant that other factors (for example, increasing age and frailty, recurrent stroke, declining cognition or death) might have adversely affected patients’ ability to be interviewed.

Recruitment and data collection

We recruited adult patients who had been admitted to a stroke unit in a large teaching hospital in the United Kingdom. Our aim was to explore decision-making regarding stroke treatments that increased the likelihood that the patient who has been left with significant disability as a result of stroke will survive longer; to be eligible to participate, the stroke needed to have caused significant physical disability and the patient needed to have had decisions made regarding treatments (such as enteral feeding, parenteral fluids, antibiotics or intermittent pneumatic compression). Therefore, we recruited patients whose extent of disability was at least a modified Rankin scale score of four as a result of the stroke (i.e. they were unable to walk or attend to their own bodily needs without assistance).

The medical team (consultants, registrars and other trainee doctors who looked after the patients on a daily basis) identified eligible individuals who had the capacity to consent to participating in the study (i.e. the patient was able to understand and retain information and communicate decisions). If the patient’s speech or swallow was affected, the medical team assessed them to ensure this did not affect their decision-making capacity and that they were able to fully participate in an interview. In each case, before approaching the patient regarding in this study, the medical team also considered whether participating might cause distress.

Once the medical team determined that the patient was suitable to participate, they asked the patient for permission for the researcher (AV) to approach them and provide further information about the study. AV was a clinical doctor specialising in geriatric medicine who had previously worked in the stroke unit. However, patients were not informed about AV’s clinical background and they were advised that she would not be able to provide any treatment advice or medical information specific to their care.

AV then provided information about the study and obtained the patient’s formal written consent to participate. We made every effort to recruit patients of different ages, genders and ethnic backgrounds. However, we recognised that patients with an acute disabling stroke are a group who are hard to recruit: many patients are medically unwell and the impact of the diagnosis was noticeably distressing to some of them. Therefore, we made a pragmatic decision to interview all eligible individuals who had agreed to take part. As the clinical team did not keep a log, it is not possible to report the total numbers approached and, of those, how many declined to participate.

We endeavoured to recruit sufficient participants to address our study aims while not creating a dataset which would be too large and unwieldly to analyse in-depth. In practice, this meant that we stopped recruitment after on-going analysis of the interviews indicated that sufficient data had been obtained to address the study aims and that a point had been reached where patients were volunteering similar views. Hence we recognised that recruiting additional patients would not enhance the quality or diversity of the data collected.

Prior to the start of each interview, AV reassessed patients’ capacity to participate. This was done because we recognised that patients’ capacity might have changed from when they had initially been assessed as suitable to take part. Initial interviews were conducted in a private room in the stroke unit at a time convenient to the patient. Table 1 summarises the main areas explored. Where possible, we conducted six month follow-up interviews in the patient’s place of residence (i.e., either own home or care home). Telephone interviews were considered where the multidisciplinary team felt that it would not be safe for AV to visit the patient. Before contacting the patient to arrange the interviews, AV phoned their general practitioner to check the patient still had capacity to participate. This was part of the consent process at recruitment. Prior to follow-up interviews, AV reassessed the patient’s capacity. We summarize the main areas explored in these interviews in Table 1.

AV kept a reflexive diary. This detailed her interactions with the participants, including a record of why, based on what patients said in interviews, she had decided it would be insensitive or inappropriate to pursue certain lines of questioning. For instance, when AV explored patients’ feelings about potentially living with disability, only physical aspects of disability were discussed (for example, immobility and inability to perform activities of daily living). This was because recruited patients were assessed to have physical disability only, and exploring irrelevant disabilities (such as cognitive or intellectual) may cause the patient unnecessary worry. Also, as we will go onto describe, it became apparent to AV that, in the early stages post stroke, some patients did not appear to understand that the goals of some stroke treatments were to extend their life but they will be significantly disabled and that decisions needed to be taken regarding these treatments. Making this known to them either acutely or six months later could have caused undue distress.

Initial interviews took place between September 2017 and January 2018 and lasted 25 to 66 minutes. Six month interviews took place between April 2018 and July 2018 and lasted 32 to 56 minutes. All interviews were digitally recorded and transcribed in full with the patient’s consent.

Data analysis

AV (who has received training in qualitative methods, including qualitative data analysis) and JL (a very experienced, non-clinical qualitative researcher) undertook data analysis. All data were analysed thematically using the method of cross-comparison.[38] This approach entailed repeated read through of all interviews to allow familiarization with the data (immersion). Interviews were then cross-compared to identify key findings which cut across different accounts (themes). Both inductive and deductive approaches were used; this allowed unanticipated themes to emerge from data as well as identification of material needed to address the study aims. Data were also analysed longitudinally to establish whether, and why, peoples’ needs and views changed over time. AV and JL analysed the data separately and wrote separate reports. They then met to discuss their interpretations, resolve any areas of disagreement (which were found to be minimal), and reach agreement on the main findings and themes. A coding frame was then developed which captured these findings and themes. Coded datasets were subjected to further analysis to allow development of more nuanced interpretations of the data and identification of illustrative quotations. Nvivo 11, a qualitative software package produced by QSR International, was used to facilitate data coding and retrieval.

Ethical approval

This study was approved by Scotland A Research Ethics Committee (Ref: 17/SS/0029).

To safeguard participants’ confidentiality, pseudonyms are used.

Pre-stroke background and early reactions to diagnosis of stroke

Trying to cope with the diagnosis: Generating and sustaining hope

Regardless of their pre-stroke functional status, all patients described looking for hope of functional recovery. This included Nigel who described how: ‘I am, I can and I bloody well will: where there’s breath, there’s hope’ and Graham, a man in his 60s who, prior to the stroke, owned his own business, and enjoyed outdoor activities, who described how:

(Re: recovery) ‘I’ll say [I am at] about 20% now. I can see the future higher up the ladder is there for me to climb up to it.’ (Graham)

In-keeping with their wish to maintain a positive outlook, patients reported using different coping mechanisms to generate hope. Some reported comparing their functional state to that of other patients in hospital, and described how they thought that others were less fortunate than themselves as they had impairments that they felt were more severe. Such patients reported how this fuelled their hope that they may return to their pre-stroke state:

‘When I was sitting in bed and watching the others, I kept thinking I’m lucky I’ve not got it as bad as they have, you know. I can swallow but I’ve just got to watch not to take a big mouthful.’ (Dorothy, a woman in her 70s, who was unable to mobilise or independently self-care post stroke)

Others described how they valued speaking to stroke survivors they knew who had made full functional recovery and been discharged home. Indeed, such patients described seeking out stroke survivors, often with help from their family and friends, and welcoming hearing encouraging stories about their recovery:

‘That was what she [friend] went through. I mean, I know that she’s okay. She’s a hundred per cent recovered but she was in a bad way. So that was good to understand that, just that there is light at the end of the tunnel, if you like. So that’s a positive experience, positive discussion I’ve had with someone who I know and I know she’s not lying to me. She’s looking good. She talks good. She’s not got any signs of a stroke.’ (Graham)

Most patients also described how they needed information that was framed positively in order to help them maintain an optimistic outlook. In keeping with their need for hope, patients also described not wanting to engage with information that was ‘negative’; that is, which alluded to a lack of recovery or the possibility of living with significant disability. Many, like Graham, described even being prepared to accept information that was not necessarily correct, as long as it gave them hope:

‘I want to know I’m going to get back to a hundred per cent; that’s what I believe inside that I’ll get back to. I think it’s vital to move forward, even if it’s…I was going to say even if it’s not completely true, but you’ve got to have a positive outlook. But if you come along and say, well you’ve got no chance, you’re going to be, you know, wrapped up in a wheelchair for the rest of your life, that’s not going to help me.’(Graham)

Indeed, some described how they might lose trust in health professionals if given discouraging information. For example, Olivia, a woman in her early 80s recalled a situation where potentially dispiriting information would have been unhelpful:

‘Well if somebody had said to my friend who died of the stroke, oh this is a very serious condition you’ve got yourself into, I don’t know if we’ll be able to get you out of this, well it wouldn’t have done him any good at all would it? I mean, where’s the trust that staff are helping?’(Olivia)

Treatment decision-making following diagnosis

It was against the above backdrop, that decisions about treatments such as intermittent pneumatic compression, enteral feeding, parenteral fluids and antibiotics were made.

Six month follow-up

Retrospective views regarding information in the early period after the stroke

When patients were asked to consider what would have been useful for to them to have known before they were discharged from hospital, all patients, regardless of their pre-stroke functional status, reported their wish to have been given information which would have helped them better understand and prepare for the situation they were now in. For instance, Irene, in her current state, was in a care home and required hoist for transfers and a catheter to manage urinary incontinence. She reported:

“Well, I think we wanted more information, you know. Excuse me. Just to understand why we’re like this, you know.” (Irene)

Likewise, Brenda who was now needing carers to help her mobilise to the toilet and to shower, described how she wished she had been given information on what her future may have looked like:

‘An idea. You can’t precisely say when it’s going to happen, an idea of what image of what’s ahead.’ (Brenda)

However, while these patients described their wish to have had information in hospital, they were not specific about the type and timing of this information.

Reactions to diagnosis- and the need for psychological support

Our study provides empirical support for recommendations provided by the Stroke Association [3] and Harrison et al [41], that stroke survivors may benefit from psychological support on discharge from hospital. Psychological support has been shown to be beneficial in patients with dementia [42], cancer [43] and myocardial infarction [44] because it reduces rates of depression, improves patients’ ability to cope with their situation, better optimises patients’ remaining abilities and improves their quality of life. This would be an important consideration for stroke patients, who, as we observed, had unmet psychological needs which seemed to contribute to a poor or suboptimal quality of life. In addition, our findings suggest that the type and amount of support required by patients who had suffered a disabling stroke may depend upon their functional status prior to the stroke and may also change over time. Therefore, different forms of psychological support (e.g. counselling[45], support groups [46], clinical psychology [47] and be-friending services[48]) may be appropriate for different patients at different time points. For example, those who had been functionally independent prior to the stroke may initially benefit from attending clinical psychology services and emotional support groups in order to come to terms with their diagnosis and loss of independence. Such patients would also benefit from be-friending services and peer support in the longer term to address social isolation as would patients who were functionally dependent prior to the stroke. A stepped care approach to psychological care after a stroke (that is, where patients are identified and treated and ‘stepped up’ to more intensive treatments based on their clinical need) was proposed by the National Institute of Health and Excellence in 2011 [49,50]., Although there are pathways into psychological services for stroke patients in the U.K, [41,51], patients in our study did not seem to have benefited from this. We recommend that the multidisciplinary team in hospital, in collaboration with specialist neuropsychology services if appropriate, assess the type and amount of psychological support each patient needs prior to hospital discharge and collaborate with community services to assess ongoing patient needs.

Involving patients in making treatment decisions

Similar to acutely unwell patients in the intensive care unit [18], we also found that it was not possible to involve some patients in treatment decision-making in the early period after their stroke. This was especially the case for those who had been functionally independent prior to their stroke. As we have shown, such patients may not appreciate that, in the early stages post stroke, some treatments might extend their life, but with significant disability. They may also struggle to engage with health professionals in order to make treatment decisions in the early stages post-stroke. We have also highlighted how the emotional impact of a life-changing diagnosis and patients’ need for hope may not be conducive to them receiving and understanding realistic information, rationally weighing up the pros and cons of treatment and/or expressing their preferences for these treatments [16,18]–these all being key to effective shared decision-making as described early on in this paper. Trying to involve these patients in making collaborative decisions may cause unnecessary psychological distress. [52] By contrast, patients who were already significantly disabled before the stroke, like frail older patients with chronic progressive conditions, may be mentally prepared for further deteriorations in their health, have considered consequences of different treatment options and have often decided not to have treatments that may prolong an already unsatisfactory quality of life. [53–56] Hence these patients may be in a better position to be more engaged in the treatment decision-making process. We recommend that health professionals explore patients’ pre-stroke functional status and consider if and whether the emotional impact of diagnosis may prove it challenging to involve them in making treatment decisions.

The hope- information paradox

The need for hope which we have reported in this paper is not exclusive to stroke patients: for cancer patients and older patients too, communication of hope by health professionals is said to help such patients adjust to their diagnosis and improve their welfare and quality of life. [21,57,58] Striking a balance between providing a patient hope that s/he will functionally recover while not providing false hope is challenging. [59] We report two further dilemmas for health professionals. First, shortly after their stroke, some patients volunteered that if their doctor were to provide them unfavourable information regarding the likelihood of their recovery at that point, they would lose confidence in their doctor, thereby detrimentally affecting the doctor-patient relationship. Second, when asked at six months to look back to the period shortly after their stroke, patients described wishing that their doctors had given them realistic information that could have prepared them better for their current (unfavourable) functional status.

While there are no easy answers with respect to resolving the “hope-information paradox”, one potential solution is to consider using existing cancer communication strategies in the stroke context. Potentially relevant cancer communication approaches include ‘Ask-Tell-Ask’ and ‘Hope for the Best, Plan for the Worst’ approaches. For example, under the ‘Ask-Tell-Ask’ approach, a doctor would communicate poor cancer prognosis, then respond to the patient’s emotions and finally transition to talking about next steps. Under ‘Hope For The Best, Plan For The Worst’ approach, a doctor would join with the patient in embracing their hopes while simultaneously asking them to explore a back-up plan based on their prognosis. This could help cultivate the patient’s hope by seeking to understand their diagnosis and prognosis and thereafter re-orient the patient’s care based on the patient’s goals and objectives. [59] Yet there are some challenges to directly adapting cancer communication strategies for using in the stroke context: the illness trajectories for cancer and stroke are different. [59] Specifically, relative to cancer, stroke is an acute medical condition, many treatment decisions need to be taken early, and the trajectory of the patient’s condition is difficult to predict.[60] By appreciating that stroke patients too, like other patients, may require different types of information at different points in their illness trajectory,[20,61] health professionals may consider reassessing stroke patients’ need for information at different points during their hospital admission.

While it is difficult to make concrete conclusions from our small study, we recommend doctors consider the following:

1. Exploring the social context and early experiences of patients to gain an understanding of their views and needs.
2. Being aware that patients may need psychological support, and that this best assessed prior to hospital discharge.
3. Being aware that, while guidelines exist, they may not apply to acute stroke patients: a shared decision-making approach may not always be appropriate.
4. Adapting strategies used in cancer when communicating hope but maintaining realism for patients who have had an acute disabling stroke.
5. Assessing the information needs of acute disabling stroke patients at various points during their hospital admission.

Strengths and limitations

We were successful in engaging and interviewing a group of patients at a time when they are often ‘hard to reach’ and therefore excluded from research. Following up these patients at six months has given us important insight into their ongoing (and changing) needs for information and support. However, due to the impact of a disabling stroke on patients’ physical and psychological states, it sometimes proved challenging to probe and explore some of their experiences and views in-depth. Our sample size was relatively small and all participants were recruited from one tertiary teaching hospital. This reduced the diversity of our sample and hence, potentially, the transferability of the findings to other populations. [62] Patients in this study had capacity: our findings are therefore not applicable to those without mental capacity.

Recommendations for further research

Future researchers could consider investigating the views and needs of stroke patients from different socio-economic and ethnic minority backgrounds who may have different information and support needs. Interviewing doctors who look after severe stroke patients could give us important insight into how, and why, doctors may have made decisions on behalf of the patient or what other factors they may have considered in the context of treatment decision-making.