Introduction

The intensive care unit (ICU) is a distinct environment that occupies a central role in modern healthcare systems, due to the extensive human and technical resources deployed and the relatively small proportion of patients that benefit from this very specialized care. In the prevailing context of scarce resources, and limited availability of beds [1, 2], ICU physicians need to ensure equitable use of ICU resources, and attribution of those resources to patients in a manner that is commensurate with the patient’s healthcare goals and wishes (if and when these latter are known) [3, 4]. The right balance can be hard to find, especially in acute situations that were neither envisaged nor anticipated, and thus, were not discussed with the intensivist in advance. Such situations can lead to patients receiving care that the ICU staff ultimately deem to be “futile” or “disproportionate” [5, 6]. This can also cause suffering in the workplace for the nursing staff [7, 8], and/or cause them to lose motivation, prompting them to change units or even professions [6]. It is often difficult for physicians to estimate a patient’s prognosis when considering admission to the ICU, especially in emergency situations, where full information about the patient’s medical state may not be available, the family are not present or cannot be contacted, and the intensivist has to make a choice between a potential loss of opportunity for the patient if admission is refused, and the risk that admission may ultimately turn out to be non-beneficial for the patient [9]. Consequently, admission to the ICU of a patient whose healthcare goals are unclear, or of a patient who explicitly stated their desire NOT to be admitted to ICU, can ultimately lead the patient’s ICU stay to be considered “non-beneficial” [3, 10, 11]. The potential consequences of this include stress and anxiety among both patients and their families. Sometimes, there can even be some degree of incomprehension, which may sometimes escalate into open conflict with the healthcare team, and healthcare professionals need to be able to respond and adapt appropriately in such cases [12].

In order to anticipate such unfortunate situations and attempt to avoid their occurrence, the intensivist needs to be able to reach beyond their traditional sphere of practice, and be involved in defining the healthcare goals for patients before the occurrence of acute clinical events that may require ICU admission [13, 14]. It seems legitimate to involve ICU specialists, as they can provide expertise regarding critical care, the potential benefits of ICU admission, as well as the possible repercussions for patients and their families, in terms of physical, mental and cognitive health [15, 16].

In this study, we performed semi-structured interviews to explore the perceptions of nursing staff in the ICU about patient stays that are deemed to be “non-beneficial” for the patient. Through the prism of their experience, we sought to identify areas amenable to intervention, with a view to improving how the nursing staff perceive the patient pathway before, during and after intensive care.

Methods

We performed a multicentre, qualitative study by means of individual, semi-structured interviews. All qualified nurses and nurses’ aides who were full-time employees in the ICU of the three participating centres (one academic teaching hospital and two general, non-academic hospitals) at the time of the study were invited to participate by personal invitation (telephone and/or email). These invitations were either sent directly, or via the hospital administration by email. Respondents were interviewed either face-to-face, or over the phone. Trainees or students were not eligible for participation. Participants were contacted by one researcher (LM; male, resident) to organize an individual semi-structured interview at a convenient time. Interviews were performed between February 2020 and October 2021 by a qualitative researcher (LM) and a sociologist with experience in clinical research and ICU care (NMB; male, PhD). There was no prior relationship between the sociologist (NMB) and the interviewees. The qualitative researcher (LM) was a medical resident in the same ICU as the interviewees, but was not hierarchically superior, thereby ruling out any potential for undue influence. To achieve maximum variation in the sample, we took account of age, total number of years’ experience since qualifying, number of years’ experience in the ICU, and the type of hospital (university teaching hospital vs non-academic general hospital).

We developed an interview guide using methods that have previously been described elsewhere [7, 17, 18]. The interview guide focused on 4 main topics, namely: (1) The mission of the ICU; (2) eligibility for ICU care (objective criteria / subjective criteria / healthcare goals / clinical status); (3) “non-beneficial” admissions; and (4) the consequences of a “non-beneficial” stay in the ICU. The interview guide was pilot tested with 3 nurses from our department; the findings from their informal interviews were not included in the analysis. The pilot interviews did not give rise to any major changes in the interview guide.

During the semi-structured interviews, the questions were open ended. The interview guide serves to prompt the respondent to talk about the aspects most important to them, in their own words. Interviews were performed in a dedicated medical office or by phone. No other persons were present during the interview. Participants were informed that quotes from their interviews might be used to substantiate the results in scientific publication, after translation into English. All participants provided written informed consent to participate in the interviews, and provided written informed consent to the use of translated quotes from their discourse in future publications. The Committee for Ethics in Research of the University of Burgundy Franche-Comté confirmed that the study did not require formal approval in compliance with French legislation.

All interviews were recorded and transcribed in their entirety for later analysis. Data were encoded to guarantee the anonymity of the participants. The discourse from the interviews was analyzed using thematic analysis as previously described by our group elsewhere [7, 17, 18]. Briefly, interviews were coded independently by 2 coauthors (LM, NMB), to identify categories, and consensually agree on themes. The different themes that arise during the interviews are noted, and classed as major themes (significant points that are of major importance and well developed by all participants) and secondary themes (less well developed by the participants). The first round of analysis was done individually by each researcher, then meetings were held to harmonize and decide on the major and secondary themes to be retained, and the optimal grouping of themes. Differences in interpretation were resolved by discussion and consensus. No software was used to assist with data management.

The analysis was validated by 4 authors (LM (reference author, male), NMB (male, sociologist, PhD), FE (female, clinical researcher, PhD), JPQ (male, critical care physician, MD, PhD). The final report was written by the same 4 authors, and approved by all. Results were not returned to the participants. Interviews were conducted in French, and translation was performed after the results were finalized. Interviews were conducted until saturation was reached (i.e. the point at which new interviews failed to bring forth any new elements on any of the points in the interview grid) [19].

Results

A total of 28 nurses and 4 nurses’ aides were contacted, of whom 21 (65%) agreed to participate in the interview. Two nurses refused due to a lack of interest, and 9 persons did not respond despite reminders. A total of 21 interviews were performed from February 2020 to October 2021, at which point saturation was reached in the data, and thus, no further interviews were performed. The average age of participants was 38.5±7.5 years. They had an average of 14.5±8 years of experience since they qualified, and an average of 10.7±7.4 years of experience working in the ICU. The average duration of the interviews was 52.7± 12.5 minutes. The characteristics of the study population are shown in Table 1.

Download:

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

Table 1. Characteristics of the study population.

https://doi.org/10.1371/journal.pone.0289954.t001

Four major themes emerged from the interviews, namely: (1) the work is oriented towards life-threatening emergencies, technical procedures and burdensome care; (2) a range of specific criteria and circumstances influence the decisions to admit patients to ICU; (3) there are significant organisational, physical and psychological repercussions associated with a non-beneficial stay in the ICU; (4) respondents made some proposals for improvements to the patient care pathway.

Illustrative quotes for all the themes are summarized in Table 2.

Download:

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

Table 2. Illustrative quotes for each theme.

https://doi.org/10.1371/journal.pone.0289954.t002

Discussion

The goals of ICU admission are to offer the patient care that will not only save their life, but also guarantee an acceptable quality of life after discharge. Indeed, there are undeniable physical, mental, spiritual and financial consequences, both during and after a stay in the ICU, and affecting the patients, their families and society in general [15, 16, 20–22]. In this context, feedback should help to enhance the relevance of ICU admissions [23].

In our study, the key points that emerged from the interviews with nursing staff were as follows: first, the ICU is a distinct unit, different from other, conventional units in the hospital, because it requires more human and technical resources; second, the patient’s medical history, their healthcare pathway and healthcare goals are all important features to be taken into account when deciding on admission to the ICU; third, the intensivists’ experience on the job, their fear of making a wrong decision, moral pressure from external colleagues or from the family, refusal of death, or refusal to give up, are all factors that may modify the physicians’ perception of an ICU admission. Fourth, stress, anxiety and a failure to find meaning in their caregiving work, not to say a feeling of mistreatment or inequity regarding access to intensive care, were all mentioned as points likely to raise problems for caregivers with regard to patient stays that are “non-beneficial”. Fifth and finally, there is a compelling need to clarify healthcare goals prior to ICU admission. Developing a healthcare project that may include ICU admission requires the participation of the patient and/or family, but also, of all the caregivers in charge of the patient. This could help to limit or avoid admissions that come to be deemed “non-beneficial”, thereby also avoiding the associated consequences. Indeed, we previously reported that a lack of knowledge of the patient’s wishes, and inability to contact the family can result in ICU stays that later come to be deemed “non-beneficial” by the physicians [10]. Furthermore, non-beneficial stays may have multiple negative consequences, including stress and anxiety for the patient/family, misunderstandings and conflict with the caregiving team. In this regard, careful discussion about the possibility of future ICU admission in the framework of the patient’s overall healthcare goals has been hailed as a means to prevent non-beneficial admissions [9].

Careful reflection is a process that can undeniably improve the patient’s healthcare trajectory, as long as this reflection includes anticipation of the possible need for ICU care. Open discussion of the healthcare goals, before ICU care becomes necessary, should be systematic, and should involve the patient, the family and the caregiving team (medical and paramedical). Involvement of nurses in end-of-life discussions, a climate of trust and good relationships between nurses and physicians in the workplace, as well as intense communication have all been shown to improve the experience of nursing staff vis-à-vis end-of-life decisions and appropriateness of care. In addition, these factors also help to reduce intent-to-leave among ICU nurses [6, 8, 24]. However, some limitations to this approach have been highlighted in the literature. Firstly, physicians outside the ICU often find it difficult to accurately estimate prognosis of their patients [25–27]. Second, physicians sometimes lack sufficient perspective regarding the patient’s clinical status, prior history, potential for recovery, prior or expected future quality of life, and life story [28–30]. Third, it can be challenging to inform patients and families about potential acute deteriorations that may or may not occur [31]. Finally, there is a pervasive lack of knowledge about the specific forms of care that can be offered in the ICU [32], and likely also, a non-negligible contribution of individual emotional factors [33].

There is thus a need to rely on the knowledge of caregivers in the ICU about the consequences of an ICU stay, which depend on multiple factors, including patient characteristics such as age, comorbidities, or chronic disease [34]; frailty [35, 36]; socio-economic conditions [37, 38]; the motives for admission to ICU and the complications that occur during the ICU stay [39]. Conversely, the ICU physician, if contacted sufficiently early during the patient’s healthcare pathway, can be a key consultant (not to say the best qualified), to guide decision-making about admission (or not) to intensive care. There are several reasons for this [40]. First, the intensivist knows best what treatments (including life support therapies) can be offered in the ICU. Second, the intensivist is capable of evaluating a patient’s prognosis based on the existence or not of organ failure [24, 41]. Third, the expertise of the intensivist is essential to inform and explain to patients and their families, the detailed indications for intensive care, as well as the potential limits and consequences of ICU admission [15, 16].

In several English-speaking countries (including the UK, USA and Australia, amongst others), a process called “advance care planning” (ACP) exists, in addition to written advance directives [42]. ACP brings together the patient, the family and physicians, prior to hospitalization, in a dynamic and continuous process to define medical care that is consistent with their personal values, life goals, and preferences. It is a proactive and anticipatory approach, intended to facilitate decision-making in emergency situations, or in situations where the patient is no longer capable of clearly expressing their wishes; both often precede ICU admission. Ideally, ACP should incorporate advance directives, when they exist, but can also prompt the patient to think about writing advance directives, if none have been written. In addition to ensuring care that is in line with the patient’s preferences, numerous other benefits of ACP have been touted, especially among patients followed for chronic diseases [43–45]. It has been shown that implementing ACP increased the rate of preparation of advance directives, yielded a better balance between the care dispensed and the patient’s wishes, as well as improving communication between patients, families and healthcare professionals [4]. Nevertheless, some limitations have also been highlighted, notably the timing of ACP vis-à-vis the onset of disease, training of physicians in ACP, the difficulties for physicians to discuss death and end-of-life issues with their patients (also found in the present study), and the lack of knowledge among patients about the progressive nature of their chronic disease [44, 46].

This study was carried out during the COVID-19 pandemic, which may have modified the perceptions of the participants due to the crisis context, especially at the peak of the various waves, when ICU activity was at its most intense and stressful. We have previously reported that the pandemic led to a heavy excess burden in terms of physical work, but also in terms of mental repercussions for healthcare staff [47–51]. Indeed, healthcare workers reported psychological suffering, especially when there were organisational difficulties in the workplace, when they were working in close proximity with the patients, when they were female, and when they had to train support staff in addition to their ordinary work [47–51]. However, the unprecedented numbers of people dying during the early phases of the COVID-19 pandemic has shone a spotlight on end-of-life issues, and raised awareness among healthcare professionals and the public about the need to hold end-of-life conversations [52].

In terms of perspectives, organising pluri-professional consultations (involving physicians and nurses) after discharge of a given patient from the ICU could help to identify areas of uncertainty, and promote ethics review of the relevance of admissions (or re-admission) to ICU [53]. Similarly, the use of telemedicine could also help to better identify patients likely to benefit from intensive care [54, 55]. These avenues for future research and reflection, involving the participation of caregivers in the definition of healthcare goals, including the question of ICU admission, need to be addressed going forward. They are likely to contribute to recognizing and acknowledging the work of nursing staff, thereby improving quality of life at work, particularly the potential negative psychological impact of ICU stays that rankle, because they were “non-beneficial” [56, 57].

Study limitations

This study has some limitations. First, as with all qualitative studies, there may be potential for desirability bias in the responses, especially when addressing questions of ethics. However, the potential for this bias was minimized by asking respondents to describe concrete examples of situations they personally experienced. Second, the interview guide may have not have encompassed the full spectrum of issues relating to non-beneficial ICU stays. We chose to limit our target study population to nurses and nurses’ aides. Consequently, the findings reported cannot be extrapolated to other professional groups working in the ICU. The number of participants may seem low, but is deemed to be sufficient since saturation was achieved [7, 17, 18, 58].

Conclusion

Nursing staff have a similar perception to physicians regarding admission decisions and non-beneficial ICU stays. The possibility of future ICU admission needs to be anticipated, discussed systematically with patients and integrated into healthcare goals that are consistent with the patient’s wishes and preferences, in a context of pluri-professional collaboration, collegial and multidisciplinary decision-making.

Supporting information

Acknowledgments

We are grateful to all the nurses and nurses’ aides for their participation.