Article Text

Download PDFPDF

Creating spaces in intensive care for safe communication: a video-reflexive ethnographic study
  1. Su-yin Hor1,
  2. Rick Iedema2,
  3. Elizabeth Manias3
  1. 1Faculty of Arts and Social Sciences, Centre for Health Communication, University of Technology, Sydney, Broadway, New South Wales, Australia
  2. 2Agency for Clinical Innovation, NSW Ministry of Health, Faculty of Health, University of Tasmania, Australia
  3. 3School of Nursing and Midwifery, Deakin University, Melbourne School of Health Sciences, The University of Melbourne, Melbourne, Victoria, Australia
  1. Correspondence to Dr Su-yin Hor, Faculty of Arts and Social Sciences, Centre for Health Communication, University of Technology, Sydney, PO Box 123, Broadway, NSW 2007, Australia; suyin.hor{at}


Background The built environment in acute care settings is a new focus in patient safety research, with few studies focusing primarily on the design of ward environments and the location and choice of material objects such as light fittings and hand-washing basins.

Methods We report on an interventionist video-reflexive ethnographic (VRE) study that explored how clinicians used the built environment to achieve safe communication in an intensive care unit (ICU) in a metropolitan Sydney hospital. We conducted 40 semistructured interviews, 5 weeks of observation and four reflexive focus groups with a total of 87 participants (including medical, nursing, allied health and clerical staff).

Results We found that the accessibility of staff and patients in the open spaces of the ICU was both a safety feature and a safety risk, enabling safe communication flow, but also allowing potentially unsafe interruptions. Staff managed interruptions while allowing for a safe degree of accessibility by creating temporary protected spaces, using physical markers such as curtains, tape and signs as well as behavioural cues, movement and the development of policies restricting activities at certain areas. Furthermore, clinicians were able to use the VRE method to gain insight into their own practices and problems, and to develop meaningful solutions for other problematic spaces.

Conclusions ICU staff enable safe communication in their wards by creating temporary spaces that are both ‘connected’ and ‘protected’. The flexibility of these ‘soft’ strategies is especially well suited to the fast-paced clinical context of intensive care.

  • Communication
  • Interruptions
  • Patient safety
  • Quality improvement methodologies

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


The field of patient safety research is nearly 15 years old, with researchers now reflecting on its achievements, failures and future directions.1 As the field becomes more sophisticated, new foci are added, and one such new focus is the built environment—namely, the physical environment of healthcare spaces, including their spatial layout, design features, furniture, equipment and other physical characteristics—and how it impacts on patient safety. A critical question in particular is how environmental features impact on communication and safety.

Effective communication is now recognised as being at the heart of patient safety.2–4 However, the evidence remains ambiguous for whether we can design environments to maximise communication effectiveness. Across commercial industries, open-plan design has been adopted to capitalise on ‘generative space’:5 space that creates new opportunities for communication and socialisation, enabling employees to resolve challenges more quickly and creatively. However, although added opportunities for staff interaction increase their ability to negotiate complex issues by accessing richer cues and providing rapid feedback,6 spatial openness also presents more opportunities for interruptions, which in healthcare may increase the risk of medical errors and patient harm.3

This article reports on a recent study conducted in a large metropolitan teaching hospital's general intensive care units (ICUs) in Sydney, Australia. Using video-reflexive ethnography, an interventionist approach to practice improvement,7 researchers worked closely with ICU clinicians to explore how space, communication and safety issues intersected in their ward.

Hospital design and patient safety

Discussions about the impact of the built environment on safety are rare in the current safety literature. However, the field of healthcare facility design contributes a comprehensive review on the impact of facility design on health outcomes.8 The review recommended features such as ventilation systems, appropriate lighting, noise control, and easy-to-clean surfaces and floor layouts, making clear that built environmental design can and does impact on patients’ safety outcomes. However, attributing unequivocal safety benefit to specific environmental attributes is difficult. For instance, carpeting may mitigate the impact of falls but is more difficult to clean and therefore more likely to harbour infectious substances than hard surfaces. Similarly, sufficient lighting is important for preventing medication errors and in supervising patients’ movements, but it can also disrupt patients’ sleep.8 Addressing the safety of healthcare environments is therefore more complicated than evaluating the safety of any particular element in isolation. Instead, we need to grapple with the multiple, changing and, at times, competing needs of clinicians and patients using these spaces.

The installation of acuity-adaptable rooms in a coronary care unit, for instance, was found to improve safety in one study by minimising patient transfers.9 Other studies have found that staff independently adapted spaces designed for other purposes, such as using multipurpose rooms as staff-only spaces10 or ordinary chairs mobilised as triage tools in emergency departments.11 In short, designs do not always produce intended or planned behaviours. Staff transform the spaces they inhabit in order to meet their needs and those of their patients and, in many cases, staff can also create the spaces they need to deliver safe care. An example of ‘space creation’ is provided by Mesman12 who described how neonatal nurses and doctors created boundaries in their work environment to protect the sterility of a central line insertion procedure. Doctors limited their mobility in the (non-sterile) ward space, while nurses prepared a protected space through a sterile field around the crib using the instrument cart, trolley and folding screens as boundary markers.

From these studies, we learn that spatial design does not necessarily predetermine how space is used, and its use can multiply the functions of a space as well as compensate for its shortcomings. Furthermore, clinicians can create safe spaces not just by manipulating the physical environment but also through their behaviour. We are interested therefore in how these resources for the manipulation of spaces are handled with regard to achieving effective and safe communication in healthcare.


This paper reports on a video-reflexive ethnographic (VRE) study that was part of a larger 3-year study, examining the intersection of space, safety and communication in healthcare. The present study was conducted over 4 months in 2011 as a learning project for participating staff in two general ICUs in a major metropolitan teaching hospital in Sydney, Australia. A total of 87 staff participated in the study.

Video-reflexive ethnography

VRE is an established approach for engaging frontline hospital staff in bottom-up and meaningful quality improvement.7 ,13 The method engages directly with participants to make sense of their own practices and contexts, beginning with ethnographic methods such as observation and interviews, followed with researcher (and/or participant) videoing of work practices and participant accounts of work. Selected clips are then shown back to participants and their colleagues in researcher-facilitated ‘reflexive sessions’, carried out as minimally-structured focus groups designed to encourage discussion on issues of concern.

VRE is an interventionist research method, aiming to deliver opportunities for participants to see their work practices from a different perspective and develop for themselves alternative practices and solutions to problems identified using the method.7 The reflexive component of VRE therefore differs from non-interventionist research approaches in its focus on engendering learning and change. The principle of ‘exnovation’ guides this process,7 ,14 focusing on the existing practices, competencies and creativity of clinicians as resources for improvement. VRE seeks to make these resources explicit, and to draw on clinicians as the experts in analysing and improving their own practices. At the focus group stage therefore, coding and analysis of reflexive session data are guided by clinician input, first drawing on the themes produced by thematic analysis carried out on earlier observational and interview data, then narrowing them in scope in order to concentrate on specific issues for discussion in subsequent sessions.

The study consisted of two overlapping phases (see table 1): phase I involved audio-recorded and transcribed interviews with staff and the observation and mapping of ICU spaces by a researcher (SH). Following analysis of data from phase I, phase II began with videoing of spaces in the ICU identified as important in phase I, and the feedback of video footage to staff in reflexive focus groups. Our guiding research questions were: do clinicians use space to manage complexity and achieve safe communication in the ICU and, if so, how?

Table 1

Study design using video-reflexive ethnography

Ethics approvals for this study were obtained from the University of Technology, Sydney, as well as the relevant local health district for this hospital. The consent process undertaken for this study was continuous, with information distributed through one-page handouts on information boards and via email. Consents for observation, interviews and videoing were obtained in person, with consents for video footage continually negotiated; for instance, consent was sought from participants prior to videoing, as well as afterwards when clips were shown to different audiences such as colleagues or management. Participants were also able to ask for video recording to be stopped at any time.


Our overarching finding was that space, safety and communication are highly intertwined in the ICU environment. We found that clinicians do use space to manage the complexity and hence safety of their work, of which communication plays a critical role. As anticipated, we also found that manipulation of space was also sometimes a matter of ‘creating’ spaces for use through ‘soft solutions’ such as policies, temporary barriers and behaviours. We report on these findings in three broad categories below: maximising accessibility, dealing with interruptions and managing accessibility using temporary protected spaces. Following these, we detail two spatial solutions that staff adopted after devising them in the focus groups in response to viewing video footage of their own practices.

Maximising accessibility for safety

A major characteristic of both ICU ward environments was their ‘openness’. Both wards had patient beds (mostly open, with a few single rooms each) arranged around a central workstation called the ‘flightdeck’. Physical accessibility through the space was a major concern: clear corridors were needed for moving beds and equipment through the ward, and nurses took care to arrange the equipment and cords around patients’ beds to allow clear access. Another aspect of accessibility related to communication needs: staff reported that they needed access to one another in the space in order to maintain contextual awareness for safety. There were two parts to this: (a) visibility for monitoring and (b) overhearing useful information.

Staff valued having clear lines of sight to monitor their patients, other patients nearby and colleagues, since patients’ conditions could change rapidly and unexpectedly. …so I stood on the corner where I can see both patients and – the likelihood [of the patient] pulling a tube out, it's not happened for the last 9–10 hours, so it's not going to happen, but there's always that chance and I don't want to be the person that walks away […] I like people to be able to see me as well, as a nurse, because if I need help, I don't want to be tucked away in the end of a quiet corridor with the cobwebs and I'm just knocking about because things change. – Interview with registered nurse

Apart from attending to non-verbal cues, staff also monitored verbal communication as a way of keeping abreast of changes relevant to them. If another team is chatting outside of the flightdeck, about a patient, the ears are pricked. – Reflexive focus group (Intensivist) … when I was training one of the other ward clerks said that I had to be as nosy as I could be […] like listening on everything, just so that I would know what was going on before someone would have to tell me and I could do my job better if I just were made aware of things and was prepared for changes. – Interview with ward clerk

Dealing with interruptions

Despite the desirability of ‘openness’ and accessibility in the ICU environment, staff also frequently experienced undesirable interruptions as a result. I can't stand it, I really can't. I don't think this is a space that's conducive to doing almost anything productive. People can walk up to the window, family members will walk up to the window, phones ring, alarms go off, the buzzers, the patient buzzer is too loud and because it's tied into the alarm system, we can't turn it down because if you turn the patient buzzer down we can't hear the alarm going off. – Interview with registrar

The drawbacks of being accessible (and hence interruptible) were easier to manage for senior staff such as consultants, registrars and nurse managers who had the option, unlike their junior staff, of leaving the wards during their shifts. Their availability however still needed to be managed to maintain safety both in the short-term and long-term. For example, it could be safer for nurses to interrupt an available consultant for advice because it was faster and more direct: Sometimes I ask the consultant because they're just as approachable as the intern. And I think, sometimes the communication from the intern will get lost to the consultant, or the same message isn't passed on. – Reflexive focus group (registered nurse)

In the longer-term however, senior staff who were too available could also have negative implications for training junior staff by limiting their opportunities to make independent decisions: … part of [leaving the ward] is that we don't get disturbances like this, not that it's bad that a nurse comes and asks a question - it's good, but I would prefer that they ask the resident the question, not because I don't want to be asked, but because otherwise the residents don't learn how to manage problems like this. – Interview with registrar

The spatial openness of the wards was therefore both a safety feature and a safety risk. The benefits for staff of facilitating connectedness and contextual awareness conflicted with the distractions and interruptions that hindered their concentration, disrupting tasks and causing mistakes and frustration. At the same time, interruptions were not necessarily unsafe; for instance, when crucial information needed to be communicated urgently or when advice was urgently needed from the most authoritative person. Dealing with interruptions could also be educational for junior staff learning how to operate in an intensive care environment.

Managing accessibility using temporary protected spaces

We found therefore that accessibility needs to be managed by ICU staff moment-to-moment and that clinicians had a range of strategies for doing so, primarily by creating what we call temporary protected spaces. For instance, clinicians sometimes withdrew into ‘protected’ spaces such as meeting rooms located within the wards but away from patient beds. These rooms were mainly used by staff for meetings, education sessions, and as private waiting and consult rooms for visiting families. Senior staff also had the option of leaving the patient-care area of the wards altogether, whereas junior medical and nursing staff were tethered to the wards.

Other temporary strategies included the drawing of curtains around beds during patient examinations, as well as using coloured duct tape on the floor and coloured signs on the poles to demarcate the bedspaces of patients with multi-resistant organisms (MROs) (when single rooms were not available). When staff wore yellow gowns in the bedspaces of patients with MROs to prevent cross-contamination, these gowns also signalled to others that those clinicians were in an infectious (and hence protected) space, and did so more clearly at times than tape on the ground or a sign placed too high to be easily seen. These strategies therefore flagged the limited accessibility of staff working inside these spaces, giving them and their patients a degree of privacy, as well as preventing cross-contamination. At the same time, because these spaces remained permeable (not completely closed off), staff were still accessible enough to maintain contextual awareness of happenings in the ward and to respond to urgent or important interruptions.

Another strategy was the use of mobility to protect a space within a moving boundary. One example of this use of mobility was the daily handover between senior nurses who moved from patient to patient as they conducted the handover in the mornings. This movement was carried out for clinical reasons, to check directly on patients during handover, but a corresponding side-effect was that the nurses experienced fewer interruptions during these mobile handovers compared with their previously static ‘paper-based’ handovers at the nursing workstation.

There were other general behavioural and technical strategies to avoid interruptions described by staff such as ‘standing up and looking busy’, ‘looking grumpy’ or, in one unpopular move, taking the phones off the hook at workstations during medical X-ray rounds. These behaviours however were thought to result in overly impermeable boundaries, potentially preventing crucial communication from taking place: If people are worried they're going to get their head bitten off, they almost always avoid having to talk to you [….] some specialists may have a reputation for being nasty to deal with and they would get called less and that may be bad for patient safety. – Interview with intensivist

Generating new spatial-behavioural solutions

Over the successive reflexive focus groups carried out in phase II, participants were invited to use the video footage and the researcher's analyses to discuss issues relating to space, safety and communication. Content-wise and thematically, the majority of discussions centred on the complex problem of managing interruptions safely, with participants encouraged to devise solutions, especially in the final focus group. As a result, two solutions were developed, targeting two open spaces in particular where activities were often interrupted in ways participants felt to be unsafe. For the nurses in one ward, it was the preparation of drugs at the controlled drugs cupboard; for the doctors in the other ward, it was their daily X-ray rounds held at one end of the central workstation. In both cases, participants devised temporary protected spaces to limit interruptions.

The nurses created a new policy restricting interruptions for nurses working at the controlled drugs cupboard. While the use of controlled drugs was already kept safe by a spatial arrangement (a locked cabinet), the area where nurses prepared drugs from the cupboard was open and visible to others in the ward. Nurses therefore were frequently interrupted by other nurses (who might be movement-restricted in isolated rooms) asking for controlled drugs to be prepared for them. Such interruptions (and their disruptive effect on the workflow and concentration of the nurses at the drug cupboard) were captured on video and shown at two focus groups. The new policy stipulated that nurses were now only allowed to get out one drug at a time from the drug cupboard, thus limiting interruptions. I think yeah we've really cut down the interruptions at the drug cupboard. ‘Cos now two people go, they're not allowed to be interrupted, go to the bedspace and then everyone else can do it. It's been really good. It's made a big difference. – Reflexive focus group 4 (Nurse manager)

As for the doctors, when they watched footage of activity at the central workstations, they identified the noisiness and interruptions that they normally experienced there, disrupting in particular the educational and clinical benefits of their weekly X-ray rounds held at one of these workstations. They then discussed how an upcoming renovation of one of their meeting rooms on the ward could make it usable for viewing patient X-rays, and decided to move their X-ray rounds into this new space as soon as the renovations were complete. I think with the new technology that's been installed here with the handovers […] might be perfect for looking at micro and X-rays. […] So that might be appropriate to do it in here. […] you're still available if there's a problem on the floor but you're still making very much clear that, look we actually need to spend our time focusing on these issues as well. – Reflexive focus group 4 (doctors)


In this study, we found that the built environment can impact significantly on communication among staff, and therefore on patient safety. In the ICU environment, the physical openness of the ward that is necessary for rapid movement and communication becomes a set-back when privacy or quietness is needed. A key finding of our study is that staff are managing their accessibility (and privacy) by using or generating a range of temporary ‘protected’ spaces with permeable boundaries; that is, spaces that are both ‘connected’ and ‘protected’. These are also ‘on-demand environments’ or environments that can be produced ad hoc and in situ.16

In these wards, we found that connectedness is not only achieved through ‘hard solutions’ associated with permanent features such as wide corridors or pillar-free wards. It is also created through ‘soft solutions’, such as bedside nurses leaving curtains open when possible, senior nurses taking care to walk around the unit and be seen, and doctors monitoring their own behaviour to come across as approachable. Likewise, limiting the accessibility of staff for privacy and safety is not only created by closing doors or by leaving the ward; it can also be created by the staff standing up and walking around, ‘looking busy’ and by instituting policies that dictate certain behaviours in certain spaces at certain times, such as those requiring personal protective equipment in the bedspaces of patients with MROs, or disallowing more than one controlled drug to be prepared at a time.

Importantly, the on-demand environments identified in this study are, by definition, temporary and permeable. This makes them uniquely suited to ICUs where urgent situations often arise that require staff to be interrupted. To impose only ‘hard’ solutions in this context could therefore be inappropriate and unsafe. The advantage of temporary and permeable boundaries is that those ‘inside’ these spaces can still attend to those ‘outside’, allowing clinicians to remain part of the network of shared contextual awareness that allows them to coordinate action quickly and responsively.

Finally, another outcome of this study was that clinicians were able to devise new soft solutions for activities that they identified as safety risks, as a result of the VRE methodology. In this study, the method delivered insights about what clinicians were already doing to manage safe communication and an opportunity for staff to draw on these strategies (now made explicit) combined with visual evidence, to develop new solutions to pre-existing problems.

Healthcare currently draws on experience-based design and codesign to enable practitioners to enrich their practice with these low-cost, self-designed and easily implemented solutions.17 Such spatial soft solutions have been observed in other ethnographic studies in creative uses of patient bedside tables and beds in space-restricted wards,18 and through careful organisation of the home environment in home-based dialysis care.19 This process also shares similarities to the human factors approach; in particular, an emphasis on understanding how the environment and broader systems interact with individuals to produce behaviours,20 as well as a focus away from individual error and training alone as causes and solutions to safety problems.21 Our approach differs, however, in locating the expertise to define and resolve safety problems with clinicians, rather than other experts—human factors or otherwise—and furthermore, aiming to establish improvement as an ongoing process integrated into everyday clinical practice rather than as a programmatic response to adverse events. Along with the principle of exnovation, this is achieved by way of facilitating an awareness in participants of how their own behaviours interact with others and their environment to produce safety (as well as risks) on an ongoing basis.7 VRE thus seeks to gain traction with the complexity inherent in healthcare systems21 by engaging with participants to assess how their interactions produce safe and/or risky situations contingently, and over time.


The limitations of our study are that it was conducted in only two intensive care wards in a single hospital, and this may limit the transferability of our findings to other units. Another limitation is that we have not been able to conduct a follow-up evaluation to assess the sustainability of the solutions proposed or of clinicians’ learning as a result of the video feedback. This lack of follow-up also prevented us from establishing if clinicians’ insights into their own spatially-mediated communication practices translated into other, still more novel, solutions. Anecdotal feedback from this and related studies suggests however that insights achieved through video reflexivity are both indelible and translatable. Insights are indelible as a result of seeing oneself on video, and they are translatable thanks to people not focusing on particular solutions, but experiencing an enhanced awareness that is applicable to all kinds of problems, not just spatial ones.7


What makes a space suitable for safe communication is not always prefigurable in healthcare, especially in volatile environments like ICUs. Instead, clinicians are reconfiguring their conducts in these spaces, and they create temporary spaces on a contingent basis. Safe communication likewise is not always predeterminable; it can mean being accessible to others for contextual awareness and quick responses to changing patients’ needs, or it can mean privacy to allow for concentration during important tasks.

Although we may wish for more certainty, the unpredictability and high context-specificity involved in determining safety is not necessarily a cause for concern in and of itself. This is because we now know that clinicians have creative ways of managing existing spaces to meet their needs. Moreover, our study suggests that clinicians are able to devise new strategies when given the opportunity to reflect on their own in situ practices. The challenge lies therefore in enabling clinicians who are expert in their own practices and well aware of their own safety and communication issues to both recognise and become more engaged in harnessing this creativity. In other words, without discounting the importance of staff engagement in the architectural design of healthcare spaces, we also need to support and allow staff to be responsive to their built environments with flexible soft solutions when hard solutions (such as renovations or new facilities) are inappropriate, too costly or unavailable for lengthy periods of time.


The authors would like to thank the participants in this study for their generous involvement in this research project.



  • Contributors SH was responsible for drafting the paper and carrying out the research described in this paper, including data acquisition, analysis and interpretation. RI and EM were responsible for the design and conception of the research study as well as significant contributions to the writing and revision of this paper.

  • Funding This research study was funded by an Australian Research Council Discovery Project Grant #DP0879002.

  • Competing interests None.

  • Ethics approval University of Technology Sydney Human Research Ethics Committee.

  • Provenance and peer review Not commissioned; externally peer reviewed.