Groups are pervasive in healthcare institutions and take on a variety of shapes. This paper uses a typology that allows us to understand the distinctive characteristics of team operations, based on interdependence and interactive dimensions. It looks at factors that influence team effectiveness in organizational settings. We review different frameworks that shed light in explaining the conditions that lead to group effectiveness. From the classical input-process-output (IPO) model to the input-mediator-output-input (IMOI) model of team effectiveness; the taxonomy of team process and emergent estates, as well as the teams understood as complex adaptive systems and also studied from the multiteam system perspective. We also report the need for more robust research designs to contribute to the field’s further advancement. There is consensus among scholars demanding further conceptual frameworks, as well as powerful research designs that capture process-oriented theory and research on team effectiveness. Some future directions and recommendations are suggested.
In recent decades, organizations have increasingly turned to using teams and made them a part of day-to-day routines [1, 2], and all for a variety of reasons, such as the ability to respond to emergencies, engage in continuous quality improvement efforts, and manage work projects through multidisciplinary teams. In the particular case of healthcare organizations, teamwork is essential to provide effective care, and the lack of teamwork has been identified in the literature as a key vulnerability in terms of service quality [3, 4]. In this chapter we propose revisiting the conditions that promote effective teamwork. We will first examine team work typology, using interaction and interdependence as the key dimensions characterizing and describing teams. We will then focus on teamwork effectiveness and review a few of the more influential frameworks that have driven research dedicated to teams. Finally, we will conclude with some directions for future teamwork research. But, first, we should briefly discuss what a team and teamwork are.
Kozlowski and Ilgen  provide a rather thorough definition of teams, describing them as “two or more individuals who socially interact (face-to-face or, increasingly, virtually); possess one or more common goals; are brought together to perform organizationally relevant tasks; exhibit interdependencies with respect to workflow, goals, and outcomes; have different roles and responsibilities; and are together embedded in an encompassing organizational system, with boundaries and linkages to the broader system context and task environment” (p. 79) . Although exhaustive, this approach defines teams in a somewhat mechanistic way in terms of their design, with an external focus. This view has been countered with a different perspective which sees teams as more dynamic and as self-constructed entities. This led Humprey and Amy  to define teams as “assemblies of interdependent relations and activities organizing shifting sets or subsets of participants embedded in and relevant to wider resource and institutional environments” (p. 450) .
On the other hand, teamwork is a process that emerges from the interactions established among team members  and it can be defined as “a set of interrelated thoughts, actions, and feelings of each team member that are needed to function as a team and that combine to facilitate coordinated, adaptive performance and task objectives resulting in value-added outcomes” (p. 562) . Teamwork reflects the minute-by-minute behaviours and interactions that take place between team members work when executing a task . As proposed by Salas et al. , teamwork is guided by a number of fundamental principles: it is characterized by a set of behaviours, cognitions and attitudes that should be flexible and adaptive; team members should monitor each other and feel safe to provide feedback and comfortable when receiving it; team members should also be willing and capable of providing support to other team members in their operations and activities; teamwork involves clear, precise, and concise communication; team members must be able to coordinate interdependently to take collective action; teamwork requires leadership that provides direction, planning, distribution, and activity coordination; and, finally, teamwork is subject to external influences as well as to the requirements of the task itself.
2. Typology of formal groups
As in all organizations, groups are pervasive in healthcare institutions and take on a variety of shapes, ranging from different units or working groups that are permanent in nature to “ad hoc” groups (committees, meetings, etc.) which are eminently temporary. In order to manage this variety of groups, establishing a typology will allow us to understand the distinctive characteristics of their operations. In addition to varying relative to the purposes they serve, formal groups (permanent or temporary) also diverge according to the basic characteristics of how they operate. The way they function is determined by two basic dimensions: team interaction and interdependence. Team interaction relates to how team members “behav[e] together, in some recognized relation to one another” (p. 12) , for the purpose of performing a task. Team interdependence is the extent to which team members cooperate, depend on each other, and work interactively to complete team tasks . Although related, the two concepts are independent in the sense that, although teams with high degrees of interdependence also have high degrees of interaction, the same does not always happen in the opposite sense. That is, teams with a high degree of interaction do not necessarily have a high degree of interdependence, since team members may interact but not depend on each other.
2.1 Team interaction
Team interaction is central to teamwork and represents complex, temporal phenomena with multilevel manifestations . It is complex because it involves a web of behavioural connections between team members; it is temporal because the very execution of team tasks has a temporal dimension unfolding over time at a specific rhythm and pace; and it manifests at several levels because it is nested in individual and collective behaviours. Team interaction is thus subject to influences from elements related to individuals, from elements within the team itself, and from relational factors. Individual factors can include, for example, team members’ attitudes towards work and the team. Collaborative attitudes will promote better interactions than competitive ones. Regarding team factors, for example, Lehmann-Willenbrock and Allen  observed that humour considered at the team level has a positive influence on the incidence of interactions within the team. From a relational point of view, differences in status and power within the team also influence the level of interaction, with that interaction increasing the smaller the differences in status and power. The team’s interaction level also has significant and positive outcomes for teams. One such consequence is the development of similar team mental models, which can be defined as a common understanding among team members about key elements in the relevant team environment . The similarity of team mental models has positive effects on several dimensions such as team performance  and adaptive capacity .
2.2 Team interdependence
Although team interdependence can be considered a single general factor, it can also be seen in three distinct dimensions: task, goal, and outcome interdependence .
In particular, task interdependence has been widely studied [19, 20] for its implications on the way teams operate and perform. For example, to determine how to assign outcomes to individual group members, the types of tasks the team performs have to be taken into account. Thompson’s  group task model (Figure 1) can help to assess the extent to which the work performed by one member affects what other group members do, as well as identifying the most effective way to distribute outcomes and/or rewards. In essence, this model reveals the form that task interdependence can take.
Group tasks based on
In tasks with
With increasing interdependence –pooled interdependence, sequential, reciprocal, and intensive–, the potential for conflict and dysfunctional behaviours can increase . However, research provides strong evidence that the relationship between team efficacy (team perceptions regarding its ability to perform a specific task) and performance is stronger when that interdependence is high compared to when it is low .
2.3 Types of groups
Based on the two team interaction and interdependence dimensions, we can distinguish four types of organisational groups (Figure 2): Staff/Crew, Remote-controlled group, Coordinated group, and Team. In the
2.4 Nature of team tasks
There are numerous dimensions by which tasks can be classified. Above we saw a classification based on interdependence, but we can look at tasks from another perspective, for example, according to the team members’ contributions. From this standpoint, tasks can be additive, conjunctive or disjunctive . A task is
In this section we have looked at some typologies of formal groups and discussed the interdependence of the teams’ tasks and their members’ interaction. In the next section we will review some of the most influential frameworks driving research on work teams.
3. Approaches to team dynamics
The last three decades have seen a significant increase in the number of articles published on teams or groups. A literature review of articles published in the
3.1 Fundamental frameworks
Scholars have developed different frameworks to attempt to explain the conditions that lead to group effectiveness. The classic input-process-output (IPO) model of team effectiveness [27, 28] guided developments in team research for several decades. Within the IPO model, the inputs are the antecedents, that is, the conditions that exist prior to the group activity (e.g., organizational context, task characteristics, and team composition). The processes are the interactions among group members that mediate the relationship between the team’s inputs and outputs (e.g., communication and coordination processes). Lastly, the outputs are the results, the consequences of group activity (e.g., productivity/performance, member satisfaction, and innovation). For example, the early IPO model proposed by McGrath  suggests that individual, group, and environmental-level factors are antecedents to group interaction processes with effects on performance outcomes such as quality, speed, number of errors, and other types of outcomes, such as member satisfaction or group cohesion.
The IPO model has been highly influential in research on teams and how members can combine their efforts and knowledge to complete a specific task. However, more recently, the model has been questioned as it has some limitations when considering the dynamic nature of teams [29, 30]. One criticism raised is that, despite involving team interactions, many researchers studying processes only assess these as static retrospective perceptions, ignoring how they emerge, their dynamics and evolution over time . Furthermore, the IPO model does not take into account that all mediational factors are not necessarily processes but can also be emergent states  as we explore below. In addition, teamwork influences create a feedback loop in which reversal causal sequences are also possible, given that the results of a team’s actions can also be an input for the following action, something not reflected in IPO models [31, 32]. To avoid some of these limitations, Ilgen, et al.  proposed the input-mediator-output-input (IMOI) model. In the latter, inputs are added at the end of the model to denote the system’s cyclical nature, and processes are replaced by mediators to reflect a wider range of variables, namely processes and emergent states.
3.2 Team processes and emergent states
As seen above, not all team mediation mechanisms are processes; some are emergent states  . The difference between the two is fundamental, since processes imply interactions while emergent states do not.
This sequential notion in which a process or emergent state is both an output and an input of subsequent processes and emergent states leads us to the recurring phase model of team processes proposed by Marks et al. . In their model, team performance episodes unfold over time, signalling specific periods in which action and transition phases occur.
Marks et al.  developed a taxonomy of team processes that considers practices that typically occur in transition phases, those that occur in action phases, and interpersonal processes that occur in both. In transition phases, team members conduct three types of processes: mission analysis, goal specification, and strategy formulation.
Recently, Mathieu, Luciano et al.  have developed a team process survey tool that allows researchers to examine team processes more systematically (transition, action, and interpersonal processes). In its more extensive version, this tool includes 50 items, while its intermediate version has 30 and the reduced version only 10, one for each process. As recommended by authors , the use of the reduced 10-item version may be tempting, but it is not the most appropriate in all situations. The longer versions offer a more complete representation of the various dimensions. For example, Marks et al.’s taxonomy  includes several sub-processes that are not revealed in the 10-item version. When the aim is to get an in-depth view of the team’s processes, the 30- and 50-item versions are more advisable. When only a quick look at how the team currently functions is desired or when this measure is included in a more extensive questionnaire along with other scales, using the 10-item version may be advantageous.
With regard to emergent states, an article by Grossman, Friedman and Kalra  summarises the emergent states emphasized the most in the literature, dividing them into affective and cognitive mechanisms. In
As far as cognitive mechanisms are concerned,
3.3 Teams as complex adaptive systems (CAS)
Since Arrow, McGrath and Berdahl  characterised teams as complex adaptive systems (CAS), multiple theoretical frameworks have emerged to capture and explain this idea. However, relatively few empirical studies have been able to examine how long it takes teams to become effective and how these effects develop over time [49, 50, 51]. CAS are open systems that are characterised by the level of uncertainty regarding their evolution over time given the interaction of their components . Ramos-Villagrasa et al.  carried out a systematic review through the nonlinear dynamical system theory lens, supporting the view of teams as complex adaptive systems. Teams are complex because they are integrated within organisations that exhibit complex behaviour; they are adaptive because they dynamically cope with environmental changes; and they are systems because their functioning depends on the team’s history and, therefore, on inputs, but also on the anticipated future, that is, on outputs. The continuous adaptive process that occurs within these teams allows them to adapt to contextual discontinuities and to make decisions according to both the team’s antecedents and projected results . The use of this new conceptual approach can help researchers to study teams in a non-linear and more dynamic way , as well as to address temporal problems [53, 54] by taking measures at different stages of the team’s evolution.
In the case of healthcare teams, they cannot always function as CAS . For example, in clinical situations where problems are identified and described in detail and solutions standardised in specific procedures, teams operate in a planned way, and guidelines are clear and executed in a simple way. However, when there is uncertainty about how to best handle a given situation, operating as a CAS may be the most appropriate option as it promotes the development of new ideas and approaches. This is based on 7 principles: (1) team members can operate autonomously guided by ground rules; (2) team members interact in non-linear ways, i.e., they are interdependent and affect other team members in different ways; (3) the team is sensitive to initial conditions; (4) interactions between team members can produce unpredictable behaviours; (5) these interactions can generate new behaviours; (6) the team is an open system interacting with the environment; and (7) team members function as attractors modelling team behaviour .
3.4 Multiteam systems
In the same complex adaptive system stream, teams can be studied from the multiteam system (MTS) perspective . An MTS corresponds to “two or more teams that interface directly and interdependently in response to environmental contingencies toward the accomplishment of collective goals” (p. 289) . These systems constitute “networks of interdependent teams that coordinate at some level to achieve proximal and distal goals” (p. 479) . In a system of this nature, the processes established between the various teams, the cross-team processes, are even more important for the system’s success than within-team processes . In the case of the healthcare industry, the use of a multiteam system logic is very beneficial, but much remains to be studied. For example, one area where team research is needed is how best to form networks that integrate patients and their families over time . Patients and their support structures are responsible for coordinating care tasks and helping interpret the information collected, extending beyond the boundaries of healthcare providers. Consequently, managing this extended multi-team system holistically will certainly have very positive results on patient care.
A literature review conducted by Shuffler and Carter  identified 7 important lessons for successful teamwork in an MTS: (1) MTS functioning is suited to contexts that are ambiguous, multifaceted, dynamic, and where there is a need for a sense of urgency; (2) MTS structures provide the specialisation, flexibility, and integration needed to deal with complex problems; (3) the teamwork phenomenon changes when moving from a teamwork logic within a team to a teamwork logic within an MTS, for example, cross-team processes take on sovereign relevance; (4) an MTS implies added barriers to collaboration that should be specifically addressed; (5) the incorporation of linking elements can benefit the system’s performance; (6) the structure of the MTS and the design of its functioning should be carefully thought out; and (7) leadership plays a crucial role in an MTS and should be integrated and managed across the system .
3.5 Facets of team effectiveness
Another relevant framework used to study team effectiveness was suggested by Mathieu et al.  illustrating the simultaneous and interrelated relationships among factors associated with team and individual outcomes. Based on a revision of team research published in the
Many of these constructs have been studied among healthcare teams. For example, O’Donovan et al.  recently developed a psychological safety measurement instrument designed specifically for healthcare teams. In this instrument, the authors combine the strengths of observation measures with survey measures, allowing for their application to longitudinal studies. Another tool has also been developed to measure the collective intelligence of primary healthcare teams . Collective intelligence can prevent repeating past mistakes and help teams to be more efficient. Jean et al.  argue that intelligent teams produce high quality clinical services, so it is essential to better understand the concept and be able to measure it accurately.
Johnson  found that intra-team communication demonstrates recurring problems that make it difficult for healthcare teams to coordinate, proposing that teams should work within a common framework represented by formal, informal, market, and professional relationships, or a unique mix based on a mutual orientation towards patient outcomes. The formal approach is based on explicit knowledge and a shared system of codes that, for example, can be translated into written guidelines for hospitals. In addition, the formal approach considers that: personal relationships are also a source of informal information that can overcome the barriers created by formal panels; market logic relates to the creation of information and knowledge-exchange relationships that tend to be maintained through the investment that has been put into the relationship; and professional relationships relate to communication within the domain of professions by creating networks of contacts between professionals based on mutual help. Information-sharing and supportive behaviours have also been observed to have a positive impact on innovation in healthcare teams .
A study conducted by Jaca, et al.  revealed that the role of the external leader in healthcare teams is quite relevant, and his/her main function is to serve as a team performance coordinator. There is also a clear definition of roles, which facilitates decision-making and conflict management. Furthermore, internal communication and participation levels tend to be high. However, team recognition and training need to be improved, as these are the weakest points in healthcare teams. Several studies have also drawn attention to the importance of teamwork in healthcare and, in particular, the importance of interventions to promote teamwork [3, 66]. One of these types of interventions is “TeamSTEPPS” (Team Strategies and Tools to Enhance Performance and Patient Safety), developed by the Agency for Healthcare Research and Quality (AHRQ) in the USA. TeamSTEPPS is based on communication, leadership, mutual support, and situation monitoring. Another useful model is CRM (Crew Resource Management), which has a significant impact on knowledge and behaviour in acute care settings, such as healthcare .
4. Future research avenues
Despite the remarkable advance in team work research, scholars agree on the need for more robust research designs to contribute to the field’s further advancement. In addition to the meta-analysis contributions summarizing past empirical findings [17, 18, 67, 68, 69], there is consensus among scholars demanding further conceptual frameworks, as well as powerful research designs that capture process-oriented theory and research on team effectiveness [29, 70].
Humphrey and Aime  call for a multilevel, multi-theoretical, and multiperiod framework to cope with the contextual dynamics and enhance the understanding of team dynamics. Likewise, Mathieu et al.  state that future advances on workgroup effectiveness will be linked with the ability to capture dynamic team properties (conceptually and methodologically); the complexity of team task environments; and the embeddedness in multilevel environments. In the special issue dedicated to
As seen, team scholars agree regarding the need for innovative research designs and new techniques to capture team dynamics over time. In this sense, Delice et al.  summarize and review existing empirical studies that use novel measurements to study team dynamics over extended periods. Some of these innovative research designs are based on techniques such as role-playing simulations, videotape and software coding, videogames, video-coding, team decision tasks, and whatsApp ICT (information and communication technology). Delice et al.  also propose longitudinal laboratory experiments and time-series analyses. Other alternatives include scenario-based studies, critical incident techniques, concept-mapping, cross-border e-business website analyses, and simulations (simulation tasks and longitudinal organizational, computer game-based, and dynamic decision-making simulations), as well as experiential learning approaches and performance assessments, among others. There is, therefore, a plethora of alternatives that should be used to further our understanding of teams that are dynamic and part of adaptive systems .
5. Concluding thoughts
In summary, some of the key ideas for future research attempt to overcome the limitations of traditional self-reported assessments, which suffer from problems such as low response rates, response bias, or intrusiveness [29, 74, 75]. Some research strategies that can help to overcome these effects:
Using more than one measurement method, potentially avoiding single-source bias as well as survey respondent fatigue .
Conceptualizing multiple levels, process dynamics, and the emergence of team phenomena over time .
Addressing and reporting on the different types of work interdependence (task or outcome interdependence) .
Thinking about new ways of obtaining team data such as emails, smartphones, video surveillance, etc., to replace multiple data collection points and traditional self-reported surveys .
At the more conceptual level, possible strategies include: