Many educators and researchers have investigated how interprofessional communication is linked both directly and indirectly to issues of patient safety and medical error.1–7 Many of these investigations were prompted by the Institute of Medicine's (IOM's) report, To Err Is Human. The IOM report suggested that analysis of the causes of error should shift focus from blaming individuals to analyzing system failures. This shift reflects a trend in the patient-safety literature that borrows from Reason's work distinguishing between two types of error: active and latent failures.6,8,9 Active failures are defined as errors made by those at the “sharp end”8 of practice, by individuals who are actively performing a task.8 In active failures, the effect of the error is felt almost immediately.8 Investigators of this approach tend to treat errors as moral issues. For these researchers, errors happen to people with aberrant mental processes such as forgetfulness and negligence (i.e., bad things happen to bad people).9 In contrast, latent failures are described as system-based failures that lie dormant within a system, only becoming evident when local triggers activate the latent condition.8 Investigators using this approach tend to focus their analysis and preventative efforts on the latent conditions that enable the error. Latent errors are seen as consequences of “upstream” systemic factors that enable errors, and not as originating in the perversity of human nature.9 To understand the role of interprofessional communication in medical error, from the latent error perspective, is to understand how interprofessional communication systems and structures enable and support the generation or the prevention of latent errors.
Despite increased research activity related to interprofessional communication and medical errors, progress in increasing safety has been slow. Investigators have proposed that little evidence exists demonstrating that systemic improvements in safety are widely available.6 When analyzing the barriers limiting progress, the first obstacle impeding progress is complexity.6 Specifically, researchers have proposed that understanding the complex nature of health care work, where there are often more than 50 different types of medical specialties and subspecialties interacting with each other and an equally large array of allied health professions, remains an elusive goal.6,10 They posit that this intra- and interprofessional complexity, compounded with a tradition of professional fragmentation, of individualism, of a well-entrenched hierarchical authority structure, and of diffuse accountability, forms a daunting barrier to improving safety.6 This barrier has hindered progress toward common-purpose teamwork and successful interdependence between and within the professions that a safety-oriented culture requires.6 A recently published literature review proposes that interdisciplinary communications are an underexplored source of medical errors and, more specifically, of latent errors.7 However, although the studies reviewed cite communication as a perceived important source of latent error, many offer only generic “improve teamwork communication” advice.7
Analysis of the literature investigating interprofessional communication and the link to medical errors reveals that studies in this field tend to be narrow in scope. They do not take on the challenge of addressing the complexity of health care work. Complexities such as team membership fluidity, rotation, and multiple forms of communication are generally overlooked. Instead, studies of interprofessional communication tend to place artificial boundaries on the setting and to place artificial emphasis on particular team members or communication media.1,11–23 For example, one study examines the communication practices of physicians and nurses, on one inpatient ward, as carried out through one paper document.22 Although this study successfully investigates this particular scenario, it leaves unanswered important systemic considerations such as the communication structures and practices that exist beyond the ward, the interactions with the multiple other health care professionals involved in care, and the many other forms of communication (electronic, oral, and other paper documents) involved in supporting patient care. These limitations are evident in much interprofessional communication and safety research where findings tend to generate local area solutions (color code patient chart pages1 or allow electronic patient record screens to be printed22) or vague dictums (provide nurses with more information24 or improve team communication25) to address system-wide problems. Although these studies have contributed to our understanding of how interprofessional communication relates to medical error, they also have limited our understanding of the contextual complexities in which interprofessional communication is supporting the generation of latent errors.26–28 But how can we take this complexity into consideration in research?
The purpose of this paper is to investigate how research into interprofessional communication and latent medical errors can take into account the broad-scope complexity of medical work. To meet this goal requires a reframing of analytic approaches and of research questions as they have traditionally been structured within the medical domain.10 To build these new research frames, our search led us to consider theoretical perspectives and data from other domains (including rhetoric, sociology, management, and computer-supported collaborative work). Through this search, we identified a number of theoretical perspectives that could support a broad-scoped examination of interprofessional communication and latent medical errors in relation to the complexity of medical work. After reviewing several of these theories, we identified two theories from the social sciences that would best support our reframing efforts: Activity Theory (AT) and its related subtheory, Knotworking. This paper reviews and summarizes these theories to generate new, theoretically informed lenses through which to reexamine the connections between interprofessional communication and latent medical errors. This paper describes how, through these lenses, new frameworks for asking (and answering) research questions can bring contextual complexity into the scope of research. Specifically, the contextual complexities that we examine relate to issues of (1) the social and personal contexts that shape the work activities of individual members of health care teams, (2) the distributed nature of patient information, and (3) the increasing use of communication technologies and media. Such complexities will have to be explored in depth if we are to adequately prepare students and practicing clinicians to work safely in interprofessional teams.
AT is a theory originally rooted in the work of Soviet psychologists Vygotsky29 and Leont'ev30 and, more recently, in the work of the Finnish researcher Yrjo Engeström.31–36 AT supports a richly nuanced conceptualization of the social interactions and relationships involved in interprofessional health care practices. AT proposes that analysis of complex social encounters, like those involved in interprofessional communication, should take into consideration entire activity systems (see Figure 1). An activity system consists of six elements: individuals, objectives, tools, communities, rules, and division of labor. An activity system models the multifaceted interactions that take place between individual humans and the social, physical, and organizational structures involved in professional work.37 Activity system-based analysis seeks to describe how individuals use tools to achieve specific objectives. An individual's achievement of these objectives is shaped and influenced by the rules and norms of his or her social communities, organizational affiliations, and by his or her conceptions of appropriate division of labor.
To illustrate, take, for example, a third-year general internal medicine resident working on an inpatient nephrology ward. The ways in which this individual resident participates in patient care is informed by the elements of his or her activity system. First, the resident is working toward the realization of specific objectives. One objective, and arguably the most important, is to provide good medical care to the patient. There are, however, other objectives influencing the resident's work. These other competing objectives may be to seem competent in front of the senior staff or to finish inputting order entries before leaving for an education session. These multiple objectives shape and influence the resident's work activities. For instance, the resident will work to provide good medical care, but may be reluctant to admit uncertainty or to request staff support. The resident may also be hurrying to complete order entries quickly.
Second, the resident works with specific tools to achieve these objectives. These tools might include, for instance, a stethoscope, a paper patient chart, an electronic order-entry system, a dialysis machine, lab requisition forms, and a telephone. While each tool is necessary to the completion of the resident's objectives, the resident will have varying levels of competence with each tool, which will shape his or her care activities. For instance, the resident may not be experienced at inputting complex pharmaceutical order entries into the electronic system. He or she may need to both improvise at the order-entry screen and to clarify orders orally with the nursing staff.
Third, the resident's care activities will also be affected by his or her affiliations with social and organizational communities. The resident may strongly associate with professional colleagues and supervisors and with the community of other residents. The resident's sense of community will also be informed by personal connections. Each of these communities has the potential to shape the resident's activities. For instance, the resident's affiliation with professional colleagues will support the resident's objective to care for the patient by carrying his or her share of the care workload. However, the resident may also be affiliated with the residency union, which insists that all residents leave post call. These two social affiliations can conflict in post call situations where the resident feels community pressures both to stay to help with the care work load and to leave to adhere to union rules. The resident will have to negotiate these conflicting community ties to meet his or her professional objective to care for the patient.
Fourth, there are several rules from each of the resident's communities that will shape his or her work activities. These rules are not all agreed on or accepted, and some are overtly stated while others are tacitly understood; however, these rules are felt and recognized to varying degrees by the resident. Within the resident's community of medical professionals, rules will include conditions about when it is appropriate to page the staff for advice, what and how many medications can be prescribed for a patient, and what is diagnostically relevant information. To carry out the objective to care for the patient, the resident will have to negotiate his or her care activities through these rules.
Finally, the resident is part of a division of labor within each community. That division of labor shapes the distribution of power, tasks, and responsibilities among collaborators. Educational and clinical experiences will shape the resident's expectations about his or her role on the ward in relation to other health care professionals. The experiences that the resident had on other wards and in other care settings will shape what he or she considers to be appropriate divisions of labor both within the medical profession and interprofessionally.
To understand the resident's interprofessional communication work activities, AT argues that all communicative actions should be examined through the resident's activity system. And, more broadly, to understand the communication activities of each member of an interprofessional health care team, AT requires that we examine the communications of each team member through his or her individual activity system. By understanding how interprofessional communications are influenced by multiple activity systems, research can uncover the layers of personal, professional, organizational, social, and physical contexts that influence the actions and interactions involved in interprofessional communication. With such research, educational interventions can better prepare clinicians to negotiate their way through the multifarious relations that complicate interprofessional communication.36
Whereas AT's model of activity systems supports the critical study of the complex relations involved in interprofessional communication, it does not model how health care team membership often involves high rotation within each profession. What is needed is a complementary theoretical framework for understanding how, within each profession involved in the care team, multiple activity systems rotate through and participate in patient care. This theoretical approach should also highlight how interprofessional communication can avoid and/or generate latent medical errors within the team structure. To fill this gap, we turn to the theory of Knotworking.38
Health care team communications regularly involve the participation of different professionals at different times. To incorporate analysis of these continually changing teams, we draw on the theory of Knotworking.38–40 Knotworking, a fairly recent development within AT, describes collaborative work as situations involving constantly changing combinations of individuals distributed over time and space. When viewed through the lens of Knotworking, health care teamwork consists of distributed and partially improvised collaborations of several different professionals.38 Each professional comes to the interprofessional collaboration through his or her unique activity system and contributes a distinctive thread of activity to resolve the patient's evolving clinical situation. If we revisit the example of the third-year general internal medicine resident, this resident may be a member of an interprofessional team with a supervising staff physician, a nurse, a patient, the patient's family member, a social worker, and a pharmacist. Each of these members brings a thread of activity, uniquely informed by his or her particular activity system, to provide patient care (see Figure 2). This collaborative performance, when the threads tie together, is a Knotworking knot.
Over time, different members of this team will withdraw from the collaborative care activity, and others will join. For example, the resident may be new to the team, having just rotated in when the previous resident left nephrology to go to rheumatology. While the resident is providing care during the day, when he or she must leave for education, the supervising staff physician will be on call. Other professions also experience team member rotation. For instance, at shift change, the nurse who was actively participating in a patient's care withdraws while another nurse enters into the activity. When collaborators leave or enter the care team, they withdraw or contribute a thread of activity into the collaborative knot. The knot of care activity is continually changing, involving the threads of different interprofessional team members at different times.
There is a notable difference between Knotworking's knots of interprofessional teams and other definitions of teamwork: in each knot there is no center that fixes the collaborative efforts of team members.38 The Knotworking knot cannot be reduced to a particular individual or organizational center of control. Instead, the locus of initiative changes from moment to moment. In each knot, as the participating care professionals work together to address a patient's care, different collaborators move the knot of activity forward at different times. For instance, the resident is moving care forward when entering a pharmaceutical order for the patient. However, the resident may not be familiar with the electronic order-entry system used on the ward and may be hurrying to finish his or her order entry to leave for an important educational session. In this situation, unwilling to ask for too much assistance in his or her first weeks on the ward, the resident inputs the complicated order using a free-text entry space found in the system. The resident, concerned that the entry will be confusing, orally clarifies the order to the nurse on the care team. The nurse then becomes the initiator of further action because he or she will use the resident's oral information to correctly administer the medication, despite the way it may appear in the order-entry system. As the patient's case continues to evolve, different collaborators instigate further action and further evolution of the knot of care activity. As this example illustrates, to conceive of health care team work as knots of activity is to make the unstable knot itself the focus of analysis.
There are several key issues related to these knots of interprofessional activity that can support a more complex understanding of interprofessional communication and its relationship to medical error. However, in the scope of this paper, only one of these issues can be fully developed: communication tools are crucial to the success of knot-based team work. Because the combination of activity systems involved in care activities is always unstable, particular analytical attention must be paid to the communication tools that coordinate actions. In our example, the resident entered confusing medication data in the electronic order-entry system. Clarification was provided orally to the nurse participating in the care team. In this example, both oral- and electronic-based communication tools coordinated the team's actions.
However, this example illustrates how local communication solutions do not necessarily eliminate latent conditions that enable latent medical errors. The confusing order continues to exist in the electronic system. The content of the oral clarification may be recorded in the patient's paper chart, but it may not be repeated in other team communications. If the staff physician is called to look in on the patient while the resident is off ward, and if the nurse who received the oral clarification is not at the bedside, the confusing medication order may be the only communication available to inform the staff of the patient's current care plan. In this scenario, the confusing order lies dormant within the system as a latent error. When the staff physician reads and acts on the information within that order, the staff physician's actions become the local triggers that activate the latent error.
As this scenario illustrates, individual communication tools (such as the electronic order-entry system and oral conversations) can be part of the activity systems of several different professionals. It is through shared communication tools that interprofessional teams coordinate care activities. When patient information is shared across multiple communication tools by rotating members of interprofessional teams, confusion or errors generated in communication tools can easily become sources of latent medical errors. To avoid the generation of such errors requires that health care professionals look beyond the immediate team members involved in care activities. Instead, professionals must take more complexity into account. They need to consider the distributed knots of activity involved in the patient's care and construct solutions that extend across multiple knots of team membership and activity.
With these two theories as lenses, we can analyze the connection between interprofessional communication and latent medical errors in new ways by exploring a more complex model of clinical activity. This theoretical framework provides a new approach for analyzing Reason's9 Swiss cheese model of how defenses, barriers, and safeguards can be penetrated by an error trajectory. Ideally, the layers of defenses, barriers, and safeguards in place in health care centers to prevent medical error would be intact and whole. However, in reality, they are more like slices of Swiss cheese, having many holes.9
Through the lens of AT, we can similarly see that each individual professional involved in a health care team is part of the defenses that guard against error. Each individual professional and his or her activity system can be seen as a layer of defense. In one's care activities, each individual is continually negotiating conflicting objectives and is juggling the use of many communication tools with which he or she has varying levels of competency. Also, each professional has multiple community affiliations, each imposing different rules and divisions of labor expectations. At any one time, one or more elements of an individual professional's activity system can create a “hole” or gap in the defenses against error. Because interprofessional care involves multiple team members, the defenses of others can diminish and even negate the effect of a hole. However, each member of the team will have gaps in his or her defenses at different times and in different aspects of his or her own activity systems. At times, these gaps will align, resulting in an error. Examining each interprofessional team member's communicative work in the context of his or her activity system enables us to better understand how and why latent errors are generated and how and why team defenses fail.
Through the lenses of AT and Knotworking, we can ask more complex questions about the different individual, social, physical, and systemic factors that contribute to interprofessional communication breakdowns: Did information get shared across the activity systems involved in the knots of activity? Did the professionals in one knot fail to pass sufficient information on to the participants of the next knot? Did one professional need to improvise the use of one communication tool, resulting in a confused message being relayed to other professionals in the team? Analysis based on activity systems and Knotworking can help us both to ask and to answer these questions by examining the broad-scope complexity of medical work. Research incorporating these approaches can develop a more complex understanding of how interprofessional communication can contribute to the existence of latent errors within health care systems. By explicitly investigating the broad-scope complexities of interprofessional team communication and health care work, we will more effectively prepare students and practicing clinicians to work safely in interprofessional teams and to face the errors that will inevitably arise.
Lorelei Lingard is supported by a CIHR Investigator Award and as the BMO Financial Group Professor in Health Professions Education Research.
The authors are grateful to Alan Bleakley for critical commentary that led to the improvement of this paper, and to Nancy Hébert and the MedTech Multimedia Division at the University of Ottawa for creating the informative figures presented in this paper.
1 Reeves S, Russell A, Zwarenstein M, et al. Structuring communication relationships for interprofessional teamwork (SCRIPT): A Canadian initiative aimed at improving patient-centred care. J Interprof Care. 2007;21:111–114.
2 Kohn KT, Corrigan JM, Donaldson MS. To Err Is Human: Building a Safer Health System. Washington, DC: National Academy Press; 1999.
3 Espin S, Lingard L, Baker GR, Regehr G. Persistence of unsafe practice in everyday work: An exploration of organizational and psychological factors constraining safety in the operating room. Qual Saf Health Care. 2006;15:165–170.
4 Thornby DCNAA. Beginning the journey to skilled communication. AACN Adv Crit Care. 2006;17:266–271.
5 Woloshynowych M, Davis R, Brown R, Vincent C. Communication patterns in a UK emergency department. Ann Emerg Med. 2007;50:407–413.
6 Leape LL, Berwick DM. Five years after To Err is Human: What have we learned? JAMA. 2005;293:2384–2390.
7 Alvarez G, Coiera E. Interdisciplinary communication: An uncharted source of medical error? J Crit Care. 2006;21:236–242.
8 Reason J. Human Error. New York, NY: Cambridge University Press; 1990.
9 Reason J. Human error: Models and management. BMJ. 2000;320:768–770.
10 Lingard L. The rhetorical ‘turn' in medical education: What have we learned and where are we going? Adv Health Sci Educ Theory Pract. 2007;12:121–133.
11 Kuziemsky CE, Borycki EM, Black F, et al. A framework for interdisciplinary team communication and its implications for information systems design. Eighth World Congress on the Management of eBusiness. IEEE 2007.
12 Lingard L, Reznick R, DeVito I, Espin S. Forming professional identities on the health care team: Discursive constructions of the ‘other' in the operating room. Med Educ. 2002;36:728–734.
13 Shaw A, deLusignan S, Rowlands G. Do primary care professionals work as a team: A qualitative study. J Interprof Care. 2005;19:396–405.
14 Cole E, Crichton N. The culture of a trauma team in relation to human factors. J Clin Nurs. 2006;15:1257–1266.
15 Lingard L, Schryer C, Garwood K, Spafford M. “Talking the talk”: School and workplace genre tension in clerkship case presentations. Med Educ. 2003;37:612–620.
16 Lingard L, Espin S, Evans C, Hawryluck L. The rules of the game: Interprofessional collaboration on the intensive care unit team. Crit Care. 2004;8:R403–R408.
17 Shendell-Falik N, Feinson M, Mohr BJ. Enhancing patient safety: Improving the patient handoff process through appreciative inquiry. J Nurs Adm. 2007;37:95–104.
18 Laxmisan A, Hakimzada F, Sayan OR, Green RA, Zhang J, Patel VL. The multitasking clinician: Decision-making and cognitive demand during and after team handoffs in emergency care. Int J Med Inform. 2007;76:801–811.
19 Cohen T, Blatter B, Almeida C, Patel VL. Reevaluating recovery: Perceived violations and preemptive interventions on emergency psychiatry rounds. J Am Med Inform Assoc. 2007;14:312–319.
20 Koppel R, Metlay JP, Cohen A, et al. Role of computerized physician order entry systems in facilitating medication errors. JAMA. 2005;293:1197–1203.
21 Spafford MM, Lingard L, Schryer CF, Hrynchak PK. Tensions in the field: Teaching standards of practice in optometry case presentations. Optom Vis Sci. 2004;81:800–806.
22 Varpio L, Schryer CF, Lehoux P, Lingard L. Working off the record: Physicians' and nurses' transformations of electronic patient record-based patient information. Acad Med. 2006;81(10 suppl):S35–S39.
23 Varpio L, Spafford MM, Schryer CF, Lingard L. Seeing and listening: A visual and social analysis of optometric record-keeping practices. J Bus Tech Commun. 2007;21:343–375.
24 Manojlovich M, Decicco B. Healthy work environments, nurse-physician communication, and patients' outcomes. Am J Crit Care. 2007;16:536–543.
25 Williams KA, Rose WD, Simon R. Teamwork in emergency medical services. Air Med J. 1999;18:149–153.
26 Plsek PE, Greenhalgh T. Complexity science: the challenge of complexity in health care. BMJ. 2001;323:625–628.
27 Scott C, Hofmeyer A. Acknowledging complexity: Critically analyzing context to understand interdisciplinary research. J Interprof Care. 2007;12:491–501.
28 Hall P. Interprofessional teamwork: Professional culture as barriers. J Interprof Care. 2005;19(suppl 1):188–196.
29 Vygotsky LS. Mind in Society: The Development of Higher Psychological Processes. Cambridge, Mass: Harvard University Press; 1978.
30 Leont'ev AN. Problems of the Development of the Mind. Moscow, USSR: Progress Publishers; 1981.
31 Engestrom Y. Learning by Expanding: An Activity-Theoretical Approach to Developmental Research. Helsinki, Finland: Orienta-Konsultit; 1987.
32 Engestrom Y, Blackler F. On the life of the object. Organization. 2005;12:307. Available at: (http://proquest.umi.com/pqdweb?did=829135771&Fmt=7&clientId=65345&RQT=309&VName=PQD
). Accessed June 13, 2008.
33 Engestrom Y. Expansive visibilization of work: An activity-theoretical perspective. Comput Support Coop Work. 1999;8:63–93.
34 Hasu M, Engestrom Y. Measurement in action: An activity-theoretical perspective on producer-user interaction. Int J Hum Comput Stud 2000;53:61–89.
35 Cole M, Engestrom Y. A cultural-historical approach to distributed cognition. In: Salomen G, ed. Distributed Cognitions: Psychological and Educational Considerations. Cambridge, UK: Cambridge University Press; 1993;1–46.
36 Engestrom Y, Miettinen R. Introduction. In: Miettinen R, Punamaki R-L, ed. Perspectives on Activity Theory. Cambridge UK: Cambridge University Press; 1999;1–16.
37 Schryer C, Lingard L, Spafford M, Garwood K. Structure and agency in medical case presentations. In: Bazerman C, Russell D, eds. Writing Selves/Writing Societies. Available at: (http://wac.colostate.edu/books/selves_societies
). Accessed June 13, 2008.
38 Engestrom Y, Engestrom R, Vahaaho T. When the center does not hold: The importance of knotworking. In: Chaiklin S, Hedegaard M, Jensen UJ, eds. Activity Theory and Social Practice. Aarhus, Denmark: Aarhus University Press; 1999:345–374.
39 Kangasoja J. Complex design problems: An impetus for learning and knotworking. In: Keeping Learning Complex: The Proceedings of the Fifth International Conference on the Learning Science (ICLS). Mahwah, NJ: Erlbaum; 2002:199–205. Available at: (http://www.edu.helsinki.fi/activity/publications/files/47/ICLS2002_Kangasoja.pdf
). Accessed June 13, 2008.
40 Engestrom Y. Activity theory as a framework for analyzing and redesigning work. Ergonomics. 2000;43:960–974.