Disrupted Routines: Effects of Team Learning on New Technology Adaptation

This paper reports on a multimethod study of 16 hospitals implementing an innovative technology for cardiac surgery. This inductive study led to propositions about new technology adaptation and how group and organizational characteristics influence this process and its outcomes. The findings reveal substantial differences in how an identical technology was integrated into ongoing practice in a set of structurally homogeneous organizations. A set of factors at the group level of analysis differentiated successful and unsuccessful adopters: characteristics of operating room teams charged with implementing the new technology—team leader behavior, team psychological safety, team learning behavior, and boundary spanning— were associated with successful adoption. Team membership stability was associated with the rate of efficiency improvement using the new technology. Organizational factors, such as size, type (academic versus community hospital) and organizational support for and experience with innovation, were not associated with either outcome. We develop new theoretical propositions, and suggest that team learning may moderate the strong relationship between structure and technology adaptation. DISRUPTED ROUTINES 2 Adopting new technologies is essential to sustained competitiveness for many organizations. In both manufacturing and service industries, new technology can lead to product and process improvements that produce tangible market advantages—but these advantages can be elusive. Organizational research suggests that failure to adopt innovations, even those with demonstrable benefits, is commonplace (e.g., Henderson and Clark, 1990; Tushman and Anderson, 1986). Organizations have been depicted as blind to the existence or advantage of external innovations (March and Simon, 1958), trapped by current competencies (Levitt and March, 1988) and business models (Christensen, 1997), paralyzed by core rigidities (Leonard-Barton, 1992) and handicapped by a lack of relevant expertise (Cohen and Levinthal, 1990)—all leading to a failure to adapt in the face of environmental changes. Understanding when and why innovations are adopted has been the subject of considerable research. Much of the focus in this literature has been on the decision to adopt, investigating the timing of adoption decisions, product features that promote adoption (e.g., Rogers, 1983) and organizational characteristics that facilitate recognition of the significance of an innovation (Iansiti and Clark, 1994; Cohen and Levinthal, 1990). Researchers have suggested that an organization’s history of innovation and the sophistication of its own research activities can allow it to adopt future innovations more easily, through building absorptive capacity (Cohen and Levinthal, 1990). An implication of much of this research is that the management challenge lies in ensuring awareness of innovations and commitment of resources to them. With some exceptions (e.g., Leonard-Barton, 1988), this literature has paid less attention to understanding the conditions under which organizations successfully integrate new technologies into their operations. The decision to adopt may be only a first—and often far from sufficient—step in integrating a new technology into the ongoing work of the organization and obtaining benefits from its routine use. Often, both the technology and the organization must adapt for adoption to succeed (Leonard-Barton, 1988). When new technologies disrupt well-established organizational routines, they are likely to be particularly difficult to implement. Another stream of research focusing on the ways in which behavioral routines and DISRUPTED ROUTINES 3 roles can be threatened and disrupted by the introduction of new technologies has used ethnographic methods to develop rich portraits of new technology implementation (e.g., Barley, 1986; Orlikowski and Hofman, 1997). From detailed investigations of work activities, these studies show that organizational adaptation to new technologies does not always proceed as planned and, instead, is fragmented and episodic (Orlikowski and Hofman, 1997; Tyre and Orlikowski, 1994). Also, organizational structures, such as patterns of interaction between different functions or role occupants, can be altered by new technology, sometimes in different ways in similar organizations (Barley, 1986). Structure consists of power dependencies and contextual constraints (Weick, 1993; Ranson, Hinings, and Greenwood, 1980); technology describes equipment that shapes and constrains tasks (Galbraith, 1974). Current structuration theory maintains that structure and technology exert a reciprocal influence, such that adaptation does occur, albeit slowly, constrained by existing status relationships (Barley, 1986). Further, organization members' beliefs about the meaning of a new technology may affect its acceptance in the organization; for example, information technology can be seen as automating jobs or as informing intelligent users to do their jobs, leading to differences in user acceptance (Zuboff, 1988). For these reasons, after the decision to adopt a new technology has been made, implementation still may be difficult and unpredictable. In general, this stream of research focuses on the organizational and human experience during new technology adaptation rather than on explaining differences across organizations implementing the same technology. Theories of organizational learning suggest that some organizations are more able to learn than others (Senge, 1990; Garvin, 2000), which may include making changes required for successful adoption of new technologies that threaten organizational routines. Despite compelling theory, empirical studies investigating differences in organizational learning processes and outcomes are rare. Many studies are limited to single organizations (e.g., Kim, 1993; Roth and Kleiner, 2000; Argyris, 1993). Organizational learning research has not collected data from a number of organizations simultaneously facing a similar opportunity for adaptation to investigate factors promoting an effective transition process. DISRUPTED ROUTINES 4 In this paper, we build on the organizational learning literature by investigating how cardiac surgery departments learn to use to a specific new technology. We frame adaptation to new technology as a collective learning process. Adoption starts with the decision to try a new technology, but successful adoption often requires adaptation, or organizational changes, to enable its use. Because little is known about factors affecting this learning process, our aim is to develop new theory and suggest propositions for future research, rather than to test hypotheses. We studied 16 hospitals and used multiple methods to investigate both groupand organization-level factors that might affect outcomes. Not all hospitals conduct cardiac surgery, but those that do share a powerful institutional context. INSTITUTIONAL STRUCTURE AND TECHNOLOGY IN CARDIAC SURGERY Across hospitals of varying size, location, history, and academic status, the structure and process of cardiac surgery—especially as manifested in roles and relationships in the operating room team—are remarkably consistent. Performing a coronary artery bypass graft (CABG) or valve replacement surgery includes many small adjustments and minor differences across procedures due to patient variation and surgeons' preferences, but overall these procedures are highly routine, involving the repetition of a precise set of moves in operation after operation. The role of each operating room (OR) team member is well understood and prescribed. Anesthesiologists put the patient to sleep and monitor vital functions during surgery; surgeons perform the actual surgical procedures (cutting and stitching), monitor the patient’s well-being and oversee the team; perfusionists are technicians who run the heart-lung bypass machine, and scrub and circulating nurses prepare equipment and support the surgeon in a variety of ways. These team routines transcend institutions. Although medical practices may vary across hospitals, professional training in surgery follows widely accepted protocols and utilizes standard technology—both derived from the research literature with which physicians and medical centers are expected to remain current. This promotes homogeneity across hospitals. Moreover, cardiac surgery places even more value on standardization than other surgical specialties; the only variation typically found involves specific DISRUPTED ROUTINES 5 surgical techniques and details of the layout of the operating room (O'Connor et al., 1996). Acting within prescribed roles, team members are able to act in perfect concert without discussion; conversation that does occur is typically about an unrelated subject, such as last night's baseball game. As an informant in our study explained, "In [CABG surgery], you look at the surgeon and you know the body language, and you act." The operating room team in a typical cardiac surgery department is likely to perform one or two, and sometimes three, open-heart operations each day and, therefore, hundreds each year. All members of the surgical department are assumed to be equally capable of doing the work of their particular discipline, and team members within a discipline are readily substituted for each other. This consistency of practice made it an ideal context in which to seek differences in patterns of new technology adaptation. The Cardiac Surgery Task Conventional cardiac surgical procedures are divided into three phases. First, the surgeon carries out a median sternotomy (cutting open the chest and splitting apart the breastbone) and stops the heart. The surgeon then directs the nurse and perfusionist to connect the patient to a heart-lung bypass machine to regulate oxygenation and support blood pressure while the heart is stopped. In the second phase, the surgeon repairs and replaces diseased components ("stitching"), and in the third phase, the surgeon restarts the heart and the patient is weaned from the bypass machine. The division of task responsibilities in the OR team mirrors powerful hierarchical status relationships in cardiac surgery. This hierarchy is further reinforced by the formal and informal status relationships between doctors and nurses in medicine overall, as well as by the formal medical and legal accountability of doctors. A New Technology: Minimally Invasive Cardiac Surgery Minimally invasive cardiac surgery (MICS), an innovation developed and manufactured by a device company called Minimally Invasive Surgical Associates (MISA), differs from the conventional approach in that the breastbone is not split apart. This reduces the extent of pain and recovery time for patients, 1 All names are pseudonyms. DISRUPTED ROUTINES 6 such that they are able to resume normal activities more quickly than after conventional cardiac surgery. Using special new equipment, the heart is accessed through small incisions between the ribs, and the patient is connected to the bypass machine through the artery and vein in the groin. A deflated "balloon"—threaded into the aorta and then inflated to prevent blood from flowing backwards into the stopped heart—replaces the traditional clamp inserted directly into the chest. This placement requires coordination among all team members; the balloon’s path must be carefully monitored with ultrasound technology (trans-esophogial echo, or "echo") because there is no direct visual and tactile data to help guide the process. The tolerances on balloon location are excruciatingly low, and correct placement is critical. Unlike conventional surgery, in which surgeons receive information from direct sensation, MICS calls for team members to supply the surgeon with vital information displayed on digital and visual monitors. Once the balloon clamp is in place, team members must continue to monitor to make sure it stays in place. The improvements for patients promised by the technology come at a high learning cost for surgeons and OR teams. As one surgeon we interviewed joked, "[MICS] represents a transfer of pain— from the patient to the surgeon." The new technology not only changes individual team members’ tasks, it dramatically increases team interdependence. The anesthesiologist and perfusionist now must work closely together to regulate blood pressure, the surgeon and the anesthesiologist must coordinate to monitor the position of the balloon clamp, and all team members must coordinate to monitor the patient’s vital pressures. The new technology thus disrupts a well established OR team routine and requires new communication patterns. Successfully enacting this change affects deeply engrained status relationships in the OR team, as the surgeon’s role shifts from that of an order giver to a team member in an interdependent process. As a nurse we interviewed explained, When you’re on bypass for the standard CABG, there’s no need for communication at all. In MICS there’s a lot more. The pressures have to be monitored on the balloon constantly. For putting in the balloon and the primary line, the communication with perfusion is critical. It is totally different. When I read the training manual, I couldn’t believe it. It was so different from standard cases. 2 Trans-esophogial echo (TEE) is a kind of ultrasound technology primarily used in cardiology. DISRUPTED ROUTINES 7 In a similar context—radiology departments introducing CT scanners—Barley (1986) studied two hospitals and found that, in one, physicians and technicians interacting in the diagnostic process renegotiated knowledge, creating greater interdependence than before the new technology was introduced. Subtle communication behaviors such as "unsought validation" by technicians of a physician’s nascent knowledge of the CT scanners and "clandestine teaching" of physicians by technicians facilitated successful use of the technology at “Suburban Hospital” without overtly violating status differences. At “Urban Hospital,” in contrast, communication remained essentially consistent with traditional patterns with physicians acting as order givers. Technicians' subordinate status thus was maintained at Urban; interdependence and reciprocal negotiation of knowledge did not increase, resulting in frustration and slow progress. Even when Urban physicians later began to ask for help from technicians, their requests were for technical knowledge only (related to equipment function), not for clinical knowledge, in contrast to what happened at Suburban. Barley attributed this difference to several factors, especially the presence of relatively less experienced radiologists and more experienced technicians at Suburban. The present study builds on this earlier research by investigating multiple factors that might influence these kinds of differences in responding to a new technology. Because the institutional context and the technology in conventional cardiac surgery are mutually supportive and reinforcing, this context—especially its status structures—presented a powerful barrier to the introduction of a new technology that required more interdependence. As Barley found, it is possible that this barrier is more difficult to overcome in some hospitals than in others, due to organizational or team characteristics. Organizational factors such as past experience with innovation in an organization (Cohen and Levinthal, 1990) and seniority of those introducing a new technology (Barley, 1986) can influence adoption outcomes. Although team factors have not been investigated in this context, research has shown that teams learning a new task are more likely to improve coordination and efficiency when they train as teams and stay together for subsequent task execution than when they change their DISRUPTED ROUTINES 8 membership (e.g., Moreland et al., 1998; Foushee et al, 1986). Thus, drawing from theories of team effectiveness (Hackman, 1987), we speculate that OR teams with stable membership may achieve efficiency in learning new techniques more quickly than unstable teams and that this may lead to adoption success. Further, team psychological safety—a shared belief that the team is safe for interpersonal risks (Edmondson, 1999)—may affect a team's ability to engage in the new behaviors required for MICS, overcoming institutional barriers described above. Thus, we consider both group and organizational factors. The research question guiding the study was what differentiates organizations that learn — successfully incorporating the new technology into ongoing routines—from those that fail to learn.