Groupthink among health professional teams in patient care: a scoping review

Abstract

There is emerging interest in understanding group decision making among a team of health professionals. Groupthink, a term coined by Irving Janis to depict premature consensus seeking in highly cohesive groups, is a theory that has been widely discussed in disciplines outside health care. However, it remains unclear how it has been conceptualized, studied, and mitigated in the context of health professionals conducting patient care. This scoping review aimed to examine the conceptualization of groupthink in health care, empirical research conducted in healthcare teams, and recommendations to avoid groupthink. Eight databases were systematically searched for articles focusing on groupthink among health professional teams using a scoping review methodology. A total of 22 articles were included—most were commentaries or narrative reviews with only four empirical research studies. This review found that focus on groupthink and group decision making in medicine is relatively new and growing in interest. Few empirical studies on groupthink in health professional teams have been performed and there is conceptual disagreement on how to interpret groupthink in the context of clinical practice. Future research should develop a theoretical framework that applies groupthink theory to clinical decision making and medical education, validate the groupthink framework in clinical settings, develop measures of groupthink, evaluate interventions that mitigate groupthink in clinical practice, and examine how groupthink may be situated amidst other emerging social cognitive theories of collaborative clinical decision making.

Introduction

Clinical reasoning can be described as the cognitive processes that a health professional engages in order to make clinical decisions. Our understanding of the cognitive processes that underlie diagnostic and treatment decisions has traditionally focused on individual decision making processes and potentially resulting cognitive biases (Bowen, 2006; Croskerry, 2003; Norman & Eva, 2010). However, there is emerging interest in understanding group decision making among a team of health professionals, and how to improve teamwork in clinical reasoning (Balogh, Miller, & Ball, 2015). Health care is increasingly becoming more collaborative, yet we still do not have a good empirical understanding of what underpins errors in decision making in a group setting. Understanding the mechanisms of errors in group decision making is critical to improve the quality and safety of patient care, and advance the training of future health professionals.

A systematic bias is a pattern of deviation from rational judgment and decision making that may affect individuals or teams (Jones & Roelofsma, 2000). Current understanding of decision making performed by health professional teams, referred to as team-based clinical reasoning, has been viewed under the lens of social cognitive theories such as situated cognition and distributed cognition (Olson et al., 2020). These theories suggest that clinical reasoning and decision making emerges from the interactions between people and their environment. For example, conversations between health professionals in a safe, non-hierarchical environment can lead to the co-construction of knowledge, insightful shared mental models, and better collaborative decision making. Understanding systematic biases in team-based clinical reasoning is a relatively new area of investigation.

One example of a potential systematic bias in group decision making is groupthink. Groupthink is a theory that describes when highly cohesive groups exhibit premature consensus seeking (i.e., premature closure on the group level) that leads to poor decision making (Janis, 1982; McCauley, 1998). Groupthink could occur at all levels of the hierarchy in health organizations, from frontline clinical teams to senior managers and leaders of the organization (Mannion & Thompson, 2014). For example, if a medical team member observes that the working diagnosis does not explain all of the patient’s symptoms, but does not mention this concern to the medical team due to the assumption that the group’s thought process and diagnostic decision must be correct, this group would be exhibiting groupthink.

Groupthink has been widely discussed in a variety of disciplines outside health care, including psychology, business and organization, political science, communication, and others (Burnette et al., 2011; Esser, 1998; Hällgren, 2010; Mpeera Ntayi et al., 2010; Pautz & Forrer, 2013; Ricciuti, 2014; Turner & Pratkanis, 2014). In contrast, prior literature and research of groupthink in health care is limited. To explore the breadth of the literature on groupthink in health care, we performed a scoping review to examine 1) the conceptualization of groupthink in health care, 2) empirical research conducted on groupthink in health care, and 3) recommendations to avoid groupthink in health care.

Janis’s Groupthink Theoretical Framework

In the early 1970’s, psychologist Irving Janis first introduced groupthink as a group decision making theory that impedes effective decision making. In his second edition book, Janis provided a theoretical framework for groupthink with causes (i.e., antecedent conditions), symptoms of groupthink, and consequences (Janis, 1982). Antecedent conditions in Janis’s framework influence the probability of groupthink and consequently the faulty decisions arising from groupthink. Janis determined that group cohesiveness was the main determining antecedent of groupthink, that group members may avoid speaking out against decisions in order to maintain group harmony.

Janis identified eight symptoms of groupthink: (1) an illusion of invulnerability; (2) moral superiority; (3) collective rationalization; (4) stereotyped views of others; (5) self-censorship of deviations from the group consensus; (6) a shared illusion of unanimity; (7) direct pressure on any member who dissents; and (8) the emergence of self-appointed mindguards who protect the group from information that might disrupt the group’s complacency about their decisions. Janis also identified consequences of groupthink that resulted from a faulty decision-making process. Finally, he provided recommendations to prevent groupthink in decision-making groups (see Table 3).

Table 3 -.

Recommendations to mitigate groupthink

Janis’s recommendations	Adaptation for clinical practice	Authors
Each member of the policy-making group should take on the role of ‘critical evaluator’. The group leader should organize this and augment this practice by welcoming criticism of their own opinions so members do not feel obligated to agree (Janis, 1982, pp. 262–263).	Individuals should engage or be appointed as ‘critical evaluator’ to encourage candidness.	Christensen (2019), Cleary et al. (2019), Gambrill (2005), Mannion & Thompson (2014)
Members of leadership should take the lead in creating a climate of impartiality by withholding assertion of their own preferences. By limiting statements of their opinions, it invites group members to participate in a culture of inquiry and into considering a wider range of decisions (pp. 263–264).	Members of leadership are responsible for promoting inquiry, constructive criticism and debate without fear of penalty and should reinforce neutrality by hiding bias.	Christensen (2019), Gambrill (2005), Hollinger (2019), Heinemann et al. (1994), Mannion & Thompson (2014)
Organizations should be in the habit of setting up several independent policy-planning and evaluation groups with different leaders. These separate groups all work on the same question or issue (pp. 264–265).	Organizational restructuring such as breaking down teams into smaller size group. Additionally rotating group members in their roles can help minimize groupthink.	Christensen (2019), Cleary et al. (2019), Mannion & Thompson (2014)
Groups should sporadically divide into multiple subgroups that meet separately and reconvene to confer on their different policy opinions (pp. 265–266)	Establish standing task groups or subcommittees that convene on critical decisions.	Christensen (2019), Cleary et al. (2019), Gambrill (2005), (1994), Mannion & Thompson (2014)
Each member of the policy-making group should reach out to a trusted third party in the same department to discuss the group’s decision-making and report the third party’s opinion (p. 266).	Teams should link up with hospital supervisors, administrators and other service teams in their unit and decrease group insulation. Reactions from the trusted third party should be reported back to the group.	Heinemann (1994), Mannion & Thompson (2014)
Groups should invite outside experts from the organization at large to join their meetings on a staggered basis. They should challenge group members’ core views (pp. 266–267).	Seek outside expertise such as new perspectives from other clinical departments or disciplines and institutions (e.g. universities, hospitals)	Allen (2007); Gambrill (2005); Harmon et al. (2019); Heinemann et al. (1994)
A member should be assigned the role of devil’s advocate at policy evaluation meetings (pp. 267–268).	Appoint and rotate the role of devil’s advocate in the group.	Cleary et al. (2019); Gambril (2005); Heinemann (1994)
Regarding policy issues involving adverse relations with a rival organization, the group should have a planning session evaluating the warning signals and possible intentions of the rival group (pp. 268–270).	N/A	None
After preliminary consensus, the policy-making group should hold a ‘second-chance’ meeting. Before coming to a final decision, the group members should air all doubts and rethink the whole issue (pp. 270–271).	Independent groups should work on critical issues through a process of reconvening and reviewing decisions. Meetings should have ‘second-chance’ motions.	Cleary et al. (2019); Gambrill (2005)
Most common other recommendations (outside Janis)
Reorient focus back to patient care. Allow for patient input, and shared decision making. Enhance patient-provider communication. Use compassion for patient as practical guide		Allen et al. (2009), Harmon et al. (2019); Snelgrove et al. (2011), Cheshire (2017)
Implement an educational program/feedback device tied to intervention and group exercises to mitigate bias, such as role play for learning members’ personalities, personal value systems, etc. to ease tension in real practice		Degnin (2009); Heinemann et al. (1994); Hollinger (2019); Snelgrove (2011)
Hold greater accountability for sharpening critical thinking skills and bolstering moral awareness. Medical professionals are responsible for being wary of latent groupthink.		Cheshire (2017); Snell (2009)

Methods

We performed a scoping review, following the methodological framework outlined by Arksey and O’Malley, in order to examine the extent, range, and nature of work and research activity of our topic arena (Arksey & O’Malley, 2005). Furthermore, scoping reviews (also called scoping studies) may identify gaps in the existing literature and inform future research. Scoping reviews also synthesize broad literature from multiple disciplines concerning interprofessional topics (Levac et al., 2010).

Research question and search strategy

Our research question was: ‘How is groupthink theory conceptualized, studied, and mitigated in the context of health professional teams conducting patient care?’ A scoping review of the literature was conducted using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) checklist (Moher et al., 2010). A medical library information specialist (KP) performed a search through May 20, 2020 of the following databases: MEDLINE (Ovid), Embase (Ovid), Cochrane Library (Wiley), Scopus (Elsevier), Academic Search Premier, Business Source Complete, APA PsycInfo, and CINAHL (last four all EBSCOhost). Keywords used to retrieve articles and details of the specific database strategies are provided in the supplementary material. Search results were exported to Covidence, an online tool for screening.

Study selection

Articles on groupthink among health professional teams in the context of patient care were included in the review. Articles were included if they discussed any of the following: 1) conceptualization of groupthink in health care; 2) empirical research conducted on groupthink in health care; and 3) recommendations to avoid groupthink in health care. Articles on healthcare administration, hospital operations, health policy, or veterinary medicine were excluded as we decided to focus our review on the direct care of patients among a team of health professionals. Articles were also excluded if the groupthink framework was not applied to clinical decisions, there was only brief mention (e.g., in the introduction or discussion) and without focusing the discussion on groupthink, or if the article used a definition of groupthink significantly different from Janis (i.e., did not view groupthink as potentially harmful or a poor decision-making process). Finally, articles were excluded if the full-text could not be accessed, unless sufficient information was provided in the abstract to meet our inclusion criteria.

The initial search identified 365 articles. A total of 153 duplicates were removed. The co-first authors (KP, HL) screened all remaining studies in a two-stage process: (1) title and abstract screening; followed by (2) full text screening. The co-first authors independently decided which articles fit the inclusion and exclusion criteria. Disagreements were resolved through group discussions involving the co-first authors and another investigator of the research team (JC), who made the final decision as the tiebreaker. The research team met weekly throughout the screening and data extraction process. Of 212 articles screened, 123 were excluded after the title and abstract screening and 67 were excluded after the full-text screening. Six articles were excluded during full-text screening because they were not able to be retrieved through library loan. The entire article screening and selection process yielded 22 articles that were included in this review (Figure 1).

Figure 1.

Literature search and article selection process.

Charting the data

Janis’s groupthink theory served as a sensitizing framework that informed our data extraction form. Categories outlined by Janis’s framework for groupthink were included in the data extraction sheet (see supplemental appendix): antecedents of groupthink, symptoms of groupthink, outcomes, and recommendations. Data was extracted by the co-first authors using individual, standardized excel spreadsheets. Articles were coded according to information on article type and characteristics, the aforementioned categories of Janis’s groupthink framework (antecedents, symptoms, outcomes, recommendations), examples (i.e., conceptualization) of groupthink in health care, and empirical research on groupthink. A final extracted data table was created through frequent meetings between the data extractors. Disagreements were settled by the principal investigator after a group discussion.

Results

Study Characteristics

Twenty-two articles were ultimately included in the scoping review (Table 1). The articles were primarily from the United States (n=12), but also included the United Kingdom (n=4), the Netherlands (n=2), Australia (n=1), Canada (n=1), Sweden (n=1) and Switzerland (n=1). Over two-thirds (n=15) were published between 2010 and 2019, with seven articles published in 2019. Five were published between 2000–2009 and only two were published prior to 2000. The article types included commentaries, editorials, or reports (n=7), narrative reviews (n=6), empirical research (n=4), case reports (n=2), quality improvement reports (n=2), and a book chapter (n=1). Two articles were responses to Kaba et al. (2016), which was also included in our review. The articles were published in a range of fields: medical education (n=4), nursing (n=4), cardiovascular medicine (n=2), clinical ethics (n=2), clinical psychology (n=2), general medicine (n=2), ergonomics (n=2), business and healthcare administration (n=1), emergency medicine (n=1), geriatrics (n=1), and patient safety (n=1). The populations studied were primarily multidisciplinary teams (n=13) and nursing (n=7), but also included physician-only teams (n=2).

Table 1 -.

Summary of Literature Characteristics

Author (Year)	Country	Article Type	Field	Healthcare team membership	Description
Allen, Stubbs & Hignett (2009)	United Kingdom	Empirical research: focus group study	Ergonomics	Nursing	Explores why student nurses continue to use the ‘draglift’, a dangerous patient handling technique, and provides recommendations for how educators can encourage safer handling techniques.
Cheshire (2017)	United States of America	Narrative review	Clinical Ethics	Physicians	Applies Janis’s eight symptoms of groupthink to healthcare professionals’ approach to victims of human trafficking.
Christensen (2019)	United States of America	Narrative review	Nursing	Nursing	Reviews the literature on diffusion of responsibility in nursing, drawing connections to Milgram’s and Zimbardo’s research. Groupthink is explored as a type of diffusion of responsibility in groups.
Cleary, Lees & Sayers (2019)	Australia	Commentary, editorial or report	Nursing	Nursing	Explores groupthink, its symptoms, and prevention strategies within nursing groups.
Degnin (2009)	United States of America	Case report	Clinical Ethics	Multidisciplinary care team: physician, psychiatrist, social worker, and nurse.	Provides a case history of a patient who was perceived as ‘difficult’ by her care team and overmedicated as a result. The author’s analysis of the case explores ‘faulty interpersonal and group dynamics’ and their consequences.
Farr & Colvin (2019)	United States of America	Commentary, editorial or report	Cardiovascular medicine	Multidisciplinary team: heart transplantation donor selection programs	Commentary on Baran et. al 2019, which hypothesized that the donor selection number may decrease desirability of a viable organ despite no correlation with risk to the patient. This commentary identifies this phenomenon as groupthink, critiques Baran’s methods and conclusions, and argues that reliance on the donor selection number is not inherently beneficial or detrimental.
Gambrill (2005)	United States of America	Book chapter	Clinical psychology	Multidisciplinary care team: social worker, health visitor, nursing officer, medical social worker, policemen, physicians, registrar, and medical student.	Discusses the factors that influence poor decision making in team meetings and case conferences, identifies strategies to improve the quality of discussion, and offers recommendations on how to improve organizational culture to improve decisions. The author presents a sample medical case conference as an example.
Harmon, Summons & Higgins (2019)	United States of America	Empirical research: focused ethnography	Nursing	Nursing	Explores how pain assessment and pain management for elderly patients are influenced by the nursing culture in an acute care unit. Groupthink is observed in the shared sensemaking of pain care provision among the nurses.
Heinemann, Farrell & Schmitt (1994)	United States of America	Narrative review	Geriatrics	Multidisciplinary team: physicians, nurses.	Reviews Janis’s groupthink theory, then provides a case study that demonstrates groupthink in a geriatric health care team.
Hollinger (2019)	United States of America	Quality Improvement Report	Nursing	Nursing	Describes the implementation of an educational program on risk and groupthink to increase reporting of near-miss occurrences. Its effectiveness was measured through annual surveys.
Jones and Roelofsma (2000)	The Netherlands	Narrative review	Ergonomics	Multidisciplinary team: emergency medicine teams	Describes the effect of social, contextual, and group biases on decision making and their relevance to ‘command and control teams.’ The group biases explored are false consensus, groupthink, group polarization, and group escalation of commitment.
Kaba, Wishart, Fraser, Coderre & McLaughlin (2016)	Canada	Narrative review	Medical education	All healthcare teams	Reviews literature on teamwork interventions and interprofessional education in health care. They identify a need for further research on contextual variables and interventions outside of medicine that have been successful in mitigating social loafing, cognitive overload and groupthink.
Kerry, Schmutz & Eppich (2017)	United States of America	Commentary, editorial or report	Medical education	All healthcare teams	Response to Kaba et al. (2016). The author disagrees with Kaba’s assertion that groupthink undermines collaboration in healthcare, and presents studies that demonstrate the benefits of teamwork in healthcare.
Kolbe et al. (2019)	Switzerland	Commentary, editorial or report	Cardiovascular medicine	Multidisciplinary team: cardiologists, cardiac surgeons, radiologists, anesthesiologists, general practitioners, geriatricians, intensive care specialists	Analyzes the evaluation of potential transcatheter aortic-valve implantation patients through the lens of social psychology. The authors discuss typical pitfalls and risks of collaborative decision-making, and make recommendations to mitigate these risks.
Madigosky & van Schaik (2016)	United States of America	Commentary, editorial or report	Medical Education	All healthcare teams	Response to Kaba et al. (2016). The authors argue that healthcare teams may not be subject to the risks identified by Kaba et al., and if they are, the benefits may outweigh the risks.
Mailoo (2015)	United Kingdom	Commentary, editorial or report	General Medicine	Physicians	Describes individual and group weaknesses in clinical reasoning such as gut feeling, colleague collaboration, and dual process theory. Groupthink is identified as a weakness of clinical decision-making. Examples of groupthink in a clinical setting are provided.
Mannion & Thompson (2014)	United Kingdom	Narrative review	Patient safety	All healthcare teams	Describes four systematic biases in group decision making at different levels of the healthcare system: groupthink, social loafing, group polarization, and escalation of commitment. Each bias is illustrated with a brief case scenario, and remedial strategies are provided.
Mcleod and Feller (2019)	United States of America	Commentary, editorial, or report	General medicine	Multidisciplinary teams: small teams, work groups, committees, or panels that approach a specific clinical problem	Explores how small groups and teams can negatively affect learning and decision making in medicine. The authors describe groupthink, its facilitating factors, and the importance of embracing dissent in teams.
Rosander, Stiwne & Granström (1998)	Sweden	Empirical research: survey	Clinical Psychology	Multidisciplinary teams*: nursing and psychiatric teams	Examines groupthink tendencies in authentic work groups. The researchers make qualitative and quantitative assessments of the mindset of these groups under stressful situations, and find that groups exhibit different types of groupthink to varying degrees.
Schols and Klein Nagelvoort-Schuit (2017)	The Netherlands	Case report	Emergency Medicine	Multidisciplinary care team: emergency medicine team.	Provides a case history of a patient who is discharged from the emergency room with an “atypical chest pain” diagnosis. The diagnostic process and the factors that influenced the care team’s reasoning are discussed.
Snelgrove, Gosling & McAnulty (2011)	United Kingdom	Quality Improvement Report	Medical education	Multidisciplinary teams: acute care clinical teams	Discusses the usefulness of a simulation-based multi-professional training program to improve ethical reasoning and teamwork in acute care physicians and nurses. The authors review the challenges to ethical decision making and effective teamwork and the need for improved education on the subject.
Snell (2009)	United States of America	Empirical research: survey	Business and Healthcare Administration	Nursing	Explores how job stress, job conflict, job ambiguity and the perceptions of nurses and nurse managers influence groupthink in hospitals through a series of questionnaires.

^*Other groups studied included police forces, a technological company, and a religious sect.

How groupthink is conceptualized in clinical practice

Fifteen articles identified precursors to groupthink existing in health care, using at least one of Janis’s antecedent conditions. Janis’s principle of group cohesiveness was identified as an antecedent to groupthink in health care by thirteen articles (Christensen, 2019; Cleary et al., 2019; Degnin, 2009; Gambrill, 2005; Heinemann et al., 1994; Hollinger, 2019; Jones & Roelofsma, 2000; Kaba et al., 2016; Mannion & Thompson, 2014; McLeod & Feller, 2019; Rosander et al., 1998; Snell, 2010). Group cohesiveness emerged either through direct interactions of team members, or at the organizational level of health care, in which health professionals perceived themselves as part of a broader medical community.

Of the fifteen articles that identified antecedents to groupthink, nine posited that health care is particularly prone to groupthink on account of some structural fault of the organization (Christensen, 2019; Cleary et al., 2019; Degnin, 2009; Gambrill, 2005; Harmon, 2019; Heinemann et al., 1994; Hollinger, 2019; Kolbe 2019; Snell 2009). In these articles, characteristics such as homogeneity of group members, insularity of the profession, and close-minded leadership were inherent in health care were identified as organizational structural faults that contribute to groupthink. Hierarchy of medical teams was described as a potential issue in five articles (Allen, 2018; Christensen, 2019; Gambrill, 2005; Harmon et al., 2019; Snell, 2010). For example, the seniority of nurses or physicians could create an inflexible environment and influence the clinical decisions of more junior nurses or physicians. Homogeneity and insularity in health care can be rooted in the similarity of personal backgrounds (Kolbe et al., 2019; Madigosky & van Schaik, 2016). A few articles acknowledged the gradual convergence of analytical thinking processes amongst health professionals as further evidence of homogeneity in health care (Harmon et al., 2019; Mailoo, 2015).

Janis’s concept of a provocative situational context was another commonly identified antecedent of groupthink. Three articles described a provocative environment due to the high-stress nature of medical decision making as an antecedent to groupthink (Heinemann et al., 1994; Jones & Roelofsma, 2000; Rosander et al., 1998). Jones & Roelofsma described the environment of health care as one that is “characterized by dynamically changing task conditions, time pressure, and stress compounded by high stakes.”

Seventeen articles discussed examples of groupthink in teams of health professionals, relating Janis’s “symptoms’’ of groupthink as well as its consequences to clinical decision-making (Allen, 2018; Cheshire, 2017; Christensen, 2019; Degnin, 2009; Farr & Colvin, 2019; Gambrill, 2005; Harmon et al., 2019; Heinemann et al., 1994; Hollinger, 2019; Jones & Roelofsma, 2000; Kaba et al., 2016; Kolbe et al., 2019; Madigosky & van Schaik, 2016; Mailoo, 2015; Mannion & Thompson, 2014; Schols & Nagelvoort-Schuit, 2016; Snelgrove et al., 2011). These articles illustrated the various structures of health professional teams, the interpersonal dynamics that create the setting for groupthink, and the ways groupthink can impact the quality of care and patient safety. For example, groupthink was attributed to nursing students continuing the unsafe practice of using draglifts for transporting patients and nurses overlooking input from elderly patients as part of a more comprehensive pain assessment. In another example, a health professional team administered medications while ignoring the suggestion of external members such as the patient’s healthcare proxy that the patient was overmedicated (Degnin, 2009). Examples of groupthink in medical decision making, particularly in diagnosis, were found during diagnostic discussions at case conferences steered by higher-status members with little opposition in their fallacies, in an emergency medicine team that developed group cohesion after spending four night shifts that led to their over-testing for a patient with “atypical chest pain,” and in another emergency medicine team that relied on a charismatic senior physician during medical decision making (Gambrill, 2005; Mannion, 2014; Schols, 2016).

Three articles discussed groupthink at the organizational level in which individual decision making is shaped by group-based medical training and the broader medical community (Cheshire, 2017; Farr & Colvin, 2019; Mailoo, 2015). Cheshire discussed how health professionals may exhibit groupthink when underestimating the health risks of trafficked individuals due to assumptions of characteristics of their patient population and the inherent morality of medicine. Farr and Colvin (2019) proposed the tacit judgment of a broader medical community as a proxy for groupthink—in their example, the assignment of a donor selection number by one ‘donor call team’ in the evaluation of a viable organ may be influenced by the previous donor selection number assigned by the other teams in the transplant community at large. It is important to note that groupthink in donor selection number assignments is not expressly negative since the implied donor quality could potentially serve as a method for more expeditious decisions when assessing the organ’s suitability for multiple sequential transplant recipients in large programs. Mailoo (2015) provided other examples of influence from a broader medical community: physicians following the decisions of fellow clinicians while eschewing their independent evaluation; copying previous doctors’ prescriptions; writing prescriptions based on nurses’ requests, and performing “focused exams” heavily informed by other doctors’ previous treatment plans in lieu of taking a thorough patient history.

Two articles provided counterpoints to the argument that groupthink is harmful in health care (Kerry et al., 2017; Madigosky & van Schaik, 2016). Health professional teams may not be subject to groupthink due to their heterogeneity in perspectives, experiences, and roles. Furthermore, health care is inherently collaborative and the concern for groupthink is outweighed by the benefits of teamwork in health care. The validity of the groupthink framework was also challenged due to the limited research on groupthink in health care.

Empirical research on groupthink

We identified four empirical studies on groupthink among health professional teams in patient care—two quantitative studies and two qualitative studies (Table 2) (Allen, 2018; Harmon et al., 2019; Rosander et al., 1998; Snell, 2010). Rosander et al. observed six different working groups and conducted the study by first proposing an alternative framework to Janis’s theory, whereby groupthink should be described as a bipolar construct. According to this construct, groupthink theory has omnipotent and depressive variants. The omnipotent variant is closest to Janis’s visualization of a cohesive group, which displays feelings of optimism and superiority over other groups. The depressive variant displays passivity to a leader and delusions about its own and other groups’ features. This bipolar groupthink dynamic was tested via questionnaires, which found that the nursing staff tended toward the mean between omnipotent and depressive groupthink poles, whereas the psychiatric team showed a tendency toward the depressive groupthink pole. This suggests that different teams may have different groupthink tendencies in how they view themselves and how they react in a provocative situation.

Table 2 -.

Empirical research on groupthink

Author (Year)	Study design	Country	Healthcare teams studied	Research aim(s)	Findings and implications for groupthink
Allen (2009)	Focus groups	United States of America	Nursing	To identify the influences on student nurses that shape their continued use of the draglift.	Nursing students experienced pressure to conform, which contributed to vicarious learning, groupthink, and theory-practice gaps.
Harmon et al. (2019)	Focused ethnography	Australia	Nursing	To explore the culturally mediated practices of registered nurses in acute care units when assessing and managing pain for older people in order to provide insight into possible barriers and facilitators.	Groupthink was inferred in relation to how the nurses undertook, interpreted and made sense of pain documentation, how they viewed and mentored the junior nurses on pain care provision, and the lack of input or inclusion from the older person in relation to their perceptions on pain care provision.
Rosander et al (1998)	Questionnaire	Sweden	Nursing and psychiatry teams	1. To identify groupthink tendencies in authentic work teams. 2. To contribute to the issue of operationalization of groupthink, including a “bipolar” concept of groupthink theory.	Psychiatry teams had the most features of the depressive groupthink variant, which is marked by a prevailing feeling of having tried everything with no results, with no solutions to their problems. Nursing teams displayed a balance of features of the depressive groupthink variant and the omnipresent groupthink variant, which is marked by feelings of invulnerability.
Snell (2009)	Survey	United States of America	Nursing	To explore whether job stress, job conflict and job ambiguity has an effect on the incidence of groupthink at an acute care hospital.	Groupthink was found less frequently in the absence of significant stress, job conflict, and job ambiguity. Of the eight main groupthink symptoms, only unanimity was present at moderate levels. The other symptoms of groupthink were at low levels of incidence.

Snell et al. surveyed nurses and nurse managers from two hospitals and found that groupthink occurred less frequently in the absence of significant stress, job conflict, and job ambiguity. Using a ‘Group Think Index’ survey created by the study team to assess how nurses evaluate the decision-making dynamic of their group, they discovered that the only symptom of groupthink in Janis’s framework that was detected at moderate levels was unanimity—all other symptoms of groupthink were at low levels of incidence.

Allen et al. used focus groups to explore patient safety and procedural training amongst nursing students. This study examined the continued use of the draglift for transporting patients, which is considered to be a dangerous practice. Groupthink emerged as a bias that contributed to continued use of the draglift—nursing students indicated that the ‘need to belong’ with the senior nurses superseded the need for safer patient-handling techniques.

Harmon et al. used focused ethnography to explore pain management and attitudes of nurses towards pain care provision for older patients in acute care units. Groupthink was observed in various aspects of the work culture: how nurses undertook and interpreted pain documentation; how they viewed and mentored the junior nurses on pain care provision; and the lack of input gathered from older inpatients on their perceptions on pain care provision.

Of note, we found two quality improvement reports in our search for empirical research, although these were not categorized as empirical studies (Hollinger, 2019; Snelgrove et al., 2011). Snelgrove et al. reported on the benefits of a simulation-based program to improve ethical reasoning and teamwork among acute care physicians. In this program, the simulations used scripted scenario triggers to encourage teams to communicate openly and challenge decisions. After the simulation, the team participated in ‘debriefing’ sessions during which the team sought feedback and reflected on successful teamwork strategies that challenged groupthink. In the second quality improvement report, Hollinger described the effectiveness of a hospital program that encouraged the reporting of ‘near-miss occurrences’ by hospital staff. The program included a 40-minute presentation on groupthink and provided tools for filing reports in an effort to “speak up for patient safety”. The effectiveness of the program was measured by the results of an annual culture of safety survey, which showed improved scores in the three years following the program.

Recommendations to avoid groupthink

Twenty-one articles included recommendations for addressing groupthink. Seven articles adapted at least one of Janis’s recommendations of avoiding groupthink to the clinical setting (Table 3) (Christensen, 2019; Cleary et al., 2019; Degnin, 2009; Gambrill, 2005; Heinemann et al., 1994; Hollinger, 2019; Mannion & Thompson, 2014). The most common recommendation for preventing groupthink outside of Janis’s framework centered around refocusing attention toward patient care and safety (Allen et al., 2009; Degnin, 2009; Harmon et al., 2019; Snelgrove et al., 2011; Cheshire, 2017). These recommendations included focusing on patient communication, opening channels to discuss unwieldy and complex issues on the patient’s goals of care, and maintaining compassion for the patient when faced with difficult decisions regarding care. A focus on patient care and safety may help facilitate and encourage appropriately divergent opinions in the decision making of healthcare teams.

The next most common recommendation outside of Janis’s framework was to implement an educational program as an intervention against groupthink. Four articles suggested group exercises in feedback, such as instructing members to actively dissent in group discussions, scripting scenarios, and engaging in role play (Degnin, 2009; Heinemann et al., 1994; Hollinger, 2019; Snelgrove et al., 2011). Implementing role play could promote the learning of team members’ personalities and personal value systems to ease tension in real practice. Two articles called for health professionals to sharpen their critical thinking skills and bolster their moral awareness (Cheshire, 2017; Snell, 2010). They lay the burden on health professionals to take personal responsibility in constantly being wary of latent groupthink.

Discussion

This scoping review aimed to understand how groupthink is conceptualized, studied, and mitigated in the context of health professional teams conducting patient care. Our results suggest that this is a topic area that is an emerging area of interest. Although it has been a half-century since the inception of groupthink theory, only two articles were published prior to 2000, over two-thirds in the past decade, and almost one-third just in the past year leading up to this scoping review. In our review process, we identified several major themes worth further discussion.

There is disagreement in the conceptualization of groupthink and whether the benefits of teamwork outweigh the potential harms of groupthink among health professional teams in clinical practice. One perspective is that health professional teams are typically homogenous in background (i.e., health professional training) and professional values, and may exhibit convergence in thinking that promotes group cohesiveness that leads to poor decision making. While most agree that group cohesiveness is an antecedent to groupthink in health care, some offer that group cohesiveness is a positive force rather than a weakness of groups (i.e., “the wisdom of the crowd”).

The disagreement on the conceptualization of groupthink may be explained by the fact that health care is delivered by multi-layered, heterogeneous teams. Teams, defined as two or more individuals who interact interdependently with a common purpose, may have members of varying levels of authority and responsibility (e.g., a supervisor and a trainee), varying expertise (e.g., general expertise versus specialty expertise), varying training and practice (e.g., by department/specialty such as cardiology versus oncology, or by healthcare profession such as medicine, nursing, physical and occupational therapy, nutrition, pharmacy, etc.). There are also other factors to consider in the make-up of a team, such as demographic factors, cultural factors, personality, and leadership style. Team heterogeneity in these various factors may foster a power dynamic between members of a health professional team. The team leader may have a special role in this power dynamic. Janis notes that “the group’s agenda can readily be manipulated by a suave leader, often with tacit approval of the members,” a condition that fosters groupthink (Janis, 1982). Janis found little evidence that the extreme of directive leadership, termed one-man rule, is an alternative to the groupthink hypothesis and instead promotes concurrence-seeking (i.e., groupthink). In health care, it may be difficult to separate directive leadership from groupthink as it appears that the former can be an antecedent condition of the latter.

We found few empirical studies on groupthink among healthcare teams in clinical practice, primarily qualitative or survey-based studies that sought the opinions of health professionals. There appears to be a significant gap in the literature on both quantitative and qualitative studies on groupthink, including those gathering validity evidence for the groupthink framework in health care, developing measures of groupthink in health care, and testing interventions that mitigate groupthink in health care. There is also a need for further investigation into how groupthink manifests in healthcare—what it is and mechanistically how it occurs.

Many of the recommendations we identified in articles to counter groupthink among healthcare teams followed Janis’s recommendations outlined in his original framework. Other recommendations were provided as well, including the need to identify and measure groupthink, discourage it in the workplace, and educate trainees on groupthink and effective teamwork strategies. Current training programs of health professionals may not be adequately teaching these skills, and there may be an increasing need for interprofessional healthcare training programs to address limitations in teamwork and group decision making.

A strength of our review is that it provides an exploration of an understudied theory and potential bias in group decision making in the context of clinical practice, a setting in which teamwork and collaboration is critical and often with high stakes. However, there are several study limitations. First, we excluded articles that made brief mention of groupthink without a substantial discussion on the topic. We mitigated potential inconsistency in this determination by performing independent reviews at each stage of screening, with extensive discussions and a tie-breaker vote for any disagreements. Second, we only included articles that specifically applied groupthink to patient care decisions and actions. By limiting our inclusion criteria to clinical decisions, a number of articles on groupthink in hospital administration and management were excluded.

This review demonstrates that there is increasing interest and awareness that health professional teams may be vulnerable to groupthink; however, there is conceptual disagreement on how to interpret groupthink in the context of clinical practice. The impact of groupthink on clinical and educational outcomes is unclear, particularly with regards to the quality of care and patient safety. There are significant gaps in ways to identify, measure, and mitigate groupthink in clinical practice. We recommend the following research priorities to advance the field of teamwork in clinical practice and interprofessional education: (1) developing a theoretical framework that applies groupthink theory to clinical decision making and medical education (2) gathering validity evidence of Janis’s groupthink framework through empirical study; (3) developing measures of groupthink, as well as positive group decision making processes; (4) elucidating the “pathophysiology” of groupthink theory and other group decision making processes in real-world situations; (5) evaluating team interventions that mitigate groupthink and faulty group-decision making processes; (6) examining how groupthink may be situated amidst other emerging social cognitive theories of collaborative clinical decision making; (7) exploring the role of patients and families in groupthink among health professionals.

Practice Points.

• Improve health professionals’ understanding of groupthink through educational programs to enhance patient care quality and safety.

• Be wary of groupthink in the presence of a directive leadership style and team hierarchy—team leaders should welcome criticism of their own opinions and encourage candidness.

• Appoint and rotate the role of ‘devil’s advocate’ in the health professional team to promote critical evaluation of group decisions.

• Allow for patient input and shared decision making in health care team decisions to mitigate negative consequences of groupthink.

Glossary

Groupthink

A theory that describes when highly cohesive groups exhibit premature consensus seeking that leads to poor decision making

Systematic bias

A pattern of deviation from rational judgment and decision making that may affect individuals or teams

Biographies

Karissa DiPierro is an undergraduate student at Cornell University concentrating in Microbiology. She is expected to obtain her Bachelor of Science in Biological Sciences in May of 2022.

Hannah Lee, BA, MSt, MPhil, is a research associate in the Department of Medicine at Weill Cornell Medicine in New York, NY. She is currently pursuing studies in a post-baccalaureate premedical program.

Kevin J. Pain, BA, is an Information and Library Research Specialist at Weill Cornell Medicine in New York, NY.

Steven J. Durning, MD, PhD, is a Professor of Medicine and Pathology at the Uniformed Services University in Bethesda, MD, and is the Director of Graduate Programs in Health Professions Education.

Justin J. Choi, MD, MSc, is an Assistant Professor of Medicine in the Division of General Internal Medicine at Weill Cornell Medicine in New York, NY.