“It’s Parallel Universes”: An Analysis of

Communication Between Surgeons and Intensivists*

Barbara Haas, MD, PhD1,2; Lesley Gotlib Conn, PhD3; Gordon D. Rubenfeld, MD, MSc1,3;
Damon Scales, MD, PhD1,3; Andre C. Amaral, MD1,3; Niall D. Ferguson, MD, MSc1,4;
Avery B. Nathens, MD, PhD2,3

Objectives: The intensivist-led model of ICU care requires surgical actionable opportunities exist to improve communication in the
consultants and the ICU team to collaborate in the care of ICU intensivist-led ICU. (Crit Care Med 2015; 43:2147–2154)
patients and to communicate effectively across teams. We sought Key Words: communication; consultants; critical care; intensive
to characterize communication between intensivists and surgeons care units; qualitative research; surgeons
and to assess enablers and barriers of effective communication.
Design: Qualitative interview study. An inductive data analysis
approach was taken.

C
Setting: Seven intensivist-led ICUs in four academic hospitals. ommunication failures between health professionals
Subjects: Surgeons (attendings and residents), intensivists are a common cause of medical errors and significantly
(attendings and residents), and ICU nurses participating in the impact on the quality of patient care (1–4). Effective com-
care of surgical patients in the ICU. munication between healthcare providers is also directly related
Interventions: None. to patient satisfaction (5). Interventions targeting improvements
Measurements and Main Results: Communication enablers and in interprofessional communication within critical care teams,
barriers existed at two distinct levels: 1) organizational and 2) cul- such as multidisciplinary rounds and daily goals, improve out-
tural. At an organizational level, participants identified that formally comes in the ICU (6–11). To date, discussions of communica-
sanctioned communication structures and processes often acted tion among healthcare providers caring for ICU patients have
as barriers to communication. Participants had developed informal largely focused on communication among ICU physicians and/
strategies to improve communication. At a cultural level, surgical and or between ICU physicians and nurses (7, 12–16).
ICU participants often expressed conflicting perspectives regarding With surgical patients accounting for more than 30% of
patient ownership, scope of practice, and clinical expertise. patients admitted to ICU, surgical teams are regular ICU con-
Conclusions: Major barriers to optimal communication between sultants (17). However, communication between ICU and
surgical and ICU teams exist in the intensivist-led ICU environ- surgical teams has largely not been studied. Indirect evidence
ment. Many are related to the structures and processes meant suggests that existing communication patterns and strategies
to facilitate communication across teams and others to how may be suboptimal for interspecialty collaboration. Conflicts
some aspects of care in the ICU are conceptualized. Multiple occur frequently in the ICU, particularly between ICU staff and
consultants (18). Medical errors and adverse events among sur-
*See also p. 2261. gical patients often occur in the ICU and many involve conflict
1
Interdepartmental Division of Critical Care, University of Toronto, Toronto, over decision making (19). Finally, previous reports suggest that
ON, Canada.
there is often discordance between surgeons’ and intensivists’
2
Division of General Surgery, Department of Surgery, University of Toronto,
Toronto, ON, Canada. perceptions and communication of prognoses (20–22).
3
Sunnybrook Research Institute, Toronto, ON, Canada. A variety of factors likely predispose surgeons and intensiv-
4
Toronto General Research Institute, Toronto, ON, Canada. ists to poor communication. First, cultural differences between
This study was performed at the University of Toronto, Toronto, ON, surgical and medical teams are well described (22). Group-level
Canada. factors, such as intergroup power differences, group norms and
Dr. Nathens was supported by funds from a Canada Research Chair Pro- values, as well as other contextual factors, may influence how
gram. The remaining authors have disclosed that they do not have any well surgeons and intensivists communicate (23). In addition,
potential conflicts of interest.
For information regarding this article, E-mail: [email protected]
the relatively recent adoption of the “high-intensity staffing” or
Copyright © 2015 by the Society of Critical Care Medicine and Wolters “intensivist-led” model of ICU care challenges long-held beliefs
Kluwer Health, Inc. All Rights Reserved. about decisional authority held by surgeons, which may in turn
DOI: 10.1097/CCM.0000000000001187 generate conflict between intensivists and surgeons (24, 25).

Critical Care Medicine www.ccmjournal.org 2147

Copyright © 2015 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.
Haas et al

Given these data, as well as the close relationship between institutions. Participants were considered to be key informants
effective provider communication and patient safety, we based on their institutional roles and medical, surgical, or nurs-
sought to characterize communication between intensivists ing specialty areas. Surgical residents and ICU fellows from the
and surgeons and to assess enablers and barriers of effective University of Toronto were included. Twenty-four interviews
communication. were scheduled and completed via criterion sampling. One
additional participant was recruited using the snowball sam-
pling technique.
MATERIALS AND METHODS
Study Design Data Collection
This was a qualitative study of communication between inten- Semi-structured interviews were conducted (Appendix 1).
sivists and surgeons across four academic hospitals. By means Semi-structured interviews are organized around themes and
of in-depth, semi-structured interviews, we sought to under- contain open-ended questions that are then supplemented
stand how members of ICU and surgical teams perceive and with questions that emerge from the dialogue between the
define the quality of communication across teams. We aimed interviewer and interviewee (27). One experienced qualitative
to identify barriers and enablers of good communication. The researcher (L.G.C.) conducted all interviews in-person; inter-
definition of “good” or “bad” communication was participant views were audiorecorded and subsequently transcribed. Data
led, rather than being established a priori by the research team. were analyzed inductively and iteratively (28). Two coinvesti-
We used a constructionist approach (26), which assumes that gators (L.G.C., B.H.) independently read and coded the first
multiple valid viewpoints are operating in the production of seven transcripts and then discussed emerging themes. Tran-
providers’ experiences and understandings of quality commu- scripts were iteratively reviewed to identify categories within
nication. This approach aims to explore these experiences and these themes, comparing them against one another, that is, the
viewpoints accepting each as valid and true. This study was constant comparison method (29). During data analysis, the
approved by the Research Ethics Board of each participating interview guide was adjusted accordingly to explore the cat-
institution. egories in more depth. Consensus between coders regarding
the major thematic findings was achieved by discussion. Data
Study Setting collection ceased when saturation was reached (30). Summary
The four participating academic hospitals are affiliated with a data were subsequently reviewed for face validity by all inves-
single university. Across these institutions, there are a total of tigators with backgrounds in critical care, surgery, and qualita-
15 ICUs and coronary care units. Due to the highly special- tive research.
ized nature of the care provided in coronary care units, cardio-
vascular ICUs, and burn units, these ICUs were excluded. The RESULTS
study focused on communication regarding patients in the In total, 25 participants were interviewed: eight surgeons,
seven remaining ICUs: two units admitting exclusively trauma seven intensivists, five critical care nurses, and five residents/
and neurosurgical patients, four mixed medical-surgical ICUs fellows (three surgery residents/two ICU fellows). Surgical
(excluding trauma patients), and one medical-surgical ICU subspecialties represented were surgical oncology, colorec-
that also admits trauma patients. tal surgery, transplant surgery, thoracic surgery, and neuro-
All included ICUs are intensivist-led, with surgical services surgery. Base specialties represented among intensivists were
typically rounding on their patients daily. All of the ICUs in internal medicine, respirology, anesthesia, and general surgery.
the study have been intensivist-led for at least 20 years. Base Among attending surgeons and intensivists, the majority had
specialties among attending intensivists in the study units are been in practice between 10 and 20 years (n = 7), followed by
internal medicine, respirology, anesthesia, and general surgery. those in practice fewer than 10 years (n = 5) and those in prac-
Two of the four centers (representing two medical-surgical tice over 20 years (n = 3).The average length of interviews was
ICUs) have no surgeon-intensivists. At the other two centers, 35 minutes, ranging from 14 to 46 minutes.
surgeon-intensivists are in the minority: four among a total of
20 intensivists at one institution and two among 15 intensiv- Defining “Good” and “Bad” Communication
ists at the other site. Residents and fellows participate in the The majority of participants were able to identify instances of
care of ICU and surgical patients at all study sites. Study ICUs high-quality communication between surgical and ICU teams,
have no written guidelines outlining communication practices and all participants were able to describe instances of poor-
between surgical teams and ICU teams. quality communication between the two teams. Participants
perceived that ICU and surgery teams had “good” communica-
Study Population tion when 1) the two teams faced a shared challenge that they
A purposive sampling strategy combining two sampling tech- overcame together and 2) the participant felt their expertise
niques was used for participant recruitment: criterion sam- was valued by the counterpart team (Table 1). Communication
pling and snowball sampling (27). The criterion sample was was perceived as “bad” when 1) the two teams were perceived
composed of key informants from the Departments of Sur- to be working toward different goals or 2) the participant felt
gery and Critical Care Medicine in each of the participating their expertise was not valued or given appropriate weight in

2148 www.ccmjournal.org October 2015 • Volume 43 • Number 10

Copyright © 2015 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.
 Clinical Investigations

Table 1. Characterization of Good Table 2. Communication Structures and

Communication Processes: Organizational Factors
Surgical team and ICU team faced a common challenge that Formal communication structures and processes
they overcame together
 Documentation in the medical record
  Surgeon: And there were no divisions between us and no
“Well, why didn’t you do this? Why didn’t you do that?” It  Paging system
was just sort of dealing with [the family’s] grief and lament.  Morning rounds
And very equally. So I thought that was a good example of
sort of satisfying communication and teamwork.  Communication networks based on hierarchy
Participant felt their expertise was valued by the Informal communication structures and processes
counterpart team
 Texting, instant messaging, e-mail
  Intensivist: So even if I don’t agree with the final
 Unplanned, face-to-face communication
management, if I feel like my concerns have been heard and
considered, then I’m usually happy with the discussion even  Communication networks based on personal relationships
if it’s not necessarily what I thought would be the outcome.

uncertainty and when there was trust in the person being con-
decision making. As such, “bad” communication often resulted tacted (e.g., the junior resident who is the first call is known to
in the participant’s perception that the patient had received be reliable by the ICU team). If any one of these criteria were
inappropriate care. For example, a surgeon explained that absent, formal communication channels were abandoned in
when his patient received dialysis in the ICU without his con- favor of informal ones.
sultation, “The resident or fellow in the ICU stuck in the dialy- The existence of both formal and informal practices often
sis line in the middle of the night ‘cause the creatinine went up led to redundant or incomplete information transfer across
a little bit. [That] patient has had an unnecessary procedure and within teams. Participants expressed a desire to implement
from our point of view. And one that carries potential risks, more effective, formal communication structures that would
which, if they were under our care, we would never have done incorporate features of existing informal practices, in particu-
because there’s no reason to do it.” lar relating to the timeliness of communication. As one sur-
geon reflected, “I mean, if rioters in London can communicate
Enablers and Barriers to Good Communication with each other on a minute to minute basis then we should be
Thematic findings revealed that the quality of ICU-surgery able to do better as well.”
team communication was viewed to be influenced by both Formal Communication Structures and Processes. Docu-
organizational and cultural factors that were both tacit and mentation in the medical record. The primary formal com-
explicit. These factors influenced whether participants per- munication mode between teams was documentation in the
ceived teams to be working together or at cross purposes and medical record. However, both intensivists and surgeons
whether participants perceived their expertise was valued or expressed a high degree of dissatisfaction with written notes as
not by the counterpart team. When reviewed, these factors fell a means of interteam communication. The majority of inten-
into one of two categories: 1) communication structures and sivists expressed frustration with the quality of notes written
processes (organizational factors) and 2) competing constructs by the surgical team in the medical record, as intensivists often
of care (cultural factors). used these notes to gain insight into a surgical team’s plan.
Intensivists felt that surgical notes focused on recording labo-
Communication Structures and Processes: ratory information, vital signs, and ventilator settings, without
Organizational Factors including information about the surgical plan. Surgical partic-
Participants described two parallel communication systems ipants did not view the patient chart as a means by which care
(Table 2): the formal structures and processes as managed by plans are communicated; medical records were seen as a legal
institutions, such as documentation in the medical record, the document providing evidence that the patient had been seen.
paging system, and team hierarchy, and the informal commu- Paging system. In addition to the written patient record, the
nication system, such as texting and instant messaging. Gen- second official interteam communication mode was paging.
erally, participants disfavored the formal systems designed to Participants noted, however, that pages were often answered
enable interteam communication; the informal systems were in a significantly delayed manner (particularly when surgeons
preferred and were used to bypass the more formal communi- were in the operating room), were sometimes not answered,
cation channels. The formal system of communication was felt and often resulted in the wrong person being contacted (e.g.,
to significantly increase the likelihood that the ICU and surgi- the wrong surgical team). While participants described efforts
cal teams were working at cross purposes. to use this official means of communication, paging was not
Participants understood that they “should” be using formal viewed as enabling effective communication.
communication channels, but used them only in low-priority Morning rounds. In virtually all cases, the only time point
situations, when there was a minimal level of decision-making at which surgical and ICU teams interacted predictably was

Critical Care Medicine www.ccmjournal.org 2149

Copyright © 2015 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.
Haas et al

during surgical teams’ morning rounds. Typically, these rounds quality of interteam communication (Table 3). Specifically,
involved trainees from the surgical team, but not the attending participants made frequent references to patient ownership in
physicians. Among ICU team members, only the bedside criti- the intensivist-led unit and to the role of surgeons’ and intensiv-
cal care nurses were consistently present for these rounds. In ists’ expertise in the care of surgical ICU patients. When agree-
general, after surgical teams had made their morning rounds, ment existed between surgeons and intensivists in defining
either the bedside nurse or post-call ICU fellow were expected these constructs, participants expressed satisfaction with the
to act as an intermediary between surgical teams and the day- quality of communication with their colleagues. Conversely,
time ICU team; participants felt this practice led to informa- competing definitions of these key constructs were associated
tion loss and did not permit collaborative interactions. with perceptions of poor interteam communication.
Communication networks based on hierarchy. Formal Patient Ownership. The majority of intensivists acknowl-
communication processes followed a clear hierarchy. If either a edged a special relationship between surgeons (as compared to
surgical or ICU team had an inquiry for the counterpart team, internists) and their patients (Table 3). Some characterized this
the sanctioned process described by participants involved relationship in a positive way, whereas others described it nega-
paging or otherwise contacting a junior member of the coun- tively as causing surgeons to be overly involved in the ICU care
terpart team. The junior physician contacted would then con- of surgical patients. There was a clear tension between surgeons’
tact their senior resident or fellow, who would then contact sense of patient ownership and the parameters of the intensivist-
the attending physician. Although participants acknowledged led ICU (Table 3). Additionally, the degree to which intensivists
that this was the “right” thing to do, the hierarchical escalation were willing to accommodate for surgical input varied greatly; as
of inquiries led to delays; such delays were associated with the a result, “intensivist-led” care existed on a spectrum from open
perception that teams were working at cross purposes. and flexible to closed and rigid. Surgeons were significantly
Informal Communication Structures and Processes. more satisfied with communication when they perceived a less
­Texting, instant messaging, and e-mail. Most participants rigid approach to the intensivist-led model of care.
preferred using text messaging, e-mail, or other direct mes- The tension between the intensivist-led model of care and
saging communication technology, despite the fact that tex- surgeons’ sense of patient ownership was highlighted during
ting and e-mail might not protect patients’ personal health
end-of-life care. The majority of intensivists described involv-
information.
ing surgeons in end-of-life discussions in order to preserve
Unplanned, face-to-face communication. Face-to-face
surgeons’ sense of ownership. However, some intensivists
communication was universally acknowledged as the most
maintained a rigid approach to the intensivist-led model of
favored mode for quality communication. All participants
care even surrounding end-of-life care of critically ill surgical
highly valued the opportunity for the ICU physicians and sur-
patients. Surgeons perceived exclusion from end-of-life deci-
geons to interact during the day, when attending physicians
sion making for their patients as the most significant violation
from both teams were present. Such interactions only occurred
of their sense of patient ownership and considered it a prime
informally, however, and were unpredictable. Despite this
example of poor communication between teams.
strong preference, participants noted that there were few or no
structures in place to facilitate face-to-face communication. The Role of Surgeons’ and Intensivists’ Expertise. Participants
Face-to-face communication occurred opportunistically and identified specific areas of patient care that were outside their
was largely dependent on chance and personal effort by indi- own scope of expertise, regardless of who held the status of most
vidual participants. Virtually all participants expressed a desire responsible (i.e., attending) physician. For example, intensivists
for joint bedside rounds between surgery and ICU teams. generally felt that decisions regarding enteral nutrition follow-
Communication networks based on personal ­relationships. ing bowel surgery should be left to surgeons, and surgeons felt
Despite acknowledging that the formal communication pro- that decisions regarding mechanical ventilation should be left to
cess followed a hierarchy, participants frequently circumvented intensivists. When participants perceived that ICU and surgical
this. Participants overwhelmingly preferred communication team members had each participated in care to a degree com-
with individuals with whom they were familiar based on pre- mensurate with these expectations regarding scope of expertise,
existing, personal relationships. As such, participants identified such comanagement was described as good communication.
the high turnover of trainees and the lack of familiarity with There were, however, significant discrepancies between how
certain colleagues as major barriers to good communication. clinicians described their own expertise and how their exper-
The informal communication network frequently excluded tise was viewed by others. Both surgeons and intensivists felt
residents and fellows because they were unfamiliar to mem- that their counterparts understood only a narrow aspect of
bers of the counterpart team. In addition, attending physicians their patients’ clinical situation, but that they themselves had
would avoid communicating with other attending physicians if a better sense of the patient’s overall trajectory and progno-
they did not have a good interpersonal relationship with them. sis. Differences in how surgeons and intensivists defined their
scope of expertise were a significant source of tension. Many
Competing Constructs of Care: Cultural Factors participants described instances in which other clinicians
Several conceptual constructs were repeatedly referenced by attempted to communicate information or suggestions that
participants and appeared to strongly influence perceived were perceived to be beyond their scope of expertise; instances

2150 www.ccmjournal.org October 2015 • Volume 43 • Number 10

Copyright © 2015 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.
 Clinical Investigations

Table 3. Competing Constructs of Care: Cultural Factors

Patient ownership
 Special relationship between surgeons and patients (positive and negative characterization)
  Intensivist: And I think they feel like they have, and appropriately, an ownership of the patient because it’s their patient who is
now in the ICU. And they stay obviously involved and they care.
  Intensivist: Because [the surgeons have] already had their hands in the patients so to speak and they have an ownership issue there.
 Spectrum of definitions for intensivist-led model of care
   Critical care nurse: So we have what we like to term an “openly closed” unit [laughs].
  Surgeon: I think that, number one, is that our closed unit actually operates a lot like an open unit, not in name but actually
in function so the intensivists listen carefully to what we have to say, we listen carefully to what they have to say and our
suggestions are taken seriously and listened to.
  Intensivist: So we did have a situation a while back where the Chief of General Surgery at the time said to the nurse, “You can
pull out the nasogastric tube” and the nurse said, “No, I’ve gotta check with my team.” The surgical intern [rotating through the
ICU] got all embarrassed ‘cause he was being asked to check on the Chief of Surgery’s decision. So that’s how closed we are.
 Highlighted sense of patient ownership by surgeons during end-of-life care
  Surgeon: The other frustration is with decisions about withdrawal of care ‘cause that’s fairly frequent. If you just did some huge
operation that took you 18 hours and now four days later, somehow there’s a withdrawal of care, how did that happen? Who
spoke to the patient’s family?
The role of surgeons’ and intensivists’ expertise in the care of surgical ICU patients
 Adherence to perceived scope of expertise perceived as good communication
  Surgeon: You get the sense that you’re working for the collective good of looking after a patient. And the roles are fairly
well-defined.
 Perception that counterpart team understands only a narrow aspect of patient care
  Surgeon: [Describing disadvantages of postoperative ICU care] You want them to get a post-op suite of care. You don’t want
them to get “Oh, let’s give him a little metoprolol and let’s give him some adrenaline.” I think that there are different types of care.
  Intensivist: And [surgeons] play the role of answering specific questions that are related to the surgical plan of care. And then
our job is to put the big picture together and work out a care plan that’s patient-centered and values-based and goal-directed.
 Surgeon-perceived exclusion from important aspects of patient care
  Surgeon: [name of intensivist] often says “We don’t need you to talk about the ventilator, the inotropes. We just wanna know what
to do with the NG tube.” I don’t think we do want to just think about the NG tube. I think we wanna think about all aspects of patient
care. And we have things that we could learn from them about that and perhaps we could also impart some of what we know.

where other clinicians crossed an expertise barrier were per- DISCUSSION

ceived as highly disruptive (Table 3). We performed an exploratory analysis of communication
Whereas surgeons’ perceived limitations of intensivists’ between surgeons and intensivists across seven ICUs in four
expertise were described in general terms (“If it’s something to academic centers. Formal structures and processes meant to
do with the surgery, then we obviously know more about it that facilitate communication across hospital teams fall short of
then they do”), intensivists expressed relatively precise catego- promoting good-quality communication. Critical care nurses,
ries of care which they felt were and were not within surgeons’ intensivists, and surgeons have developed informal communi-
scope. For example, information they sought from the surgical cation practices that they prefer and associate with improved
interteam communication. These informal practices depend
team was “Whether or not the patient can eat that particular day
primarily on individual provider motivation and existing
or be fed enterally. Whether or not one can start anticoagulants
interpersonal relationships, leading to their use only on an
and aspirin and those sorts of things.” Significant differences
opportunistic and unpredictable basis.
emerged between surgeons and intensivists in how they felt Although an intensivist-led model of care is associated
their colleagues perceived them. Surgeons described a sense of with a significant reduction in mortality among critically
exclusion from aspects of patient care that they felt were impor- ill patients (31, 32), our data demonstrate that the inten-
tant and in which they wished to participate. Surgeons’ sense of sivist-led ICU environment has significant, unintended
exclusion was closely associated with feelings of frustration and consequences on interspecialty team communication. The
mistrust in the ICU team and was perceived by surgeons as one intensivist-led ICU directly challenges surgeons’ traditional
of the most important causes of poor interteam communication. view of their relationship with their patient, in which they

Critical Care Medicine www.ccmjournal.org 2151

Copyright © 2015 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.
Haas et al

have a “covenant to cure” (22) and should “hold themselves These data emphasize the influence of group identity and
entirely accountable for the outcomes of their patients” (25). intergroup dynamics on communication between surgeons
Surgeons described attempts to integrate their professional and intensivists.
worldview with the intensivist-led ICU model. Nevertheless, This study has a number of limitations. Our interviews
we identified significant discrepancies in how members of with surgeons did not include data from all specialties. It is
surgical and ICU teams operationalize this model and how possible that the communication practices described in this
they value the spectrum of expertise required to care for criti- study do not fully capture the variety of practices across all
cally ill surgical patients. The lack of a common understand- surgical specialties. In addition, our study did not specifi-
ing of these concepts contributed significantly to perceptions cally seek to identify differences in communication practices
and experiences of poor communication and dissatisfaction across general surgery, thoracic surgery, and neurosurgery.
with the counterpart team. As a result, there may be nuanced differences in how each of
Improving patient safety by improving teamwork within the these subspecialties interact with ICU teams that are not cap-
ICU has been the focus of a large number of studies (4, 33). tured. Tensions may also exist between ICU teams and other
In particular, the development of the interprofessional team, healthcare providers (e.g., medical oncologists); the possibil-
consisting of physicians, nurses, and other healthcare profes- ity of such tensions is not addressed in this study. However,
sionals, has become the cornerstone to patient care in the ICU. we hypothesize that, given the unique nature of surgical
Facilitating teamwork between members of this team and iden- culture and the “surgical personality” (22), barriers to high-
tifying methods to ensure all team members are working toward quality communication between ICU and surgical teams are
similar patient goals have been the subject of many reports (3, likely unique.
4, 6, 9, 14, 16, 33). These data also frequently demonstrate that Given that the interview guide was designed and modified
interventions can improve how individuals within the ICU team around the primary objective of our study, there are areas of
communicate and collaborate, despite a broad range of pro- interest where we did not achieve data saturation, which may be
fessional and group identities (6, 8–11, 34). Notwithstanding worthy of further study. For example, we did not have adequate
advances made by ICU teams to provide patient-centered, data to comment on the relative differences between surgeon-
interprofessional care, our data suggest that communication intensivists and other intensivists or the relative importance of
between ICU and surgical teams is fraught with tension—likely each type of communication barrier across different contexts.
to the detriment of communication with critically ill patients Finally, as with all qualitative research, there is an inherent
and their families. balance between the in-depth nature of our inquiry and the
Several studies have assessed communication between sur- generalizability of our data. Clearly, not all of our findings are
geons and intensivists; however, these data examined com- applicable to all intensivist-led ICUs, since institutions likely
munication at a single point in time (35–39). Furthermore, vary significantly in how they operationalize the intensivist-led
these studies defined effective communication as accurate ICU model. Given that this study was conducted in academic
transmission of information in a linear fashion (40). Such centers, our findings may not be generalizable to community
an approach ignores the way in which communication across centers. Nevertheless, the overarching themes and concepts we
teams not only serves to transfer information but also defines have identified are likely to be relevant to many interactions
the social context in which people work and the “spoken and between surgeons and intensivists.
unspoken frameworks the team develops regarding appropri-
ate goals, roles, and behaviors” (40). Our study focused on CONCLUSIONS
this broader, “social construction” view of communication; Our data suggest that significant opportunities exist in improv-
this approach is particularly relevant to interactions between ing communication between surgeons and intensivists in
intensivists and surgeons, where nuanced, close collaboration intensivist-led ICUs. First, existing institutional structures and
across disparate medical cultures is required for the optimal processes to facilitate communication across teams are inad-
care of complex patients and their families. Notably, in our equate. Approaches that harness and formalize existing infor-
data, participants did not define good or bad communication mal communication networks should be explored. For example,
simply through the lens of information transfer but rather as including attending physicians in joint rounds and replacing
a complex process of interpersonal and intergroup relation- texting with hospital-wide microblogging might serve to facili-
ship building. tate daily, structured, meaningful interactions and foster inter-
Both surgeons and intensivists had clear beliefs about their personal relationships. In addition, cultural differences between
own group identity and role and strongly identified with their surgeons and intensivists must be addressed if collaboration
own group. Participants demonstrated attribution bias when is to improve. The creation of the ICU interprofessional team
describing the behavior among members of the counterpart approach required the acknowledgment and breakdown of cul-
group (41). Negative behaviors were attributed to peoples’ val- tural, professional, and historical barriers between physicians,
ues and their lack of consideration or lack of understanding. nurses, and other health professionals; in a similar fashion, these
Even when institutional structures and processes were acting as same barriers between surgeons and intensivists will need to be
barriers to good communication (e.g., paging system), partici- addressed in order to optimize the intensivist-led model of criti-
pants felt that the counterpart team should make more effort. cal care delivery for surgical patients.

2152 www.ccmjournal.org October 2015 • Volume 43 • Number 10

Copyright © 2015 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.
 Clinical Investigations

REFERENCES 21. Aslakson RA, Wyskiel R, Shaeffer D, et al: Surgical intensive care
1. Proctor ML, Pastore J, Gerstle JT, et al: Incidence of medical error and unit clinician estimates of the adequacy of communication regarding
adverse outcomes on a pediatric general surgery service. J Pediatr patient prognosis. Crit Care 2010; 14:R218
Surg 2003; 38:1361–1365 22. Penkoske PA, Buchman TG: The relationship between the surgeon
2. Haller G, Myles PS, Taffé P, et al: Rate of undesirable events at and the intensivist in the surgical intensive care unit. Surg Clin North
beginning of academic year: Retrospective cohort study. BMJ 2009; Am 2006; 86:1351–1357
339:b3974 23. Kreindler SA, Dowd DA, Dana Star N, et al: Silos and social identity:
3. Reader T, Flin R, Lauche K, et al: Non-technical skills in the intensive The social identity approach as a framework for understanding and
care unit. Br J Anaesth 2006; 96:551–559 overcoming divisions in health care. Milbank Q 2012; 90:347–374
4. Reader TW, Flin R, Cuthbertson BH: Communication skills and error 24. Schwarze M, Bradley C, Brasel K: Surgical “buy-in”: The contractual
in the intensive care unit. Curr Opin Crit Care 2007; 13:732–736 relationship between surgeons and patients that influences decisions
5. Chow A, Mayer EK, Darzi AW, et al: Patient-reported outcome mea- regarding life-supporting therapy. Crit Care Med 2010; 38:843–848
sures: The importance of patient satisfaction in surgery. Surgery 25. Buchman TG, Cassell J, Ray SE, et al: Who should manage the dying
2009; 146:435–443 patient? Rescue, shame, and the surgical ICU dilemma. J Am Coll
6. Rehder KJ, Uhl TL, Meliones JN, et al: Targeted interventions improve Surg 2002; 194:665–673
shared agreement of daily goals in the pediatric intensive care unit. 26. Guba EG, Lincoln YS: Competing paradigms in qualitative research.
Pediatr Crit Care Med 2012; 13:6–10 In: Handbook of Qualitative Research. Denzin NK, Lincoln YS (Eds).
7. Sexton JB, Berenholtz SM, Goeschel CA, et al: Assessing and Thousand Oaks, CA, SAGE, 1994, pp 105–117
improving safety climate in a large cohort of intensive care units. Crit 27. Bernard HR: Research Methods in Anthropology: Qualitative and Quan-
Care Med 2011; 39:934–939 titative Approaches. Fifth Edition. Lanham, MD, Altamira Press, 2011
8. Kim MM, Barnato AE, Angus DC, et al: The effect of multidisciplinary 28. Bernard HR, Ryan GW: Analyzing Qualitative Data: Systematic
care teams on intensive care unit mortality. Arch Intern Med 2010; Approaches. Thousand Oaks, CA, Sage, 2010
170:369–376 29. Strauss A, Corbin J: Basics of Qualitative Research. Los Angeles,
9. Jain M, Miller L, Belt D, et al: Decline in ICU adverse events, nosoco- CA, Sage, 2008
mial infections and cost through a quality improvement initiative focus- 30. Morse JM: The significance of saturation. Qual Health Res 1995;
ing on teamwork and culture change. Qual Saf Health Care 2006; 5:147–149
15:235–239
31. Pronovost PJ, Angus DC, Dorman T, et al: Physician staffing patterns
10. Shaw DJ, Davidson JE, Smilde RI, et al: Multidisciplinary team training and clinical outcomes in critically ill patients: A systematic review.
to enhance family communication in the ICU. Crit Care Med 2014; JAMA 2002; 288:2151–2162
42:265–271
32. Wilcox ME, Chong CA, Niven DJ, et al: Do intensivist staffing patterns
11. Dodek PM, Raboud J: Explicit approach to rounds in an ICU improves influence hospital mortality following ICU admission? A systematic
communication and satisfaction of providers. Intensive Care Med review and meta-analyses. Crit Care Med 2013; 41:2253–2274
2003; 29:1584–1588
33. Reader TW, Flin R, Mearns K, et al: Developing a team performance frame-
12. Thomas EJ, Sexton JB, Helmreich RL: Discrepant attitudes about work for the intensive care unit. Crit Care Med 2009; 37:1787–1793
teamwork among critical care nurses and physicians. Crit Care Med
2003; 31:956–959 34. Petrovic M, Martinez E, Aboumatar H: Implementing a perioperative
handoff tool to improve postprocedural patient transfers. Jt Comm J
13. Boyle DK, Kochinda C: Enhancing collaborative communication of Qual Patient Saf 2012; 38:135–142
nurse and physician leadership in two intensive care units. J Nurs
Adm 2004; 34:60–70 35. Agarwal HS, Saville BR, Slayton JM, et al: Standardized postopera-
tive handover process improves outcomes in the intensive care unit:
14. Reader TW, Flin R, Mearns K, et al: Interdisciplinary communication in A model for operational sustainability and improved team perfor-
the intensive care unit. Br J Anaesth 2007; 98:347–352
mance. Crit Care Med 2012; 40:2109–2115
15. Cardarelli M, Vaidya V, Conway D, et al: Dissecting multidisciplinary
36. Karakaya A, Moerman AT, Peperstraete H, et al: Implementation of a struc-
cardiac surgery rounds. Ann Thorac Surg 2009; 88:809–813
tured information transfer checklist improves postoperative data transfer
16. Pronovost P, Berenholtz S, Dorman T, et al: Improving communication after congenital cardiac surgery. Eur J Anaesthesiol 2013; 30:764–769
in the ICU using daily goals. J Crit Care 2003; 18:71–75
37. Joy BF, Elliott E, Hardy C, et al: Standardized multidisciplinary pro-
17. Wunsch H, Wagner J, Herlim M, et al: ICU occupancy and mechan- tocol improves handover of cardiac surgery patients to the intensive
ical ventilator use in the United States. Crit Care Med 2013; 41: care unit. Pediatr Crit Care Med 2011; 12:304–308
2712–2719
38. Segall N, Bonifacio AS, Schroeder RA, et al: Can we make postop-
18. Azoulay E, Timsit JF, Sprung CL, et al; Conflicus Study Investiga- erative patient handovers safer? A systematic review of the literature.
tors and for the Ethics Section of the European Society of Inten- Anesth Analg 2012; 115:102–115
sive Care Medicine: Prevalence and factors of intensive care unit
conflicts: The Conflicus study. Am J Respir Crit Care Med 2009; 39. Zavalkoff SR, Razack SI, Lavoie J, et al: Handover after pediatric heart
180:853–860 surgery: A simple tool improves information exchange. Pediatr Crit
Care Med 2011; 12:309–313
19. Gawande AA, Zinner MJ, Studdert DM, et al: Analysis of errors
reported by surgeons at three teaching hospitals. Surgery 2003; 40. Eisenberg EM: The social construction of health care teams. In: Improving
133:614–621 Healthcare Team Communication: Building on Lessons From Aviation and
Aerospace. Nemeth CP (Ed). Burlington, VT, Ashgate, 2009, 9-202008
20. Aslakson RA, Wyskiel R, Thornton I, et al: Nurse-perceived barriers to
effective communication regarding prognosis and optimal end-of-life 41. Turner RN, Hewstone M: Attribution biases. In: Encyclopedia of
care for surgical ICU patients: A qualitative exploration. J Palliat Med Group Processes and Intergroup Relations. Levine JM, Hogg MA
2012; 15:910–915 (Eds). Thousand Oaks, CA, SAGE, 2009, pp 43–46

Critical Care Medicine www.ccmjournal.org 2153

Copyright © 2015 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.
Haas et al

Appendix 1. Interview Guides Interview guide for members of surgery teams:

Interview guide for members of ICU teams: 1. Can you please tell me about your training and on the job
experience?
1. Can you please tell me about your training and on the job 2. Can you describe a typical scenario when a surgery team
experience with patients in the ICU? becomes involved in the care of an ICU patient?
2. Can you describe a typical scenario when a surgery team 3. How do you communicate information regarding an ICU
becomes involved in the care of an ICU patient? patient to ICU teams?
3. How do you communicate information regarding an ICU 4. How do ICU teams typically communicate with you about
patient to surgery teams? ICU patients?
4. How do surgery teams typically communicate with you 5. Overall, how would you describe the quality of communi-
about ICU patients? cation that happens between ICU teams and surgery teams?
5. Overall, how would you describe the quality of communi-
cation that happens between ICU teams and surgery teams? a. When do you think communication between the two
a. When do you think communication between the two teams is at its best? EXAMPLE please? What specific fac-
teams is at its best? EXAMPLE please? What specific fac- tors influenced the situation?
tors influenced the situation? b. When/where do you think there are opportunities to
b. When/where do you think there are opportunities to improve communication? EXAMPLE please? What are
improve communication? EXAMPLE please? What are the specific factors that hindered the situation?
the specific factors that hindered the situation?
6. In your experience, over the past X number of years, how
6. In your experience in the ICU, over the past X number of has communication between ICU teams and surgery teams
years, how has communication between ICU teams and changed?
surgery teams changed? 7. In an ideal world, if you could instantly change one thing
7. In an ideal world, if you could instantly change one thing about communication between the ICU and surgery teams,
about communication between the ICU and surgery teams, what would that be?
what would that be? 8. Is there anything else you would like to add about commu-
8. Is there anything else you would like to add about commu- nication between ICU and surgery teams, or about the ICU
nication between ICU and surgery teams, or about the ICU environment generally, that would be helpful for us to bet-
environment generally, that would be helpful for us to bet- ter understand this topic?
ter understand this topic?

2154 www.ccmjournal.org October 2015 • Volume 43 • Number 10

Copyright © 2015 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.