In routine medical practice, the most commonly encountered situation is that of a patient who is conscious and capable of receiving relevant information about diagnosis and management, integrating that information, understanding its implications, consenting to care and complying with the proposed therapy or exams. Ideally, the patient’s representative or surrogate would be an integral part of this process.

Despite the existence of a legal framework, difficulties arise in situations of uncertainty, such as unconscious patients, persons with dementia, emergencies, psychiatric cases and patients with neurological disorders. All of these situations are frequent among patients admitted to the intensive care unit (ICU), and several of them may co-exist in a single patient. Admission to the ICU is a situation where it is only rarely possible for the patient to exercise their right to autonomy. In the vast majority of cases, at admission to the ICU, the patient is unconscious, under sedation, or their life is in the balance, and they are therefore quite unable to express their wishes or make decisions. In addition, the rapid decisions about urgent treatment initiation that such situations often require are not compatible with the time-consuming process of trying to find out what the patient’s wishes are.

Consequently, in such situations, the information is often provided to the family, or to the surrogate if one is designated, and to the patient’s general practitioner or other treating physicians. Despite generally being perfectly willing to provide all the relevant information, the intensivist is often prevented from doing so, because the admission may be an emergency or unexpected, or there may be no officially designated surrogate, and sometimes even no family members at all present at the time of admission. Frequently, at admission or during the ICU stay, the family are incapable of assimilating and understanding medical information, because of the anxiety and worry caused by the serious, not to say life-threatening illness that has suddenly beset their loved-one. Most of these aspects are beyond the direct control of the intensivist, and may not change in any substantial way during the patient’s stay in the ICU.

It is important to remember that a surrogate decision maker only exists if the patient him/herself has designated one. Neither the caregiving staff nor the family can designate an official surrogate if the patient has not done so. Therefore, if there is an officially designated surrogate, then medical information can be given to this person, and this person only. The physicians can carry out their job within the framework laid down in the legislation, and there are no limitations on the form or content of medical information that can be given, except if the patient has expressly opposed the transmission of certain information ^[1][5]. By designating an official surrogate, the patient establishes the type and extent of information that can be transmitted. However, this means that the physician caring for the patient has to accept that the designated surrogate is the “right” person, and the only person capable of receiving medical information, and of relaying and respecting the patient’s wishes. This situation is relatively straightforward when an official surrogate for the patient has been designated prior to admission. However, what should you do when there is no officially designated surrogate, a situation that is encountered almost daily in routine ICU practice? Or when an official surrogate has been designated, but does not have the patient’s best interests at heart, or the best interests of the family? In addition, how should one deal with the patient’s relatives, who are naturally worried and eager for information? It can happen that patients do not choose the most appropriate person as their surrogate, especially as regards the sharing and communicating of medical information. In a study about the information provided to families when further tests are required for a patient in the ICU ^[2][20], it was shown that most physicians almost systematically refer to “the family” or “the relatives” in a general way, and not specifically to “the surrogate”. In reality, while the legal provisions for an official surrogate exist in France, very few people actually have an officially designated surrogate, and therefore, the presence of “a surrogate” is quite rare in French ICUs. The reasons for this are manifold: emergency admissions for acute and unpredictable situations, failure to designate a surrogate in advance, as well as lack of awareness among healthcare professionals and the general public about the possibility to designate a surrogate. It is not a intention to call into question the principle of surrogate designation simply because it is difficult to apply in the context of critical care. While the confines of the 2002 law about patients’ rights and end-of-life issues are strict, it is almost certainly possible to take into account the specificities of ICU patients.

It is logical to assume that the people who are at the patient’s bedside in the ICU are the family and close relatives (or people whom the patient would consider as such). These persons surely already possess a great deal of information about the patient’s prior health status, notably the pre-hospital course, the patient’s previous state of health at home, etc. Sometimes, the person(s) present at the bedside were also there when the doctor or the emergency services came to treat and transfer the patient. So, in general, they do know quite a lot about the patient and can be a precious source of information when the patient is admitted to the ICU. Given that the majority of people understand that admission to the ICU is synonymous with a very serious situation, it is also logical to assume that the persons present are already aware, to a certain extent, of the patient’s state, before they ever meet with the ICU physician.

When no official surrogate has been designated, which is the case in the vast majority of ICU admissions, it is useful to identify a “reference person” among the relatives who can act as a surrogate for the patient. This reference person must meet the caregivers’ expectations ^[3][21]. The physician should be able to identify, among all the relatives present, which of them is the best suited to act as surrogate, the most benevolent, perhaps the closest to the patient, and in any case, the most capable of receiving and understanding relevant medical information. The choice is far from being easy. Close relatives, especially the patient’s spouse, are usually the most available, and often the most capable of understanding the medical information and reporting the patient’s wishes ^[4][22]. However, the sudden hospitalization of their loved-one may cause them acute psychological distress, and prevent adequate comprehension or cloud their judgement. They may even require specific support themselves to manage their distress ^[5][6][7][18,23,24]. Indeed, it has been shown that those who struggle most in the role of surrogate decision maker are those who lacked prior experience as a surrogate and those who had no prior discussions with the patient about treatment preferences ^[8][25]. Accordingly, choosing a surrogate from among the relatives should not put that surrogate in difficulty, either personally, or vis-à-vis the other members of the family. It is also not acceptable to exclude any members of the entourage or family from the communication, once a surrogate has been chosen from among the relatives. On the contrary, it is equally impossible for the ICU physician to provide information to all the relatives, every day ^[2][3][5][9][18,20,21,26]. The involvement of families in decision making for critically ill loved ones is a vast and complex subject that has been widely discussed in the literature. A group has previously reported on the challenges of obtaining consent in the ICU setting ^[10][27], while Azoulay et al. have extensively reported on the challenges of involving families in the care of, and decision making for critically ill patients in the ICU ^[6][11][12][23,28,29]. In these studies, it was shown that participating in decision making for critically ill loved ones increases the family’s satisfaction, but also that a substantial number of families do not wish to be involved in decision-making processes. Indeed, making decisions against a background of the psychological stress caused by the sudden hospitalization of a loved-one can cause significant distress and be an unwelcome burden for some. Therefore, any reluctance on the part of the family to participate in decision making must be heeded by the ICU physicians.

To cater for these complicated situations, it could be of interest to propose that a key contact person be chosen among the family members using a collegial decision-making process. The family, general practitioner, ICU staff and physicians could meet to designate the person to serve as the surrogate, according to a process that remains to be determined, but which could be modelled on the “family conference”, for example. Such a procedure would guarantee the legitimacy of the surrogate, avoid problems with the other family members, and perhaps also avoid having surrogates who struggle with their new responsibility ^[3][21]. There could even be two or more surrogates designated to share the responsibility. This type of procedure would not interfere in any way with the official designation of a surrogate by the patient while still in good health, but would be nonetheless well suited to the specific challenges of the ICU context. Similarly, as discussed below, an exception to the rule of medical secrecy, limited to the duration of the ICU stay, could also be envisaged, justified by the principle of beneficence towards the patient’s family ^[3][21].

The family’s capacity to understand medical information is an important aspect to take into account. Researchers reported that the family’s poor capacity for comprehension was cited by physicians as a factor in their choice of a suitable surrogate. The physicians seem to engage in more intense communication in such cases, but may also tend to limit the amount or type of information imparted, to avoid putting the family into difficulty, either out of beneficence towards them, or because the physician feels that the family will simply not be able to understand or assimilate the information. Conversely, it is noteworthy that compared to their younger counterparts, older and/or more experienced physicians are significantly less prone to limit information, but rather adapt their language appropriately ^[2][20].

In the ICU, the vulnerability of the patient’s family evolves in parallel to that of the patient. The family are eager for a word, a gesture, or a titbit of information that might relieve their stress even a little. The family is in a situation of complete dependence on their loved-one’s outcome, and on the medical and caregiving teams, who hold the key to the diagnosis, treatment and information. A lack of communication can give rise to post-traumatic stress disorder, with resultant repercussions on daily functioning, and the possible need for psychological assistance ^[13][14][30,31].

In the ICU setting, numerous factors contribute to almost instantaneous and constant transmission of medical information to the families, such as: the constant presence of medical staff, the availability of the caregivers, the intensity of the care delivered, the large range of complementary exams performed, the increasingly long visiting hours, the wide spectrum of variability in the clinical state of patients, and the uncertainty surrounding prognosis. An oversight or an inaccurate piece of information can usually be corrected in time, before—aided by the family’s anxiety—a state of incomprehension or loss of trust in the medical team sets in.

Absent the possibility to have a meaningful relationship with the patient, the caregivers in the ICU have a very particular relationship with the family. They must at all times show benevolence towards the family, and assist them as they deal with the difficult situation of their loved-one’s hospitalization. In this way, there is a shift from “what is right for the patient” towards “what is right for the family”. Building and cultivating this relationship of trust, notably via pertinent communication, certainly contributes to creating the climate of trust that is essential between the caregivers and the family. In addition, this also promotes high quality management of the patient, and most likely also increases the family’s overall satisfaction ^[8][15][25,32].

Apart from being a key component of the quality of care, provision of adequate information meets a strong need expressed by the families, whose anxiety and “need to know” are understandable. In addition to the legal aspects of providing information in the ICU setting, it is important to take account of the relatives’ need for information. They are constantly waiting for some news about the status of a loved-one with whom they cannot communicate, and the information given to them must therefore be transparent, relevant and adapted to their level of comprehension. When a family is worried and desirous of information about the health status of a loved-one who is unable to express themselves, then providing that information is undoubtedly an act of benevolence.

There is one final issue that can arise with regard to information to families of ICU patients, and that is, what would the patient have wanted? Would the patient have agreed for specific information about his/her prognosis to be delivered indiscriminately to the family members? Is there any possibility that the information about the patient’s prognosis, given to the family by the physician, could affect the family, or their personal lives, or their future projects or future commitment to the patient? When the prognosis is uncertain, poor or abysmal, telling the family as much will inevitably cause them distress. This must be considered during discussions with the family, and it behoves the ICU physician to be particularly prudent and empathetic when addressing this point in conversations with the family.

The question of the type of information to be given to the families of ICU patients is at the interface between medical confidentiality and the family’s capacity to understand. Medical confidentiality is intended to protect the patient, and preserve their image, their future and their past. It is easy to understand that the information given to the family of an unconscious ICU patient should not be detrimental to the patient in any way. In this context, it would be acceptable to envisage a principle whereby the content and/or extent of information could be modulated according to personal, clinical, diagnostic and therapeutic data, but also according to the person receiving the information. It would be up to the physician to judge the quantity and quality of information that it is appropriate to deliver in each situation.

In the ICU setting, it is established that the type of information delivered to the patient’s family usually concerns the diagnosis, the expected course of disease and the key components of the treatment ^[5][18]. Estimates of the prognosis are more difficult, because the prognosis may often be hard to predict, especially in the early stages of the ICU stay. Therefore, information about prognosis should be cautious, and various communication strategies can be used to facilitate such conversations ^[16][17][33,34].

A second important element in the information to families relates to the need for additional investigations ^[2][20]. For the family, the fact that their loved-one is undergoing additional tests can reassure them as to the medical management. However, they may also be worried, and will inevitably be asking for the results of any additional investigations. It may even happen that the family expresses surprise that no such investigations are being performed, or that a patient as seriously ill as their loved-one should be in the ICU without undergoing additional investigations—as if the management of critically ill patients absolutely had to comprise a battery of complicated tests. Clinical judgement can pale in comparison to “indisputable” results from a blood test or CT scan. It can be hard to justify the “clinical judgement” approach when faced with questions from the family. It is reasonable for the family to believe that the results of additional tests might be important to counterbalance what the physician has told them, because they may perceive the physician’s discourse to be variable or inaccurate, despite the physician’s best efforts to be cautious and patient. Additional investigations, and their results, can appear to be more reliable, more concrete that just the doctor’s word. Indeed, it is commonplace, when non-urgent tests are being scheduled, such as a CT scan for persistent coma, for the family to consider it as a milestone in the patient’s pathway. For the physician, upcoming tests can provide an opportunity to inform, but also to stall for time, give perspective or materialize the stages of the patient’s clinical course.

Despite the increasing trend towards electronic health records in the ICU, one of the main sources of information remains the patient’s chart, a veritable treasure trove of information that brings together data about vital signs, treatment, biology and microbiology results, traceability of all additional tests and procedures, and key clinical events. The patient’s chart is almost universally available to the family in the ICU, and it is not unusual to see family members rush to check the chart first thing, as soon as they enter the patient’s room. It represents reliable and indisputable testimony to the events that occurred during their absence. It is also not unusual for the family to question the physicians about any changes or inconsistencies that they notice when reading the patient’s chart.

The first and easiest approach to keeping medical information confidential is to give information to the official surrogate only. However, this policy has some shortcomings, as previously outlined. Another approach is the more pragmatic approach practiced in many ICUs, whereby a wider range of information is given, or to a wider circle of family members, justified by their right and need to know. This raises the question of the confidentiality of medical information in this context. It is evident that in view of current practices, absolute secrecy is increasingly difficult to uphold in the setting of the ICU. Indeed, the current legal framework was not conceived with the ICU setting in mind. It is impossible to ignore the high levels of distress suffered by the families of ICU patients, and they cannot be denied the empathy that is their due. Consequently, and inevitably, these considerations override the duty of medical confidentiality. If an individual wants information and has a right to that information (especially if they are related to someone hospitalized in the ICU), then information covered by the rule of medical confidentiality can be revealed with the sole aim of informing and reassuring the worried relatives. When seen in this light, transgressing the rule of secrecy could be seen as legitimate, and in fine, merely translates the reality of observed practices in the ICU, which may not follow the legal framework.

In summary, information is most often provided to family members out of necessity, in the absence of an officially designated surrogate, and out of empathy, due to their legitimate state of worry. The ICU physician is obliged to adapt his/her practices in order to meet the obligation to provide transparent and full information. This approach to providing information may lead to practices that fall outside the framework defined by current legislation, particularly as regards respecting medical confidentiality.

In clinical research involving human subjects ^[18][35], clear and honest information must be given to the potential participant, and the research may commence only after the patient has provided informed consent ^[19][13]. It should be underlined here that accepting to be enrolled in a clinical research project implies that one also accepts the limits of medical knowledge. It is easy to imagine the difficulties that this assent creates for patients and families, at a time when they are experiencing an unexpected and serious clinical situation, whose outcome remains uncertain. The information process cannot and must not be reduced to obtaining consent alone, if it is signed, written consent. It also cannot simply be limited to the content of the leaflets given to the patient, because such an approach is not appropriate for the clinical situation or the doctor–patient relationship ^[20][6]. Information about participating in research must be the starting point for a relationship of trust between the patient and the physician, which should be maintained throughout the duration of the research, and even beyond. Accordingly, sending the results of the research project to the patients may give them insights into their personal situation ^[21][15]. Here again, for the purposes of research in the ICU, the question arises of how decisions can be made by patients who are unable to communicate. In such situations, the patient’s surrogate decision making gives consent on the patient’s behalf, although often without knowing whether the patient would have wished to participate or not, and sometimes even in the knowledge that the patient probably would not have accepted to participate ^[22][36].