Ethnography of "Local Universality": Admission Practices in an Intensive Care Unit Among Guidelines, Routines, and Humour

Abstract: The article analyses the existing gap between the formal dimension of evidence-based medicine (EBM), as constituted by protocols, procedures, and guidelines, and actual professional practices in relation to a specific issue: the admission of patients to an intensive care unit (ICU). The results of a case study, carried out in the ICU of a hospital in the north of Italy between 2006 and 2007 are reported. The study was performed using ethnographic methods: participant observation, ethnographic interviews, and semi-structured interviews. Empirical data have been analysed using a grounded theory approach. The results show how three dimensions (macrosocial, organisational-interactional, and individual) become intertwined with the operational guidelines that have been drafted on the basis of international evidence. The standardisation process that the guidelines presuppose results in the adoption of a variety of different local styles with respect to the approach that individual doctors take in relation to the admission of a patient to an ICU. These styles can range from strict adherence to the international criteria to a greater compliance with medical–legal, organisational, and individual needs. Furthermore, the results of the study demonstrate how relational knowledge, as a form of situated knowledge, can allow the personnel involved to activate local resources (organisational, professional, and personal) in order to incorporate the formal prescriptions of EBM in professional practice.

Key words: evidence-based medicine; intensive care; situated practice; humour; ethnography; participant observation; interviews; grounded theory methodology

Evidence-based medicine (EBM) currently represents the mainstream perspective in medical epistemology and is applied in every type of setting for treatment and care (for example, from evidence-based clinical practice to evidence-based nursing and from evidence-based oncology to evidence-based prevention). According to many scholars, the success of this epistemology, which was introduced into the medical discourse at the beginning of the 1990s, seems to result from its effectiveness in facilitating the standardisation of medical practices. However, recent studies have shown that when evidence-based instruments (such as guidelines, protocols, and systems to support decision-making) are applied locally, they undergo a process of interpretation and adaptation (BERG, 1997; TIMMERMANS & BERG, 2010). This process is necessary because social variables are usually excluded from standardisation criteria and medical-scientific publications (GREENHALGH, 2014). [1]

The aim of this article is to analyse the existing gap between the formal dimension of EBM, as constituted by protocols, procedures, and guidelines, and actual professional practices in relation to a specific issue: the admission of patients to an intensive care unit (ICU). Whether or not to admit a patient to an ICU is clearly a critical decision. It might determine whether the patient survives or dies (and thus the adequacy of the treatment must be stressed in the decision-making process, especially in the case of elderly people or patients in critical conditions), and will entail the use of considerable resources. Decisions about the legitimate use of highly complex treatments that require advanced technological instruments and constant care must take into account the structural limitations of funding to healthcare organisations and the relevant criteria for the distribution of these funds (LAPSLEY & MELIA, 2001). [2]

The study described herein is based on qualitative empirical data that were gathered during field research carried out in the ICU of a hospital in the north of Italy between 2006 and 2007. The ethnographic data reveal that daily practice in an ICU is influenced by two factors that constantly interact: the formal dimension, which emphasises the rhetoric of universalism within western medicine and health organisations (TIMMERMANS & KOLKER, 2004), and the actual behaviour of professionals, in which formal protocols and guidelines are reinterpreted and become embedded in practices that have been adapted to the local situation. The aim of the study was to analyse how the guidelines that are provided in relation to the admission of patients to an ICU are incorporated into the actual practice of medical personnel. [3]

The article is organised as follows: First, I briefly review and discuss propositions about the interaction between EBM and medical practice (Section 2). After having illustrated the data and methods (Section 3), I present the research results, focusing on the different styles of the integration of EBM guidelines in a unit's everyday practice (Section 4). Lastly, I discuss the findings, providing some requirements for further research (Section 5). [4]

Within the discourse on medical/nursing matters, EBM focuses on what has been defined in the scientific community as "the conscientious, explicit, and judicious use of current best evidence in making decisions about the care of individual patients" (SACKETT, ROSENBERG, GRAY, HAYNES & RICHARDSON, 1996, p.71). From the first formulation of what was meant explicitly to be a new "paradigm for medical practice" (EVIDENCE-BASED MEDICINE WORKING GROUP, 1992, p. 2420), the practice of EBM has spread rapidly throughout the field of medicine—in clinical and research fields, as well as in cure and care settings. [5]

It has been noted that the development of EBM has been supported by the strong boost that EBM has given to the process of formal rationalisation, which began in the 1980s (BERG, 1997). In particular, as TIMMERMANS and BERG (2010) have pointed out, the standardisation of practices that has been fostered by EBM has been welcomed by the relevant players in the healthcare system. From a management point of view, the process of standardisation meets the need to identify a "newer form of accountability" that "attempts to manage the entire clinical process by making it transparent" (p.112), whereas, as far as the professionals are concerned, it contributes to attaining "the elusive fusion between scientific knowledge and clinical practice, attempting to turn the art of medicine into a science" (p.87). [6]

However, as various studies of science and technology and other studies that have stemmed from the socio-anthropological field have emphasised, the EBM paradigm is characterised by a rhetoric of the formalisation of practices that is strongly reductionist and does not correspond to the daily life of medical professionals or to the practical implementation of treatments (CARMEL, 2013; GOODWIN, 2008; GREENHALGH, 2014; MOREIRA, 2007; PRASAD, 2005). [7]

Through the empirical analysis of professional practices, Marc BERG (1997) has stressed how the introduction of technologies and artefacts (e.g. software or intervention protocols) into an existing practice is a process of continuous negotiation. As a result, all the elements involved, including the technologies and artefacts, acquire a new significance. [8]

With respect to daily medical and nursing practice, prescriptions or directions that have been developed by international agencies are not simply reproduced locally. There is an empirical process of interpretation that involves several subcultures and different systems of significance (MOL, 2002, 2008). TIMMERMANS and BERG (1997) have introduced the concept of "local universality" to describe the processes of adjustment and partial transcription of protocols, guidelines, and other instruments that support the decision-making process, which accompany their incorporation into medical practice. [9]

In the light of these studies, it appears that the evidence that EBM aims to produce is a contextual and disconnected from many of the factors that usually affect treatment processes and the daily practice of medical professionals (e.g. organisational resources, the level of compliance of patients, interactions among different professional groups, and communities of practice, just to name a few). As TIMMERMANS and EPSTEIN (2010) claim, when assessing the gold standard of clinical tests, i.e. the randomised group-controlled double-blind clinical trial, life, with its chaos-like character, severely tests the purity of the abstract model. [10]

MANGHI (2005), in relation to the nature of medical knowledge, stresses how the analytical and instrumental knowledge and skills that are required to observe and interpret the symptoms and to implement the appropriate therapeutic processes have a cognitive character. In other words, these skills "are or can be codified in repeatable discursive forms which are expected to be universally shared or shareable" (p.25; my translation). The evidence-based "objects" (clinical trials, guidelines, protocols, and so on) are consistent with this knowledge scheme and can be perceived as external and impersonal by their users, and "deprived" of any individual or local meaning. [11]

MANGHI adds that there are two other knowledge schemes that are relevant to medical knowledge: normative knowledge, which concerns the ideal and practical orientations of a deontological and moral nature, and relational knowledge, which refers to the emotional dimension that is inevitably involved in any human interaction. Relational knowledge is mainly situated, produced, and maintained "by means of the actors' participation in the interactive, communicative and social contexts" (p.28; my translation). It is particularly relevant for social actors, although to a large extent it is applied unconsciously; it is characterised by a strong character of prescriptivity on the social agency (TOMELLERI, LUSARDI & ARTIOLI, 2015). According to the taxonomy of rationing in health care that has been defined by LAPSLEY and MELIA (2001), it is a form of soft rationing, defined as "a situation in which there may exist general constraints (of a financial or physical nature) which limit activities, but which may be attenuated by the actions of clinical managers" (p.730). [12]

There is no acknowledgement of relational knowledge in EBM reports, in the operative tools that have been developed from those reports, or in the academic programs that are implemented in medical and nursing education. However, relational knowledge is embedded in the practices of medical professionals, and contributes to the production and maintenance of the moral order, in which potentially disruptive conflicts are dealt with through specific organisational structures (LATIMER & SCHILLMEIER, 2009). [13]

The empirical data that are presented in the following section demonstrate that even in an ICU, which is characterised by a high degree of formalisation of knowledge and practices, the formal elements (protocols and guidelines) concerning the admission of a patient are incorporated into situated practices that are influenced strongly by informal elements (social, organisational, and relational aspects). [14]

The data analysed in this study came from an ethnographic case study that was carried out between March 2006 and March 2007 in the ICU of a hospital in the north of Italy. The aim of the study was to analyse the changes that are currently occurring in contemporary medicine by investigating the working practices of the ward and the symbolic representations of the health care professionals involved. [15]

The study sought to take into account of both the pragmatic component of human agency, i.e. the empirically observable behaviours and activities of individuals, and the symbolic framework within which these behaviours and activities are carried out. In the first case, the unit of analysis has been the working practice while in the latter it has been the moral tales (SILVERMAN, 2011), or the competent ways in which people describe themselves and their actions. [16]

The case study was performed using ethnographic methods: participant observation, ethnographic interviews, and semi-structured interviews. In this study I assumed the position of the observer as participant, following the categorisation proposed by KAWULICH (2005). This means that "the researcher is an observer who is not a member of the group and who is interested in participating as a means for conducting better observation and, hence, generating more complete understanding of the group's activities" (§6). [17]

During the period that I spent within the unit, I observed situated practices and the daily activities of the ICU for at least 25 hours a week, transcribing the observations in ethnographic notes (EMERSON, FRETZ & SHAW, 2011). I performed several ethnographic interviews (SPRADLEY, 1979) with the 15 physicians and 60 nurses who worked permanently on the ward. With the nurses, I also carried out 15 semi-structured interviews (WENGRAF, 2001) more devoted to the symbolic dimension related to practice. I participated in 30 daily briefings of the medical équipe (namely meetings about the therapeutic plans for the patients), and assisted in 60 meetings with the relatives of patients. The interviews lasted from 20 to 90 minutes, the staff meetings from 30 to 45 minutes, and the meetings with relatives from 5 to 30 minutes. The observation was semi-covert, i.e. the goals of the study were known only by the personnel of the ICU. Interviews, staff meetings, and discussions with relatives were audio-recorded and transcribed into digital files. The resulting empirical materials, both observational and narrative data, were coded using ATLAS.ti (MUHR & FRIESE, 2004). [18]

I adopted a grounded method of analysis, as it is widely regarded as effective in revealing the invisible work involved in many kinds of tasks (STAR, 2007). It allows brightening domains of everyday life that other analytical methods (more theoretically oriented) may have underestimated. I followed the traditional coding procedures described by CHARMAZ (2006), as a heuristic tool for understanding the social world I have studied but without generating a theory. [19]

In the initial phase of coding, I used the incident-to-incident approach, as it is reported as being particularly effective when the empirical base consists mainly of observations and descriptions of daily practices. Then, the analysis passed through a focused coding phase: in the generated codes I have identified the main categories that contribute to synthesising the data through the construction of homogeneous families of meaning. In the third phase of the theoretical coding I looked for the connections among the macro categories in order to weave an analytic choreography that was consistent with the empirical base. The three phases were accompanied by a process of constant comparison of analytical codes (GLASER & STRAUSS, 1967), to mark analytic distinctions and make comparisons at each level of the analytic work. [20]

At the end of the analysis process, ICU working practice was determined to be the result of the interaction between three main dimensions: the macrosocial dimension, connected to the cultural, legal, and epistemological domains; the mesosocial dimension, connected to administrative and technological process; and the microsocial dimension, connected to body work and individual motivation. [21]

Humour emerged as an apparently marginal element in the overall frame. I found its place in the general theory when I came back to the empirical basis for understanding how the staff could reconcile the contradictions that characterise intensive care work in everyday practice (for example, as discussed in this article, the tension between evidence-based guidelines and organisational habits and social expectations towards hospitalisation in the ICU). Humour appears as a strategy for maintaining the group's social cohesion despite the different personal perspectives, which are often radically divergent. The empirical data presented in this article have been translated into English from the original language, Italian. In the following pages, surnames and other personal information have been modified to guarantee anonymity. [22]

The results section is structured in 3 parts: In the first part, I will present the admission protocol containing the formal rules for patients admission and I will introduce the concept of local habits and its articulation; in the second part I will analyse how local habits can shape everyday medical practice; in the final part, I will discuss how the complex configurations of formal rules and local habits are embodied in different local styles of interpretation of the admission protocol and how humour can become a strategy to manage the coexistence of opposing styles. [23]

"Procedures that meet the criteria for admission to and discharge from the Intensive Care Unit and for limitation of care" is the title of a document that hangs between administrative notices and internal communications at the top of a cork notice board in the "medical staffroom" of the ICU in the ward where the study was performed. The staffroom is where doctors usually gather at the end of shifts, to fill in clinical files, to examine medical images, or simply to rest. The document was drafted by the head physician, namely the director of the ICU, subsequently approved by the chief administrative officer, and finally signed by the director of the hospital. It defines the guidelines to which medical personnel must refer when assessing the appropriateness of the hospitalisation or continuation of treatment of a patient. [24]

The document was drafted in accordance with the guidelines of the Italian Society of Anaesthesia, Analgesia, Resuscitation, and Intensive Care (SIAARTI, 2003). It identifies two elements that define the appropriateness of therapeutic intervention: the acuteness of the physiological problem and the margin of improvement. As was confirmed many times by the head physician of the unit, the ICU must admit patients susceptible of cure, namely ill persons who can continue a therapeutic course in a specific ward, after discharge from the ICU. [25]

The admission protocol defines a scale with four levels of priority, which range from a maximum expected potential benefit for the patient to the complete inappropriateness of the treatment:

Priority 3: patients in a critical condition with a reduced possibility of recovery;

Priority 4: patients whose admission to the ICU is generally not appropriate. These can be represented by three categories:

These categories have a clear purpose: to standardise the procedure for admission to the ICU on the basis of a priority scale produced by a national organisation, SIAARTI, which also takes into account the clinical evidence produced at an international level. However, as is shown below, the application of this protocol requires the existence of broad social dynamics whose functional and essential requisites cannot be aligned with the linear and reductionist logic upon which evidence-based artefacts are founded (TIMMERMANS & BERG, 2010). [27]

The ICU personnel are well aware of this gap, as demonstrated by the paradoxical words of Dr. Greco, a physician with more than 20 years of professional experience: "all protocols are made for intelligent people, who are able to use them going beyond the protocol". Primarily, the medical professionals include in the therapeutic strategy the contingent and social elements that are neglected by the protocol itself, but of which the professionals are inevitably aware. [28]

Showing some sociological sensitivity, during an ethnographic interview Dr. Greco identifies a set of practices that occur alongside the admission protocol and take into account various sociological dimensions. He calls them habits and they correspond to the following analytical levels:

As a consequence, the local implementation of protocols and guidelines involves a series of negotiation processes, which each incorporate elements that belong to different domains, ranging from the sociocultural dimension to the individual moral dimension. As a result, these negotiations end in specific articulations of local universalities (TIMMERMANS & BERG, 1997). [30]

The following case shows how the three above-mentioned habits are connected indissolubly in relation to the practice within wards. The case has been reported by Dr. Marino, the physician in charge of the unit, during an ethnographic interview. While he was explaining the case, the doctor reported the interaction with his colleague otolaryngologist, the other protagonist of this accident:

"Once I was called into Otorhino [the Otolaryngology ward] because of a [cardiac] arrest. I start [practising cardiopulmonary resuscitation] trac trac trac [onomatopoeic sounds], the otolaryngologist just watching. He starts again [the patient's heart starts beating again], I was there, ventilating him [blowing air into the patient's lungs]. When he is out of danger, I asked [the otolaryngologist]: 'What is his problem?'¹⁾

[O:] No, no, no, no chemotherapy at all. On the contrary, it is necessary to determine whether he can receive pain therapy to relieve him from the pain.

We argued like this for a bit, then the fellow had another cardiac arrest, I didn't touch him and he died. I said to my colleague: 'What are you doing? You make him come back to give him pain therapy so he can die next month ...'.

You know that he filed a complaint against me to the Medical Director because I didn't want to resuscitate the patient? This is the mentality. In the wards, people don't make this type of decision" (Ethnographic Note 34). [31]

This excerpt is clearly characterised by the emergence of a conflict between the otolaryngologist and Dr. Marino. The conflict was generated by the different assessments that the professionals made about the appropriateness of the resuscitation and subsequent ICU admission of the patient in question. [32]

In the practice of the observed ICU only one of the doctors on duty was responsible for making the decision to admit a patient. Other hospital units could request the advice of the doctors in the ICU through a pager. The doctor who was carrying the pager during a given shift would in turn call the operational unit in need of help and intervene. If the situation appeared to be particularly complex, the doctor carrying the pager could ask for the assistance of a colleague. However, the individual doctor would usually go alone, accompanied by a nurse. [33]

When an ICU doctor is called for advice, especially during an emergency such as a cardio-respiratory arrest, (s)he has insufficient time to review the patient's medical history, evaluate the benefits of treatment on the basis of a hypothetical prognosis, and decide whether or not to implement the treatment. Instead, the doctor relies on some information that can be grasped visually and some instrumental data that has been collected by the personnel on the ward, such as blood pressure and temperature, although the main reference is the doctor who first requested the intervention. When there is no data available on the patient, as in the above-mentioned case, the ICU admission protocol states that "it is good practice to hospitalise the patient in the ICU". However, this often conflicts with the subsequent evaluation of the decision taken. [34]

In the episode described above, the ICU doctor became aware that the treatment was probably inappropriate only at the end of the resuscitation. According to the guidelines of the unit, the patient was a Priority 4 case, due to his terminal cancer. Usually, the ICU doctor trusts the analysis of the other doctor, and assumes that the latter must have already considered the general state of health of the patient and the appropriateness of the treatment. However, in the case under consideration, one of the hospital habits had come into play, that is, to leave the decision with respect to terminal or end-of-life cases to the ICU medical personnel, regardless of the general condition of the patient. [35]

The decision of the ICU doctor can also be affected by state habits, among which the juridical dimension is particularly important, especially in relation to the decision of whether or not to resuscitate the patient and admit them to the ICU. It is possible that the relatives of a patient might decide to file a complaint for failure to assist or for negligence. Alternatively, as in the episode described above, the doctor on the ward where the patient is being treated might file a complaint to the Medical Director concerning the manner in which the ICU doctor dealt with the situation, for example in this case because he refused to resuscitate the patient during the second cardiac arrest. Medical personnel are well aware of (and perhaps afraid of) the legal consequences and can consolidate this fear into a type of personal habit, because it arises constantly when decisions must be made. This has been called defensive medicine, in which medical-legal considerations tend to prevail over clinical ones (CATINO, 2009). [36]

In relation to hospital habits an opposite attitude emerges in which formal prescriptions and clinical data are given greater importance than organisational and legal considerations. This attitude is characterised by a greater strictness of professionals with respect to analysis of the patient and adherence to the ICU admission protocol, as evidenced by the following excerpt from ethnographic notes. Dr. Ricci, a senior physician with a vast professional experience in both intensive care and internal medicine, received a call from a doctor on the general medicine (GM) ward about a patient whom he had examined just a little earlier in the emergency room (ER) and for whom he had considered hospitalisation in the ICU to be inappropriate:

"Dr. Ricci and I are in the medical staffroom. He is filling out some clinical files. The telephone rings, he picks up the phone and answer.

[On the other end of the phone somebody says that there is a patient in ER and that GM has no beds available.]

Ricci: Listen, that patient doesn't need to be in the ICU, otherwise I would have found a way to take him. [...] The fact that you don't have room only means that he must be sent somewhere else, but not to our ward, because I wouldn't admit him even if the ICU was empty, because he is not ...

Ricci: Then, again, he's not going to be hospitalised in the ICU because he cannot do ...

Ricci: Listen to me, I was an internist for six years and we used to keep these patients in GM. Then, I don't know what to tell you, please hold on, I'm not gonna take him, you go find him a place somewhere else, it is not my problem. OK?

Dr. Ricci was struggling with another hospital habit: the directors of operational units resorted to the ICU as a reservoir of beds when other units were full. This functional allocation can be linked to the frequent availability of beds in the ICU due to the need to ensure that emergencies are covered. In other words, the personnel of other units considered the ICU to be a sort of subsidiary unit of last resort. In the conversation with the doctor who called from GM, Ricci maintained his stance firmly. He disagreed with hospitalisation in the ICU because the patient did not need intensive care. In fact, according to the protocol, the patient could be assigned to the Priority 4 category, because he did not present altered vital functions, but rather a partially compromised general picture. The ICU doctor contradicted the idea that the lack of available beds in GM justified the hospitalisation of the patient in the ICU. The condition of the patient suggested that intensive care treatment would be inappropriate, regardless of the availability of beds. As a consequence, the ICU doctor took upon himself the clinical and juridical responsibility of denying the treatment to the patient. [38]

However, this behaviour is not the only way to interpret the admission protocol. The following two excerpts from ethnographic notes present an informal conversation that I had with Dr. Conti and Dr. Bruno. During this conversation they discussed their point of view about patients' admission in ICU. Both the doctors have a large working experience in ICU and are also close friends. Dr. Conti, talking to me and Dr. Bruno, stresses how the decision to admit a patient is ultimately an individual choice:

"How many times do we throw in each other's faces, rudely or nicely, the fact that we brought patients in for no reason? It has happened to all of us, it could happen for a number of different reasons. To her [Dr. Bruno], because she's, let's say, nice, and therefore she never makes the decision that a patient doesn't have the right to have a chance. Or it happens to me, because, in the ward, I find myself unable to deny the opportunity to the relatives, who are for instance particularly aggressive, or to a patient who has been mistreated. It is therefore hard to say how many patients should not have been hospitalised, because it is almost impossible to have an absolute definition through which you can ... [decide who to admit and who not to admit to the ICU]" (Ethnographic Note 38). [39]

The decision to hospitalise a patient in an ICU involves the professional and personal characteristics of the doctor concerned. It can be influenced not only by the rational evaluation of specific clinical criteria, but also by the personal ethics of the doctor who has to decide at that precise moment, and the circumstances in which the decision is taken. [40]

Ethical values that are linked to religious beliefs can affect the decision about whether to admit a patient or not, in the same way as a personal habit would. For instance, ethical values that are linked in particular to Catholic morals can have a crucial influence on such decisions, because the official bioethics of the Catholic Church emphasise the "sanctity of human life" (and of the individual person, as a consequence). In other words, first and foremost, human life is not subject to the will of individuals, and thus must be preserved at all costs (KELLY, MAGILL & TEN HAVE, 2013). As a consequence, for doctors with such religious beliefs, ethical-religious considerations might prevail over clinical ones, which might result in their hospitalising a patient whom other doctors would not. [41]

Dr. Bruno is a practising Catholic who has regular involvement in the pastoral and recreational activities that are organised by her parish, as she described during an interview. Her religious belief is particularly important in relation to her professional practice. She objects as a matter of conscience to practices such as assisting in voluntary interruptions of pregnancy (during which the support of a reviver/anaesthetist is necessary, being a surgical intervention). Moreover, she admits patients to the ICU whose hospitalisation is considered to be inappropriate by her colleagues. [42]

Continuing the conversation Dr. Bruno and Dr. Conti argue about precisely this issue:

"Bruno: In the end how many times have you brought downstairs [to the ICU] patients who afterwards woke up with just a little effort? How many times have I been told that the old lady could stay upstairs [in another ward]? Well, not so often, though ...

Conti: Of course! This is of primary importance! She clings to these things. She says: those times things went well ...

Conti: I've just said a random number! And nobody will be able to say it wasn't worth it ...

Conti: Therefore nobody can say that those two times can be overlooked because of the other ten times things went wrong" (Ethnographic Note 38). [43]

Conti seems to suggest that hope is not grounded on statistical analysis. As a consequence, Dr. Bruno's decisions are the result of a fideistic attitude. [44]

A second relevant aspect stressed by Conti is that circumstances might modify the result of an assessment. Particularly belligerent relatives, who are unable to accept the imminent death of a member of the family, patients with a problematic history within the healthcare organisation (mistreated patients), or a personal acquaintance between the doctor and the patient or their relatives can influence the doctor to admit the patient even though the clinical conditions would not support this. [45]

In the practice of the ICU, the interaction between state, hospital, and personal habits resulted in three well-distinguished local styles in relation to the application of the admission protocol and the definition of the criteria for the evaluation of patients who need intensive care: the compliant, the intransigent, and the situational style: [46]

The compliant style is represented by doctors such as Dr. Marino and Dr. Bruno, who appeared to be more willing than their colleagues to admit patients, even if the condition of the patients did not match the standards of the protocol. Their greater tolerance could be attributed to a fear of legal consequences and adherence to ethical or ethical-religious values. [47]

The intransigent style is represented by doctors such as Dr. Ricci and Dr. Russo (mentioned below), who were less flexible with respect to the analysis of clinical conditions. They tended to deny hospitalisation for patients whom they did not consider in need of intensive care due to the absence of indicators defined by the admission protocol. [48]

Finally, the situational style is represented by doctors such as Dr. Conti and Dr. Greco, who displayed an intermediate approach. They tended mainly to consider the clinical conditions. However, in certain circumstances, personal considerations did lead them to admit unsuitable patients. [49]

These different styles occurred among a group of professionals who were found within the same work context and who demonstrated strong personal relationships even outside the work place. [50]

From the ethnographic observations, it emerged that humour represented a major strategy for maintaining the relationships within the group despite the existence of different styles of working. As can be seen in the following excerpt from an ethnographic note, the doctors tended to refer to admissions filed by colleagues with witty remarks. Dr. Longo mentioned below is a young physician, recently hired in the ICU.

"15:45 – The ward is particularly quiet, because the relatives have departed. There are some nurses who are chatting, looking at something on a computer screen. Dr. Longo is filling out a patient's clinical file, which is hanging on the metal cart at the bottom of the bed. Dr. Ricci comes into the ICU, glances around, goes towards Longo and says: 'How is the girlfriend doing?'

Ricci: The lady that Marino has authorised for admission [he points to the old lady in bed Number 6]. He must really be in love, to admit somebody like her ...

Dr. Ricci's joke refers to an old lady who had been admitted the day before by Dr. Marino, after a consultation with the GM ward. The patient is 85 years old. Except for a minor problem with respect to her breathing ability, she does not have any serious issues. The joke highlights Dr. Ricci's opinion that this is an inappropriate admission because it does not comply with the standards defined by the protocol. [52]

Analogous phenomena have been described by GOFFMAN (1961), who used his observations among operating teams as a starting point. Even in such a setting, humour, especially from the one who is directing the intervention, can be used as a strategy to highlight inappropriate types of behaviour without compromising the cooperation of the working group. [53]

Another example is shown by the following excerpt. Dr. Russo, another senior physician, well known among colleagues for his cynicism, is sarcastically teasing Dr. Bruno, who has admitted an obese, 88-year-old woman who is hardly breathing and is already in a coma:

Marino: Well, she arrived yesterday ... it's all her fault! [He points to Dr Bruno.]

On other occasions, the humour does not refer directly to the doctor. Instead, mockery is directed towards medical aids in order to stress their uselessness when it comes to certain patients. For example, during the change of shift in the afternoon, Marino proposes the use of a quite expensive medical aid for a patient whose ICU admission had already been criticised. Russo ridicules the use of that medical aid for the patient:

"Bruno: Rossi [a patient] is doing fine. He's been breathing spontaneously all day long today.

Marino: We are going to apply the helmet [non-invasive respiratory aid] to the other patient.

Russo: The helmet? Then we can call the hairstylist and apply curlers to him too at that point!

In the practice of the unit, humour represents a chance to disperse tensions within the team and allows solutions to disputes to be found while avoiding the disintegration of the group. As MEAD has pointed out (in BATESON, 1953), humour appears when "there is a discrepancy between what is correct to express and what everybody feels" (p.11), as can occur when it is necessary to integrate very different styles of decision-making in order to coordinate a team. [56]

The analysis of the process for the admission of patients to the ICU suggests that the formal artefacts that comply with EBM standards (i.e. the guidelines that describe in detail the procedure for the admission and discharge of patients) must be implemented within a social context that is unavoidably unique for each admission. The guidelines must interact with human beings (primarily professionals, patients, and relatives), symbolic actors (in particular, in Italy, the medical management, the Italian legislative system, and the Catholic system of morals), and different circumstances (economic, organisational, and relational). [57]

The elements that are involved in the admission process for patients and that contribute to the local implementation of international guidelines belong to three different dimensions:

As the above-mentioned dimensions become intertwined with the operational guidelines that have been drafted on the basis of international evidence, the standardisation process that the guidelines presuppose results in the adoption of a variety of different local styles with respect to the approach that individual doctors take in relation to the admission of a patient to an ICU. These styles can range from strict adherence to the international criteria to a greater compliance with medical-legal, organisational, and individual needs. [59]

Protocols, guidelines, and other instruments that support the decision-making of medical/nursing personnel undergo an adjustment process within local environments that implies a new attribution of significance to the artefacts within consolidated representations (BERG, 1997). In the eyes of professionals, these representations, which are consolidated in the form of routine, can appear to be truly rooted habits. [60]

As a result, the local peculiarities and singularities that the EBM paradigm is intended to remove actually interact with the evidence-based artefacts. In other words, the characteristics of the ward, hospital, and district organisation of the healthcare services, together with the personal biographies of the professionals involved and the histories of the patients and relatives, make a major contribution to the production of a local universality (TIMMERMANS & BERG, 1997) from the evidence-based artefacts during the implementation phase. [61]

The results of this empirical study further underline the relevance of relational knowledge (MANGHI, 2005) in the process of integrating protocols and guidelines into medical practice. It is through the application of this knowledge that medical personnel can cope on a day-to-day basis with the gap between formal prescriptions and the actual circumstances that surround individual cases. Relational knowledge, which is constantly produced and reproduced, is a situated knowledge that allows the personnel involved to activate local resources (organisational, professional, and relational) in order to implement the formal prescriptions. For instance, the frequent recourse to humour among professionals appears to be an effective strategy of coordinating the different decisional styles in a ward, and of managing the medical équipe if different individuals interpret the protocol for the admission of patients differently. [62]

Thus, the article shows that relational knowledge plays a fundamental role in the process of the incorporation of operational guidelines and other EBM artefacts in professional everyday practice. In doing so, the article claims that analysing the specificities of empirical settings is the key to understanding how those artefacts become effective in medical practice. Although I am aware of the impossibility of recording the infinite variety of organisational and biographical contingencies that generate daily practice, other evidence is required to make EBM truly effective in orienting professional practice in healthcare. Further empirical research could be addressed to investigate different strategies through which professionals achieve incorporation and in which ways different macro-, meso-, and microsocial variables affect this process. [63]

1) In the following passage, I have transformed the indirect speech made during the interview in a dialogue between the two doctors to make the interaction more understandable to the reader. <back>

Bateson, Gregory (1953). The position of humor in human communication. In Heinz von Foerster. (Ed.), Cybernetics: Circular causal and feedback mechanisms in biological and social sciences (pp.1-47). New York: Josiah Macy, Jr. Foundation.

Berg, Marc (1997). Rationalizing medical work: Decision-support techniques and medical practices. Cambridge, MA: MIT Press.

Carmel, Simon (2013). The craft of intensive care medicine. Sociology of Health & Illness, 35(5), 731-745.

Catino, Maurizio (2009). Blame culture and defensive medicine. Cognition, Technology & Work, 11(4), 245-253.

Charmaz, Kathy (2006). Constructing grounded theory: A practical guide through qualitative analysis. London: Sage.

Emerson, Robert; Fretz, Rachel & Shaw, Linda (2011). Writing ethnographic field notes. Chicago, IL: University of Chicago Press.

Evidence-Based Medicine Working Group (1992). A new approach to teaching the practice of medicine. Journal of the American Association, 268, 2420-2425.

Glaser, Barney G. & Strauss, Anselm L. (1967). The discovery of grounded theory. Chicago, IL: Aldine.

Goffman, Erving (1961). Encounters. Two studies in the sociology of interaction. Indianapolis, IN: Bobs-Merril.

Goodwin, Dawn (2008). Refashioning bodies, reshaping agency. Science, Technology & Human Values, 33(3), 345-363.

Greenhalgh, Trisha (2014). How to read a paper: The basics of evidence-based medicine. New York: Wiley.

Kawulich, Barbara B. (2005). Participant observation as a data collection method. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 6(2), Art. 43, http://nbn-resolving.de/urn:nbn:de:0114-fqs0502430 [Accessed: January 10, 2015].

Kelly, David F.; Magill, Gerard & Ten Have, Henk (2013).Contemporary Catholic health care ethics (2nd ed.). Georgetown: Georgetown University Press.

Lapsley, Irvine & Melia, Kath (2001). Clinical actions and financial constraints: The limits to rationing intensive care. Sociology of Health and Illness, 23(5), 729-746.

Latimer, Joanna & Schillmeier, Michael (Eds.) (2009). Un/knowing bodies. Hoboken, NJ: Wiley & Sons.

Manghi, Sergio (2005). Il medico, il paziente e l'altro. Un'indagine sull'interazione comunicativa nelle pratiche mediche [The doctor, the patient and the other. An investigation on the interaction of communication in medical practice]. Milan: Franco Angeli.

Mol, Annemarie (2002). The body multiple: Ontology in medical practice. Durham, CA: Duke University Press.

Mol, Annemarie (2008). The logic of care: Health and the problem of patient choice. New York: Routledge.

Moreira, Tiago F. (2007). How to investigate the temporalities of health. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 8(1), Art. 13, http://nbn-resolving.de/urn:nbn:de:0114-fqs0701139 [Accessed: December 2, 2014].

Muhr, Thomas & Friese, Susan (2004). User's manual for ATLAS.ti 5.0. Berlin: ATLAS.ti Scientific Software Development GmbH.

Prasad, Amit (2005). Making images/making bodies: Visibilizing and disciplining through magnetic resonance imaging (MRI). Science, Technology & Human Values, 30(2), 219-316.

Sackett, David; Rosenberg, William; Gray, Muir; Haynes, Bryan & Richardson, Scott (1996). Evidence-based medicine: What it is and what it isn't. BMJ, 312, 71-72.

SIAARTI (2003).Guidelines for admission to and discharge from intensive care units and for the limitation of treatment in intensive care. Minerva Anestesiol, 69, 102-118.

Silverman, David (2011). Interpreting qualitative data: Methods for analysing talk, text and interaction (4th ed.). London: Sage.

Spradley, James P. (1979). The ethnographic interview. New York: Holt, Rinehart and Winston.

Star, Susan (2007). Living grounded theory: Cognitive and emotional forms of pragmatism. In Antony Bryant & Kathy Charmaz (Eds.), The Sage handbook of grounded theory (pp.75-94). London: Sage.

Timmermans, Stefan & Berg, Marc (1997). Standardization in action: Achieving local universality through medical protocols. Social Studies of Science, 27(2), 273-305.

Timmermans, Stefan & Berg, Marc (2010). The gold standard: A sociological exploration of evidence-based medicine and standardization in health care. Philadelphia, PA: Temple University Press.

Timmermans, Stefan & Epstein, Steven (2010). A world of standards but not a standard world: toward a sociology of standards and standardization. Annual Review of Sociology, 36, 69-89.

Timmermans, Stefan & Kolker, Emily (2004). Evidence-based medicine and the reconfiguration of medical knowledge. Journal of Health and Social Behavior, 45, 177-193.

Tomelleri, Stefano; Lusardi, Roberto & Artioli, Giovanna (2015). The metaphors of collaboration, or the social construction of collaborative interactions between health professionals. Acta Bio Medica, 86(1S), 7-18.

Wengraf, Tom (2001). Qualitative research interviewing: Biographic narrative and semi-structured methods. London: Sage.

Roberto LUSARDI is research fellow at University of Bergamo (Italy). His main research interests include qualitative research, particularly ethnographic studies and action research. His personal areas of research interest involve medical practice, interprofessional collaboration and organizational sensemaking.

University of Bergamo
Dept. of Human and Social Sciences
Piazzale S. Agostino n. 2
24129 Bergamo, Italy

Lusardi, Roberto (2015). Ethnography of "Local Universality": Admission Practices in an Intensive Care Unit Among Guidelines, Routines, and Humour [63 paragraphs]. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 16(2), Art. 26,
http://nbn-resolving.de/urn:nbn:de:0114-fqs1502261.

Forum Qualitative Sozialforschung / Forum: Qualitative Social Research (FQS)

ISSN 1438-5627

Creative Commons Attribution 4.0 International License