ArticlePDF Available

Making It Safe: The Effects of Leader Inclusiveness and Professional Status on Psychological Safety and Improvement Efforts in Health Care Teams

November 2006
Journal of Organizational Behavior 27(7):941 - 966

Authors:

This paper introduces the construct of leader inclusiveness—words and deeds exhibited by leaders that invite and appreciate others' contributions. We propose that leader inclusiveness helps cross-disciplinary teams overcome the inhibiting effects of status differences, allowing members to collaborate in process improvement. The existence of a professional hierarchy in medicine and the differential status accorded to those in different disciplines is well established in the health care literature, as is the need for quality improvement. We build on this foundation to suggest that profession-derived status is positively associated with psychological safety (H1)—a key antecedent of speaking up and learning behavior—in health care teams. We hypothesize that this effect varies across teams (H2), and furthermore, that leader inclusiveness predicts psychological safety (H3) and moderates the relationship between status and psychological safety (H4). Finally, we suggest psychological safety predicts engagement in quality improvement work (H5) and mediates the relationship between leader inclusiveness and engagement (H6). Survey data collected in 23 neonatal intensive care units involved in quality improvement projects support our hypotheses. These results provide insight into antecedents of and strategies for fostering improvement efforts in health care and other sectors in which cross-disciplinary teams engage in collaborative learning to improve products or services. Copyright © 2006 John Wiley & Sons, Ltd.

A model of professional status on psychological safety as a precursor to engagement in quality improvement work: the moderating effect of leader inclusiveness and the mediating effect of psychological safety

…

Psychological safety as a function of professional status and leader inclusiveness

…

. Results of general linear model analysis

…

Content may be subject to copyright.

Content uploaded by Amy Edmondson

Content may be subject to copyright.

Making it safe: The effects of leader

inclusiveness and professional status on

psychological safety and improvement

efforts in health care teams

INGRID M. NEMBHARD

*AND AMY C. EDMONDSON

Graduate School of Arts and Sciences, Graduate School of Business Administration, Harvard

University, Boston, Massachusetts, U.S.A

Graduate School of Business Administration, Harvard University, Boston, Massachusetts, U.S.A.

Summary This paper introduces the construct of leader inclusiveness —words and deeds exhibited by

leaders that invite and appreciate others’ contributions. We propose that leader inclusiveness

helps cross-disciplinary teams overcome the inhibiting effects of status differences, allowing

members to collaborate in process improvement. The existence of a professional hierarchy in

medicine and the differential status accorded to those in different disciplines is well

established in the health care literature, as is the need for quality improvement. We build

on this foundation to suggest that profession-derived status is positively associated with

psychological safety (H1) —a key antecedent of speaking up and learning behavior —in health

care teams. We hypothesize that this effect varies across teams (H2), and furthermore, that

leader inclusiveness predicts psychological safety (H3) and moderates the relationship

between status and psychological safety (H4). Finally, we suggest psychological safety

predicts engagement in quality improvement work (H5) and mediates the relationship between

leader inclusiveness and engagement (H6). Survey data collected in 23 neonatal intensive care

units involved in quality improvement projects support our hypotheses. These results provide

insight into antecedents of and strategies for fostering improvement efforts in health care and

other sectors in which cross-disciplinary teams engage in collaborative learning to improve

products or services. Copyright #2006 John Wiley & Sons, Ltd.

Introduction

In today’s complex organizations, teams are increasingly valued for their potential to innovate, solve

problems, and implement change. A growing literature on team learning identiﬁes factors that allow

Journal of Organizational Behaviour

J. Organiz. Behav. 27, 941–966 (2006)

Published online in Wiley InterScience

(www.interscience.wiley.com) DOI: 10.1002/job.413

* Correspondence to: Ingrid M. Nembhard, Graduate School of Arts and Sciences, Graduate School of Business Administration,

Harvard University, Soldiers Field Road, Boston, MA 02163 U.S.A. E-mail: inembhard@hbs.edu

Contract/grant sponsor: Harvard Business School Division of Research.

Received 28 January 2005

Revised 15 September 2005

Accepted 29 June 2006

teams to experiment, reﬂect, and improve across a range of industry settings (Bunderson, 2003a, 2003b;

Edmondson, 1999; Gibson & Vermeulen, 2003). Few industries have more at stake when teams learn —

or fail to learn—than health care. Increasingly, cross-disciplinary teams are responsible for delivering

care to patients in settings ranging from primary care to critical acute care, chronic care, geriatrics, and

end-of-life care (Institute of Medicine, 2001). These teams face not only a daunting expansion of

medical knowledge but also increasing specialization that divides critical knowledge among

individuals— knowledge that must be integrated for the delivery of quality care, as well as for

improving care.

In this environment, the combined challenges of teamwork and learning are emerging as central to

the health care delivery enterprise, in particular because research has shown that 70 to 80% of medical

errors are related to interactions within the health care team (Schaefer, Helmreich, & Scheideggar,

1994). This article seeks to advance theory in organizational behavior and health care management by

proposing and testing a model of engagement in team-based quality improvement work. In the next

section, we review critical trends in health care delivery to set the stage for our model and empirical

research.

Dynamic trends in health care

Health care professionals today face a staggering rate of change in medical knowledge. Whereas in

1966, only 100 published articles reported on randomized control trials —in medicine, the ‘‘gold

standard’’ for recognizing new knowledge—1995 brought more than 10,000 (Chassin, 1998). In terms

of sheer volume of new information, the Medline bibliographic database adds 30,000 new references

each month, and the Federal Drug Administration reviews thousands of applications for new devices

and drugs annually (Shine, 2002). No single individual can absorb all of this new knowledge in a timely

manner. Nevertheless, new knowledge must be absorbed for continued effectiveness in health care

delivery.

A second crucial trend is the increasing specialization of health care professionals (Hafferty & Light,

1995). Prior to 1930, there were only two boarded medical specialties (ophthalmology and

otolaryngology). Today, there are 26 specialties and 93 subspecialties within the major specialties, 8 of

which were approved during the 2002–2003 accrediting year (Accreditation Council for Graduate

Medical Education, 2004). Thus, the scope of an individual physician’s domain of expertise has

diminished, while depth of expertise has increased. At the same time, a growing number of non-

physician professions have joined the patient care enterprise. Specialists in nutrition, respiratory

therapy, physical therapy, phlebotomy, and so on, have joined nurses as non-physician caregivers,

playing vital roles in the health care system. In 1900, the ratio of physicians to non-physicians was 1 in

3; by 2000, it had exploded to 1 in 16 (Shine, 2002), implicating greater fragmentation of expertise, and

more hand-offs in the patient care process (Leape et al., 1995). Today, an increasing number of different

caregivers treat each patient at the bedside. Each brings information necessary and relevant for

development of a cohesive care plan.

A third trend—almost a necessary outcome of the ﬁrst two— is increasing interdependence. Many

new technologies and care practices involve reciprocal (as opposed to sequential) interactions.

Caregivers cannot simply do their jobs and assume others will come along at some point to do theirs.

Instead, their knowledge and efforts must be integrated to deliver quality care.

These trends—increasing knowledge, specialization, and interdependence—which are more

prominent in health care than in other industries, together imply a need for collaborative learning in

groups of professionals from different disciplines. The modern intensive care unit (ICU) is said to

DOI: 10.1002/job

942 I. M. NEMBHARD AND A. C. EDMONDSON

exemplify the conﬂuence of these factors (Wachter, 2004), but all health care settings have been

touched by these trends. Collaboration — deﬁned as ‘‘physicians and nurses [and other caregivers]

working together, sharing responsibility for solving problems, and making decisions to formulate and

carry out plans for patient care’’ (Baggs et al., 1999, p. 1991)— requires open communication and

mutual respect in addition to collective decision making (Baggs et al., 1999; Brown, Ohlinger, Rusk,

Delmore, & Ittmann, 2003; Zimmerman et al., 1993) and is critical to care delivery. Health

professionals themselves have recognized the imperative for teamwork in clinical care and quality

improvement, even suggesting that the latter needs to be based in cross-disciplinary teams (Donaldson

& Mohr, 2000), a suggestion that has been embraced by many (e.g., Berwick, Godfrey, & Roessner,

1990; Horbar, 1999).

Barriers to collaborative learning in health care

Despite the need for collaborative learning in cross-disciplinary teams in the health care setting, team-

based quality improvement efforts may stall for a number of reasons. First, the stakes are undeniably

high. Human life is at risk when processes fail, creating understandable risk aversion that can inhibit

willingness to engage in the chaos and uncertainty of team brainstorming and experimentation. It is

noteworthy that studies of other high stress environments have shown that improvement efforts tend to

be centralized and hierarchical rather than collective and democratic (Driskell & Salas, 1991; Foushee

& Helmreich, 1988; Hermann, 1963; Klein, 1976; Staw, Sandelands, & Dutton, 1981).

Second, cross-disciplinary teamwork —intended to integrate knowledge and expertise from different

sources—is difﬁcult to carry out in practice (e.g., see Edmondson, Roberto, & Watkins, 2003 for a

review). Improving the quality of care delivery processes necessarily requires different viewpoints,

each grounded in deep knowledge of a different aspect of the process. Physicians possess specialized

medical expertise, while nurses and allied health workers (e.g., respiratory therapists and dietitians)

have greater knowledge of daily patient-interaction processes. In combination, they contain a more

comprehensive information base. However, information often goes unshared. A recent study showed

that, although nurses witness and experience a variety of problems and employ a number of creative

solutions to resolve emergent issues, they generally do not communicate these to others in the hierarchy

(Tucker & Edmondson, 2003). Thus, despite its importance for improving care delivery, collaborative

learning does not occur naturally in health care.

Third, and a central focus of this paper, a well-entrenched status hierarchy exists in medicine,

making it difﬁcult to speak across professional boundaries (e.g., physician vs. nurse vs. therapists) to

collaborate for learning (Edmondson, 2003). The medical training that instills a culture of autonomy

for action can diminish professionals’ tendencies to seek opportunities to learn to communicate, share

authority, and collaborate in problem-solving and quality improvement (Institute of Medicine, 1999,

2001). Unfortunately, this reluctance can adversely affect patient care. Patient outcomes are

signiﬁcantly correlated with the degree of hierarchy in health care team interactions (Feiger & Schmitt,

1979). According to a recent report, Keeping Patients Safe:Transforming the Work Environment of

Nurses, ‘‘counterproductive hierarchical communication patterns that derive from status differences’’

are partly responsible for many medical errors (Institute of Medicine, 2003, p.361). In a content

analysis of medical malpractice cases from across the country, physicians (the high-status members of

the team) were shown to have ignored important information communicated by nurses (the low-status

members of the team), and nurses also withheld relevant information for diagnosis and treatment from

physicians (Schmitt, 1990). In this status-consciousness environment, opportunities for learning and

DOI: 10.1002/job

STATUS AND PSYCHOLOGICAL SAFETY 943

improvement can be missed because of unwillingness to engage in quality-improving communication

due to fear of reprisal by high-status others.

Aims of the present study

The present study investigates factors that promote engagement in quality improvement work when

status differences are present in teams. We examine the relationship between status and psychological

safety, and introduce the construct of leader inclusiveness, deﬁned below, as a moderator of the status–

psychological safety relationship. We then assess whether psychological safety mediates the

relationship between leader inclusiveness and engagement in improvement efforts. At a time when

quality improvement in health care is viewed as imperative, we aim to provide insight into how to

overcome an important barrier to quality improvement learning efforts.

The implications of this study extend beyond the health care industry, however. With customer

heterogeneity, a high need for customization, and a highly specialized workforce, hospitals present a

challenging, but by no means unique, service setting. Furthermore, the use of cross-disciplinary teams

continues to rise across industries as organizations seek to learn and innovate to remain competitive

(Grifﬁn, 1997; Sarin & Mahajan, 2001; Wind & Mahajan, 1997). The salience of status in health care

though provides an opportunity to investigate its effects on team-based improvement efforts and reveal

implications for other organizations that use teams encompassing status differences to improve

products or services. Therefore, our aim is to contribute to knowledge of the role of status in shaping

perceptions of psychological safety and, more broadly, the conditions that support improvement and

learning in cross-disciplinary teams.

Collaborative Learning in Cross-Disciplinary Health Care Teams

Status in professional hierarchies as a determinant of psychological safety

Status refers to the level of prominence, respect, and inﬂuence associated with an individual as a

result of some characteristic (Anderson, John, Keltner, & Kring, 2001), such as age, education,

ethnicity, gender, organizational position, profession, wealth, etc. (Bacharach, Bamberger, &

Mundell, 1993; Benoit-Smullyan, 1944). According to status characteristics theory, personal

characteristics affect self- and other-assessments and beliefs about the individual and his or her

performance abilities (Berger, Cohen, & Zelditch, 1972; Berger, Rosenholtz, & Zelditch, 1980).

Status characteristics can be ‘‘diffuse’’ meaning they are applicable over a range of settings (e.g.,

age, gender, or ethnicity) or ‘‘speciﬁc’’ meaning they provide cues about a person’s expertise

relative to a task (e.g., education or professional training). In either case, individuals that possess

the esteemed characteristic or more of that characteristic are judged superior to those with less of

that attribute. Thus, status indicates relative position within a social hierarchy (Benoit-Smullyan,

1944).

In the United States, social status is often role- or profession-based. Status differences are thus ‘‘most

salient in the work context where they often have practical implications (e.g., education level

determining pay level)’’ (Bacharach et al., 1993, p. 24). Higher status individuals receive more tangible

DOI: 10.1002/job

944 I. M. NEMBHARD AND A. C. EDMONDSON

and intangible beneﬁts in the workplace than their lower ranked co-workers. They gain power over the

actions of others (e.g., determining co-worker schedules and tasks), prestige or the right to occupy

honorary places (e.g., a windowed corner ofﬁce), a reputation for signiﬁcant contributions leading to

greater solicitations of their opinions, the beneﬁt of the doubt in ambiguous situations, and ﬁnancial

rewards (e.g., higher salary and department budgets).

The allocation of beneﬁts which favors high-status individuals over lower status individuals

shapes the environment they share as well as interpersonal interactions (e.g., Alderfer, 1987).

Individual awareness or beliefs that membership in a particular group (e.g., profession) bestows a

certain level of status creates feelings of superiority or inferiority that consistently govern behavior

so as to preserve the hierarchy (Tajfel & Turner, 1986; Webster & Foschi, 1988). Compared to high-

status individuals, those with low status are more likely to suffer low self-efﬁcacy and

underestimate their contribution to work tasks (Berger, Fisek, Norman, & Zelditch, 1985), and

therefore withhold valid information (Argyris, 1985), defer decision rights to higher status others

(Driskell & Salas, 1991), limit their organizational citizenship behavior (Stamper & Van Dyne,

2001), and speak less (Kirchler & Davis, 1986; Pagliari & Grimshaw, 2002; Vinokur, Burnstein,

Sechrest, & Wortman, 1985; Weisband, Schneider, & Connolly, 1995). Consequently,

organizations rely heavily on high-status individuals, which is beneﬁcial when status corresponds

to the expertise required for the task. However, empirical research suggests that individuals often

fail to recognize the expertise held by multiple team members— to the detriment of group and

organizational goals (e.g., see Littlepage, Robison, & Reddington, 1997). Organizational

innovation and improvement, in particular, suffer when minority opinions are ignored (Nemeth,

1986). Nevertheless, inadequate identiﬁcation of valid contributions continues because social

hierarchy leads to the domination of high-status individuals and self-censoring by low-status

individuals. The latter relates to perceptions of risk to self and fear of negative repercussions (e.g.,

public reprimand or assignment to a ‘‘bad’’ work shift).

Research on organizational silence indicates that sense of threat and/or risk is a key determinant

of employees’ willingness to speak up freely (Ashford, Rothbard, Piderit, & Dutton, 1998; Detert

& Edmondson, 2005; Edmondson, 2003; Milliken, Morrison, & Hewlin, 2003; Morrison & Phelps,

1999; Ryan & Oestreich, 1991). Speaking up freely occurs when people are not constrained by the

possibility of others’ disapproval and/or the negative personal consequences that might accrue to

them as a result—a state of psychological safety. In most organizations, those with high status have

more control over formal appraisals and resources than those with low status, and thus may

experience a certain freedom of self-expression in front of others, that low-status individuals do not

enjoy. Research on politeness shows that those with low status employ more ‘‘facework’’ (face

saving verbal strategies) when addressing those with higher status than the other way around

(Brown & Levinson, 1987). With increased status, people exercise less concern about damaging

others’ face; opinions can be freely voiced, and requests can be made of others without verbal

compensation to convey apology, humility, or deference. This well-documented inverse

relationship between status and politeness suggests corresponding differences in psychological

safety across different status groups.

In general, high-status individuals tacitly assume that their voice is valued. People with high status —

role-based and demographic—are more likely to be asked for their opinion than those with low status.

Accustomed to having their opinions sought—often in formal capacities—they learn to offer them

freely. They thus do not perceive the same level of interpersonal risk associated with self-expression

experienced by those with low status. Qualitative evidence of this difference in psychological safety

according to status was found in Kahn’s (1990) study of an architectural ﬁrm and a summer camp;

informants described their interactions with those positioned higher in the hierarchy as more stiﬂing

and threatening than their interactions with peers. Lower status individuals in Kahn’s study reported a

DOI: 10.1002/job

STATUS AND PSYCHOLOGICAL SAFETY 945

lack of conﬁdence that higher status individuals would not embarrass or reject them for sharing

contradictory thoughts.

Individuals in lower positions in the medical hierarchy (Helmreich, 2000) may feel a similar sense of

fear about speaking up across status boundaries, for example, to raise a concern or challenge a current

practice. The existence of a professional hierarchy in medicine and the differential status accorded to

those who occupy different positions within that hierarchy is well-known to health professionals and

well established in the health care literature (Coburn, 1992; DeSantis, 1980; Dingwall, 1974; Friedson,

1970a,b; Fuchs, 1974; Hafferty & Wolinsky, 1991; Shortell, 1974; Wolinsky, 1988). We know that

surgeons garner more prestige than other specialty physicians, that specialty physicians rank above

primary care physicians, that physicians possess more power than nurses, nurses than physical

therapists, and so on. The hierarchy and related status differences exist within professional groups (e.g.,

physicians: specialists vs. primary care) and between groups (e.g., physicians vs. non-physicians), with

between group status differences being most salient and largest. We therefore predict, based on prior

research on the behaviors of different status groups, that non-physicians, as lower status health

professionals, view the cross-disciplinary team climate as less psychologically safe than higher status

individuals such as physicians:

Hypothesis 1 (H1): In cross-disciplinary teams, higher status individuals will experience greater

psychological safety than lower status individuals. (In cross-disciplinary health care teams,

physicians will experience more psychological safety than nurses, who will experience more

psychological safety than respiratory therapists).

The proposed main effect between status and psychological safety may vary depending on how

status differences are handled in a work group. Prior research found that psychological safety can

vary signiﬁcantly across groups, even within the same strong organizational culture (Edmondson,

1996, 1999). Bringing together professionals with different backgrounds and expertise may

exacerbate this variance, if some groups handle the challenge of managing these differences more

skillfully than others (Bunderson, 2003b). When status differences are present in a team, team

members must manage a tension between the norms of collaboration that underlie the notion of

teamwork and the reality of status differences. Prior research showed that the degree of status

derived from different professions can vary across cross-disciplinary work groups (Bunderson,

2003a). Similarly, the formal status associated with a given profession in health care may be treated

differently in different work groups called ‘‘units’’ in health care. The unit is a cross-disciplinary

care team, consisting of all of the staff that participates in delivering a speciﬁc domain of clinical

care, for example, intensive care or cardiac care. Differences in status handling across work groups

in other settings suggests the effect of role-based professional status on psychological safety will

also vary across cross-disciplinary unit teams—indicating an interaction effect between status and

unit team membership on psychological safety:

Hypothesis 2 (H2): Status and unit team membership will interact to predict psychological safety.

Bunderson (2003a) identiﬁed two group attributes —average tenure and power centralization — that

helped explain differences in status effects across groups, and suggested that future research consider

alternative moderators such as task interdependence (Wageman, 1995) and task uncertainty (Van de

Ven, Delbecq, & Koenig, 1976). Other research suggests that how work is organized affects status

relationships. In their study of specialized AIDS units versus general medical units treating AIDS

patients, Aiken and Sloane (1997) found that the relative status of nurses was increased after the

conversion to specialized units. They attributed the elevation in status to an increased appreciation for

nurses’ specialized knowledge and client differentiation. We explore another possibility— that leader

DOI: 10.1002/job

946 I. M. NEMBHARD AND A. C. EDMONDSON

behavior, particularly in its level of inclusiveness, can frame the meaning of status differently across

units, even with similar structures or work design (Edmondson, 2003), helping to explain why status

effects might vary across work groups.

Leader inclusiveness

Team leader behaviors, in general, have been shown to affect the internal dynamics of a team, in

particular inﬂuencing team climate and learning orientation (Baker, Murray, & Tasa, 1995;

Edmondson, 1999; Hult, Hurley, Guinipero, & Nichols, 2000; Madhavan & Grover, 1998; Norrgren &

Schaller, 1999; Shortell, Rousseau, Gillies, Devers, & Simons, 1991; Yukl, 1994; Zimmerman et al.,

1993). Team members are highly attuned to the behavior of leaders and examine leader actions for

information about what is expected and acceptable in team interactions (Tyler & Lind, 1992). If a leader

takes an authoritarian, unsupportive, or defensive stance, team members are more likely to feel that

speaking up in the team is unsafe. In contrast, if a leader is democratic, supportive, and welcomes

questions and challenges, team members are likely to feel greater psychological safety in the team and

in their interactions with each other.

Preliminary evidence of leadership effects on psychological safety emerged in a study of medication

errors in nursing teams (Edmondson, 1996). In some units, nurses described nurse managers as

authoritarian and also expressed deep fears about being reprimanded for revealing mistakes. In

contrast, nurses in other units felt safe speaking up about errors because their nurse manager had

stressed the importance of using this information as a learning tool for the unit. In a later study of

cardiac surgery teams, qualitative data suggested that teams with leaders who actively invited others’

input had higher psychological safety than those in which this behavior was absent. Qualitative data

suggested that surgeon team leaders handled status differences within their teams differently

(Edmondson, 2003). Although all teams comprised four professional roles, with clear traditional status

differences, in some, the surgeons (those with the highest status) made an explicit effort to invite others’

input to help the team implement a new technology. In other teams, the leaders were not similarly

proactive.

Building on these qualitative insights, we propose the construct of leader inclusiveness, deﬁned as

words and deeds by a leader or leaders that indicate an invitation and appreciation for others’

contributions. Leader inclusiveness captures attempts by leaders to include others in discussions and

decisions in which their voices and perspectives might otherwise be absent. It is related to team leader

coaching behavior, which describes team leader behaviors that facilitate group process and provide

clariﬁcation and feedback (Baron, 1990; Edmondson, 1999), and to participative leadership, which

describes leaders that consult with workers, participate in shared decision-making and delegate

decision-making authority to subordinates (Bass, 1990; McGregor, 1960; Yukl, 1994). Leader

inclusiveness differs from these constructs in that it directly pertains to situations characterized by

status or power differences and pertains more narrowly to behaviors that invite and acknowledge

others’ views. The established constructs did not capture sufﬁciently the difﬁculty of lower social

position, nor the behavior of explicit invitation designed to overcome communication boundaries.

Leader inclusiveness describes behavior that, through direct invitation, should create psychological

safety for speaking up. We suggest that both invitation and appreciation are needed to convey the

More speciﬁcally, team leader coaching describes the leader as an operational facilitator; participative leadership emphasizes

harnessing employee motivation to maximize productivity. Thus, neither construct is intended— as is leader inclusiveness— to

address status salient environments, nor to identify the two-pronged strategy of invitation and appreciation.

DOI: 10.1002/job

STATUS AND PSYCHOLOGICAL SAFETY 947

inclusiveness that helps people believe that their voices are genuinely valued. Without a recognizable

invitation, impressions derived from the historic lack of invitation will prevail. And without

appreciation (i.e., a positive, constructive response), the initial positive impact of being invited to

provide input will be insufﬁcient to overcome the subsequent hurdle presented by status boundaries.

Deﬁning leader inclusiveness in this way, we hypothesize that:

Hypothesis 3 (H3): Leader inclusiveness is positively associated with psychological safety.

Finally, we predict that leader inclusiveness will alter the status–psychological safety relationship as

follows. When leaders demonstrate inclusiveness, lower status others are likely to feel supported and to

believe the leaders see them as important members of the team. An atmosphere of mutual respect across

the different professions may develop, in which the specialized expertise held by those with low status

is seen as valuable to the team’s shared task. This helps to equalize the value associated with all

members’ contributions, promoting an egalitarian context. These efforts may increase the level of

psychological safety felt by everyone in the team, but the effect is likely to be greater for low-status

individuals who have less prior experience with others expressing interest in their input than it will be

for those with high status. In contrast, when leader inclusiveness is low, a lack of opportunity to

overcome traditional status barriers allows them to prevail, such that low-status individuals fail to

experience an elevation in psychological safety, leaving high-status individuals more advantaged in this

respect. Thus, we hypothesize:

Hypothesis 4 (H4): Leader inclusiveness moderates the relationship between status and

psychological safety.

Engagement in quality improvement

The need for active quality improvement in health care has been widely recognized since the Institute

of Medicine released its 1999 report documenting rampant failures in the health care system.

Improving the quality of work processes and outcomes requires effort and engagement—which we

deﬁne, drawing on Kahn (1990) as being physically, cognitively, and/or emotionally connected to the

improvement work. Engagement is essential for overcoming powerful barriers to quality improvement

that exist in the health care setting, as well as in other busy and chaotic service contexts. Health care

professionals are often stretched thin, barely able to complete their required tasks in the workday, let

alone devote time to improving the system (Tucker & Edmondson, 2003). Participating in quality

improvement efforts thus requires deliberate and effortful allocation of time. Yet, despite time and

resource constraints, many in health care are embracing quality improvement projects, because of what

is at stake when systems fail. The construct of engagement captures the commitment and effort these

individuals devote to quality improvement.

We argue that engagement in quality improvement in health care is likely to be enabled by

psychological safety. First, engaging in quality improvement requires team members to be willing to try

new technologies and procedures, remaining cognitively ‘‘mindful’’ of relationships between tasks and

team members (Weick, 2002) and emotionally open to giving and receiving feedback in these states of

transition. These behaviors can be interpersonally risky and thus more likely to be found when

psychological safety is present. In a psychologically safe environment, team members do not feel they

must be guarded in their behavior, instead feeling encouraged to question current practices and to share

what may be regarded as provocative ideas, challenging the group to develop more innovative

solutions. Without psychological safety, suggesting new procedures, overstepping professional status

boundaries (Kahn, 1990), or offering unsolicited feedback would seem overly risky.

DOI: 10.1002/job

948 I. M. NEMBHARD AND A. C. EDMONDSON

Second, in a related vein, researchers have argued and shown that individuals’ willingness to

participate in problem-solving activities diminishes signiﬁcantly when they view the team as hostile

(e.g., Dutton, 1993; MacDufﬁe, 1997). In fact, they are more likely to act in ways that diminish learning

behavior (Argyris & Scho

¨n, 1978). One such way is to withdraw from the team and its work. Kahn

(1990) described this as personal disengagement. We thus expect an association between psychological

safety and engagement in quality improvement work in health care. Psychological safety creates the

willingness to change personal habits (Schein & Bennis, 1965), and should allow team members to be

enthusiastic about improvement and their role in that process.

Hypothesis 5 (H5): Psychological safety is positively associated with engagement in quality

improvement work.

Combining Hypotheses 5 and 3, we predict that psychological safety will mediate a relationship

between leader inclusiveness and engagement:

Hypothesis 6 (H6): Psychological safety mediates a relationship between leader inclusiveness and

engagement in quality improvement work.

Together our hypotheses comprise a model of engagement in quality improvement work that

includes antecedents and consequences of psychological safety for health care improvement teams

(See Figure 1). We hypothesize that professional status (H1) and its interaction with unit team

membership (H2) explain variance in psychological safety. Moreover, the magnitude of the status

inﬂuence is moderated by leader inclusiveness (H4), an additional predictor of psychological safety

(H3). Finally, psychological safety enables engagement in quality improvement work (H5), and

mediates between leader inclusiveness and engagement (H6).

Figure 1. A model of professional status on psychological safety as a precursor to engagement in quality

improvement work: the moderating effect of leader inclusiveness and the mediating effect of psychological safety

DOI: 10.1002/job

STATUS AND PSYCHOLOGICAL SAFETY 949

Organizational Context

The Quality Improvement Imperative in Health Care

In 1998, the National Roundtable on Health Care Quality issued a statement, The Urgent Need

to Improve Quality of Care, which documented signiﬁcant quality problems in the American

health care system, problems ranging from overuse of antibiotics to underuse of beta blockers

following heart attack to misuse inherent in medical error (Chassin, Galvin, & The National

Roundtable on Health Care Quality, 1998). Shortly thereafter, the Institute of Medicine

published ToErrisHuman(1999) and Crossing the Quality Chasm (2001), which further

quantiﬁed the extent of preventable medical error and identiﬁed the systemic nature of quality

problems. In response to these reports, health care organizations across the country from

individual providers to community health centers to hospital systems bolstered their efforts to

improve the quality of care they delivered (Wachter, 2004). Continuous quality improvement

became a community ideal, to which every organization strove to achieve. The health care

organizations we studied were no different.

Proﬁle of Participating Unit Teams

We collected data for this study from NICUs in the United States and Canada. NICUs provide

care for premature infants, infants weighing less than 1500 g and infants born with complications.

All of the NICUs studied were members of the 2-year, 44-member Neonatal Intensive Care Quality

Improvement Collaborative (NIC/Q 2002). Collaboratives such as this, which are intended to

facilitate improvement by facilitating the transfer of knowledge across health care organizations

(Kilo, 1999), are a growing phenomenon in health care (Mittman, 2004). In the one we studied,

cross-disciplinary teams were encouraged to work together on speciﬁed improvement areas (e.g.,

infection control, respiratory care management, or discharge planning) to develop and test clinical,

organizational, and operational changes for improving neonatal care. As such, this setting was

particularly attractive for our study of how status relationships that exist between professional

disciplines relate to psychological safety and engagement in quality improvement in cross-

disciplinary teams. To craft and execute their quality improvement projects in-house, team members

from different disciplines had to work together. Development of a new or better delivery room

plan for infant resuscitation, for example, required physicians, nurses, and respiratory therapists

to share their expertise and recommendations with one another to ensure that the developed

plan included accurate and up-to-date practices from each group, and accounted for their

interactions.

Methods

Sample and procedure

We conducted data collection in three phases. First, we visited 4 of the 44 neonatal intensive care

units (NICUs) in the collaborative. We toured the units, observed unit functioning and interviewed

DOI: 10.1002/job

950 I. M. NEMBHARD AND A. C. EDMONDSON

23 staff members (5–7 members per NICU) for 30 to 90 minutes using open-ended questions about

the NICU work climate and quality improvement efforts. We selected the four NICUs to include

differences in demographic variables (e.g., teaching status, size, etc.) and improvement efforts

(e.g., improvement focus areas and prior collaborative participation) and selected interviewees

to capture multiple professional groups (i.e., physicians, nurses, and respiratory therapists).

This increased our understanding of NICUs and facilitated the design of a meaningful survey

for this population. In the second phase, we developed and pilot tested a survey with these four

NICUs. Descriptive statistics and psychometric tests indicated no need to alter the survey. Further,

the survey results accorded with our site visit observations, supporting the validity of the

instrument.

In the third and ﬁnal phase, we invited the 40 remaining, non-pilot site NICUs in the

collaborative to participate in the survey via phone and electronic mail directed to the leader of

each NICU’s collaborative improvement team. Twenty-three NICUs agreed, for a NICU response

rate of 58%. When we compared participating sites to non-participating sites, we found no

signiﬁcant differences between the two groups on avariety of structural, clinical, and patient acuity

measures. For example, there were no signiﬁcant differences based on hospital ownership type

(not-for-proﬁt, for-proﬁt, government), teaching status, level of severity of the care provided in the

NICU, volume of extremely low birth weight babies (ELBW, less than or equal to 1000 g), number

of times the site participated in prior collaboratives (0, 1, or 2), length of stay, percentage of ELBW

babies with Apgar scores 3 one minute after birth, percentage of babies transported from another

hospital (‘‘outborn’’), average birth weight, and gestational age (see Tucker, Nembhard, &

Edmondson, 2006).

As soon as a NICU received the permission of its hospital Institutional Review Board (IRB), its

team leader provided us with a count of the number of staff and patient beds in the unit. We then

invited all team members in the NICU to participate in our survey via an invitational letter distributed

in accordance with the procedures outlined in the hospital IRB approval. In 18 NICUs, the

collaborative team leader electronically mailed or placed in team mailboxes our letter to team

members, which provided instructions on where to retrieve a paper copy of the survey in the unit, if

interested. In the remaining ﬁve NICUs, the team leader distributed our letter along with a paper copy

of the survey to team mailboxes. The invitational letter described the purpose of the study, assured

respondents’ anonymity and included instructions for completion of the paper as well as an online

version of the survey. We offered two versions to allow participants to choose their preferred format.

All individuals returned the survey directly to us via self-addressed stamped envelope or online

submission.

In all, 1440 health care professionals (46% of team members contacted) from 23 NICUs in the

United States and Canada completed the survey between July 2003 and May 2004. Of the

respondents, 1375 persons declared their profession: 100 as physicians (83 neonatologists, 13

attending physicians, and 4 neonatology fellows), 998 as nurses (65 neonatal nurse practitioners, 16

Master’s prepared nurses, 8 clinical nurses specialists, 867 registered nurses, 14 licensed

nurse practitioners, and 28 other nursing functions), 131 as respiratory therapists, and 146 as

other types of health care professionals (e.g., social worker, dietitians, etc.). Excluding individuals in

the latter group (because the presence and recognition of their discipline as part of the NICU

team varied across units), NICUs averaged 60 respondents, ranging from 10 to 164. The

average number of respondents per NICU was 3 physicians, 15 nurses, and 5 respiratory

therapists.

We regarded the professionals in each NICU from medicine, nursing, and respiratory therapy as the

‘‘unit team’’ because these interdependent professionals unquestionably work together in the unit on a

daily basis. Their patients generally stay in the NICU for an extended period of time, often months, to

DOI: 10.1002/job

STATUS AND PSYCHOLOGICAL SAFETY 951

receive the care that only a team of neonatal-trained practitioners can deliver. These teams and

members are so specialized that they work exclusively in the NICU.

Measures

Professional status

Professional status was inferred from occupational category because prior research had demonstrated

that status in the medical ﬁeld derives from role or position in the professional hierarchy (Friedson,

1970a,b, 1994; Mechanic, 1991; Wolinsky, 1988). The numerous professional roles reported in the

survey were combined into three major categories for the purposes of these analyses: (1) physicians

(neonatologists, other attending physicians and neonatal fellows); (2) nurses; and (3) respiratory

therapists. Further, because physicians—as ‘‘captains of the health care team’’ (Fuchs, 1974)— garner

higher status than all other health care professionals (Hafferty & Light, 1995), an additional

dichotomous professional status variable was created to reﬂect that (0 ¼low-status ¼nurses and

respiratory therapists, 1 ¼high status ¼physicians).

Unit team membership

Unit team membership, a categorical variable, refers to the speciﬁc NICU in which an individual

respondent worked (N¼23).

Psychological safety

Four items from Edmondson’s (1999) psychological safety scale adapted to this context were used to

assess the extent to which respondents felt safe to speak up about issues or ideas regarding the NICU.

Examples of these items are: ‘‘Members of this NICU are able to bring up problems and tough issues’’

and ‘‘People in this unit are comfortable checking with each other if they have questions about the right

way to do something.’’ Respondents’ agreement (1 ¼strongly disagree, 7 ¼strongly agree) with these

items formed a single scale (Chronbach alpha ¼0.73).

Leader inclusiveness

In the hospital context, physicians are both high-status technical experts and leaders who are

responsible for directing the efforts of others in delivering care to patients. Thus, although the construct

of leader inclusiveness is more general, in this setting, it refers to behaviors and attitudes of the

physicians-in-charge. A three-item scale assessed the extent to which NICU leaders’ words and deeds

indicated an invitation and appreciation for others as contributing members in a team endeavor. The

ﬁrst two items, ‘‘NICU physician leadership encourages nurses to take initiative’’ and ‘‘Physicians ask

for the input of team members that belong to other professional groups,’’ were adapted from Shortell

et al.’s (1991) physician leadership scale. The third item, ‘‘Physicians do not value the opinion of others

equally’’ (reversed scored), was developed by us for this research. The level of agreement with each

statement (1 ¼strongly disagree, 7 ¼strong agree) was averaged to provide a single perception for

each respondent (Chronbach alpha ¼0.75).

In this study, we assessed leader inclusiveness with data from non-physicians only, for two reasons.

First, prior research suggested that nurses provide a more accurate assessment of organizational culture

There is one possible exception to the exclusivity of the NICU teams. Some hospitals do use ‘‘rotating’’ respiratory therapists,

who work exclusively on the unit for a deﬁned period of time periodically (e.g., 2 months every 2 months), then rotate to other

units when not assigned to the NICU. That said, we believe that ‘‘unit teams’’ are accurately represented in our sample since we

asked team leaders to distribute the survey to NICU staff. We abided by team leader judgment of team membership.

DOI: 10.1002/job

952 I. M. NEMBHARD AND A. C. EDMONDSON

and leadership practices than physicians. One study showed that while nurses’ reports of the culture

showed appropriate variance, physicians’ reports were more uniform across different contexts, as well

as more positive; further, only the nurse data predicted performance outcomes (Leonard, Frankel,

Simmonds, & Vega, 2004), afﬁrming earlier studies’ ﬁndings that nurses’ perceptions of the work

environment are more predictive of the risk of adverse outcomes (e.g., Baggs et al., 1999) and quality of

care (e.g., Shortell et al., 1991) than physicians. Second, and more important, conceptually, those with

low status are in a better position to assess the degree to which high-status leaders are including them

than are those with high status— who may rate themselves as inclusive even when others would not do

so. More simply, physician ratings of their own inclusiveness toward others are unlikely to be as

externally valid as the ratings of those others will be.

Engagement in quality improvement work

Engagement in quality improvement work was measured using a four-item scale adapted from Baker,

King, MacDonald, & Horbar (2003). Sample items are: ‘‘A growing number of staff in this NICU are

participating in improvement efforts’’ and ‘‘In the coming year, I would like to be very involved in our

NICU’s quality improvement efforts.’’ Respondents rated their agreement with these statements on a

seven-point scale (1 ¼strongly disagree to 7 ¼strongly agree). Chronbach’s alpha was 0.79.

Control variables

We included gender, years working in any NICU, years as an employee of the hospital, years working in

the current NICU, and hours per week in the NICU as control variables since these demographic

variables are potential predictors of psychological safety at the individual level. These control variables

were excluded from the group-level analyses performed to test psychological safety as a mediator

between leader inclusiveness and engagement in quality improvement work (H6).

Table 1 provides the means, standard deviations, and intercorrelations between the variables. We

assessed the adequacy of our survey measures through tests of internal consistency reliability and

discriminant validity. The results supported the use of these measures for analysis. Chronbach’s alpha

for all survey scales exceeded the 0.70 threshold proposed by Nunnally (1978) for internal consistency,

and factor analysis, in which the planned constructs emerged, conﬁrmed discriminant validity.

Analysis and Results

To test Hypotheses 1and 2, that status and its interaction with unit team membership explain variance

in psychological safety; we used a univariate general linear model (GLM) with professional status

(three groups: physicians, nurses, and respiratory therapists) and unit team membership as independent

variables and the reﬂected, logarithmic transformation of psychological safety as the dependent

variable.

We also included the above mentioned control variables. Results shown in the upper portion

of Table 2 indicate that two of the ﬁve control variables were signiﬁcant and that our ﬁrst two

hypotheses were supported. The number of years spent working in any NICU and the number of years

The reﬂected, logarithmic transformation corrected for heterogeneity of variance and non-normality (i.e., negative skewness) in

the dataset (Tabachnick & Fidell, 2001).

DOI: 10.1002/job

STATUS AND PSYCHOLOGICAL SAFETY 953

Table 1. Summary statistics and intercorrelations for individual-level data

Variable Mean SD 1 2 3 4 5 6 7 8 9

1. Professional status—3 groups:

Physicians versus

nurses versus therapists

2.03 0.43

2. Professional status—2 groups:

High versus

low status

0.08 0.27 0.71



3. Psychological safety 5.31 1.08 0.15



0.14



4. Leader inclusiveness 4.62 0.55 <0.01 0.01 0.29



5. Engagement in quality improvement work 5.45 1.06 0.12



0.17



0.48



0.21



6. Gender 1.10 0.30 0.23



0.56



0.05 0.04 0.02

7. Years in any NICU

4.17 1.19 0.08



0.10



0.05 0.09



0.03 0.08



8. Years as hospital employee

3.97 1.26 0.03 0.01 0.04 0.02 0.02 0.07



0.73



9. Years in current NICU

3.84 1.31 0.05 0.03 0.04 0.02 0.02 0.08



0.81



0.94



10. Hours per week in NICU 33.84 13.46 0.35



0.38



0.05 0.03 0.10



0.12



0.06



0.08



0.08



This is a categorical variable: 1 ¼less than 1 year, 2 ¼1 to less than 2 years, 3 ¼2 to less than 5 years, 4 ¼5 to less than 10 years, 5 ¼10 or more years.



p<0.05.



p<0.01.

DOI: 10.1002/job

954 I. M. NEMBHARD AND A. C. EDMONDSON

spent working in the current NICU signiﬁcantly predicted psychological safety (F(4,1111) ¼3.54,

p¼0.01 and F(4,1111) ¼3.09, p¼0.02): More time spent in NICUs in general as well as an

intermediate amount of time spent in the current NICU (5 to less than 10 years) both were associated

with higher psychological safety. The number of years as an employee of the current hospital was only

marginally predictive (F(4,1111) ¼2.09, p¼0.08), and gender and hours worked per week not at all

(F(1,1111) ¼0.01, p¼0.93 and F(1, 1111) ¼0.24, p¼0.62).

Consistent with Hypothesis 1, professional status was positively associated with psychological

safety (F(2,67) ¼8.46; p¼0.001). Physicians felt signiﬁcantly more psychological safety than

nurses (t(1111) ¼0.33, p<0.001) who in turn reported more psychological safety than

respiratory therapists (t(1111) ¼0.11, p¼0.02); the overall planned contrast was signiﬁcant

(F(2, 1111)¼14.11, p<0.001).

To determine whether the effect of professional status on psychological safety varied across

multidisciplinary teams, as predicted in Hypothesis 2, we examined the interaction term composed of

professional status and unit team membership. The interaction was signiﬁcant (F(31,1111) ¼2.22,

p<0.001), indicating that the difference in psychological safety felt across status groups varied across

teams, even as the main effect of status on psychological safety remained signiﬁcant. In some teams,

status differences were less important, such that members of different professional groups felt similarly

safe. In others, professional status differences led to larger disparities in psychological safety among

groups.

Hypotheses 3 and 4 that leader inclusiveness predicts psychological safety, and moderates the

relationship between status and psychological safety, were tested using GLM analyses. First, however,

we assessed the properties of leader inclusiveness, as a new construct. Just as prior work (Edmondson,

1999; Shamir, Zakay, Breinin, & Popper, 1998) has shown convergent perceptions of leadership

behavior in teams, we anticipated that views of leader inclusiveness would be shared in the NICUs

studied. To test this, we performed a one-way analysis of variance with unit team membership as the

independent variable and leader inclusiveness (rated by non-physicians only) as the dependent variable.

The ANOVA results showed signiﬁcant variance at the group level of analysis for leader inclusiveness

as reported by low-status individuals (F(22,1095) ¼9.44, p<0.001), and an intraclass correlation that

Table 2. Results of general linear model analysis

Model Independent variables F-ratio p

Y¼Ln(Psychological safety) Gender F(1,1111) ¼0.01 0.93

Years in any NICU F(4,1111) ¼3.54 0.01

Years as a hospital employee F(4,1111) ¼2.09 0.08

Years in current NICU F(4,1111) ¼3.09 0.02

Hours per week in NICU F(1,1111) ¼0.24 0.62

Professional status—three groups F(2,67) ¼8.46 0.001

Unit team membership F(22,41) ¼2.08 0.02

Professional status (3) unit team membership F(31,1111) ¼2.22 <0.001

Y¼Ln(Psychological safety) Gender F(1,1126) ¼0.10 0.76

Years in any NICU F(4,1126) ¼3.98 0.003

Years as a hospital employee F(4,1126) ¼1.66 0.16

Years in current NICU F(4,1126) ¼2.43 0.05

Hours per week in NICU F(1,1126) ¼0.001 0.97

Professional status—two groups F(1,1126) ¼11.58 0.001

Leader inclusiveness F(21,1126) ¼4.27 <0.001

Professional status (2) leader inclusiveness F(18,1126) ¼1.78 0.02

DOI: 10.1002/job

STATUS AND PSYCHOLOGICAL SAFETY 955

was positive and signiﬁcant (r

ICC

¼0.35), conﬁrming leader inclusiveness as a group-level measure

(Kenny & LaVoie, 1985). We therefore entered the team’s leader inclusiveness score (the team average)

as the individual’s score for each member of the team in the individual-level dataset.

We used this dataset and a GLM to test Hypotheses 3 and 4 that leader inclusiveness is positively

associated with psychological safety and moderates the relationship between professional status and

psychological safety. The dichotomous professional status variable (high status, if physician and low

status, if nurse or respiratory therapist), leader inclusiveness, and their interaction served as

independent variables, along with all of the control variables. Again, the reﬂected, logarithmically

transformed psychological safety measure served as the dependent variable. The results shown in the

lower section of Table 2 support the hypothesized relationships. When physician leaders were

perceived as inclusive and welcoming of others’ ideas and efforts, psychological safety was greater

(F(21,1126) ¼4.27, p<0.001) (Hypothesis 3). Also, as predicted in Hypothesis 4, leader inclusiveness

moderated the relationship between status and psychological safety, as indicated by a signiﬁcant

interaction term (F(18, 1126) ¼1.78, p¼0.02). Low leader inclusiveness was associated with a greater

disparity in psychological safety between high and low-status individuals to the disadvantage of low-

status individuals. In contrast, high leader inclusiveness was associated with a lower difference in

psychological safety between the two groups, raising the psychological safety of low-status individuals

closer to that of their high-status team members. (See Figure 2).

To test Hypothesis 5 that psychological safety is positively associated with engagement in quality

improvement work, we conducted a regression analysis on the group level dataset. Using this dataset,

allowed us to not only test our hypothesis, but also as important, examine the relative inﬂuence of a

competing group-level variable on engagement: staff workload or busyness of the work environment.

Prior work has identiﬁed staff workload as a predictor of quality-related behavior (Oliva, 2001;

Tucker & Edmondson, 2003). Overburdened employees facing competing priorities are less likely to

further quality goals. Therefore, to continue with our model of engagement in quality improvement

Figure 2. Psychological safety as a function of professional status and leader inclusiveness

DOI: 10.1002/job

956 I. M. NEMBHARD AND A. C. EDMONDSON

work, the question of relative inﬂuence had to be addressed to provide assurance that our model was not

mis-speciﬁed in emphasizing psychological safety as a determinant of engagement. We ﬁrst performed

one-way analysis of variance and intraclass correlations for psychological safety (F(22, 1366) ¼8.62,

p<0.001; r

ICC

¼0.21) and engagement in quality improvement work (F(22, 1352) ¼7.67, p<0.001;

ICC

¼0.18) to conﬁrm the group level status of the variables before proceeding with the analysis. We

then regressed engagement on psychological safety and the ratio of staff-to-patient beds, (mean ¼3.31,

SD ¼0.91), our measure of staff workload. The results provided support for our hypothesis that the

more members feel they work in a team characterized by interpersonal trust and respect, the more

enthusiastic and devoted they are to participating in quality improvement efforts, which often require

the interpersonally risky act of collaborative learning (B¼0.62, p¼0.001). We found this relationship,

even after accounting for the competing inﬂuence of staff busyness, which was an insigniﬁcant

predictor of engagement (B¼0.14, p¼0.09) and thus eliminated from further analyses.

To test Hypothesis 6, that psychological safety mediates the relationship between leader inclusiveness

and engagement in quality improvement work, we estimated three regression equations to satisfy the

conditionsfor mediation (Baron & Kenny, 1986). The ﬁrst condition is that the independent variable must

signiﬁcantly affect the mediator; second, the independent variable must predict the dependent variable;

and third, the mediator must signiﬁcantly predict the dependent variable when entered in the same

regression equation as the independent variable, while the independent variable drops in signiﬁcance or

becomes insigniﬁcant. We assessed whether these conditions were satisﬁed using group-level data

because leader inclusiveness (our independent variable), psychological safety (our proposed mediator),

and engagement in quality improvement work (our dependent variable) are conceptually meaningful and

empirically demonstrated as group-level variables. We measured leader inclusiveness at the aggregate

non-physician level for reasons explained above, and analyses related to testing Hypothesis 5 conﬁrmed

psychological safety and engagement as group-level constructs as well.

As shown in Table 3, the results support all three mediation conditions: (1) leader inclusiveness

predicts psychological safety (B¼0.53, p<0.001), (2) leadership inclusiveness predicts team

engagement at the group level (B¼0.41, p¼0.004), and (3) when included in the same model, leader

inclusiveness becomes insigniﬁcant (B¼0.11, p¼0.52), while psychological safety remains

signiﬁcant (B¼0.57, p¼0.03). Thus, the data support psychological safety as a mediator of the

relationship between leader inclusiveness and team engagement in quality improvement work.

In sum, all of our hypotheses were supported by the data. Not only do high-status individuals differ

from low-status individuals in psychological safety, but also there is an interaction between status and

unit team membership, and between status and leader inclusiveness. Greater inclusiveness minimizes

the effect of status on psychological safety in the team, and vice versa. Moreover, with more leader

Table 3. Tests of psychological safety as a mediator between leader inclusiveness and engagement in quality

improvement work (N¼23)

Conditions to demonstrate mediation



Independent variable B tpR

1. Does leader inclusiveness predict

engagement in quality improvement work?

Leader inclusiveness 0.41 3.19 0.004 0.33

2. Does leader inclusiveness

predict psychological safety?

Leader inclusiveness 0.53 5.02 <0.001 0.55

3. Does the effect of leader inclusiveness

drop substantially or become insigniﬁcant when

psychological safety (the mediator) is included in

the model for engagement in quality improvement work?

Psychological safety 0.57 2.32 0.03 0.47

Leader inclusiveness 0.11 0.65 0.52



Dependent variables are in italics.

DOI: 10.1002/job

STATUS AND PSYCHOLOGICAL SAFETY 957

inclusiveness comes greater psychological safety, which in turn predicts greater team engagement in

quality improvement work.

Discussion and Conclusions

Toward a theory of role-based status and psychological safety in teams

This paper empirically examined the effect of professional status on psychological safety in health

care teams, motivated both by the theoretical role of psychological safety in team learning

(Edmondson, 1999) and by the practical need for team learning in today’s hospitals. As anticipated,

psychological safety was signiﬁcantly associated with professional status in these data. The results thus

suggest that, in health care, professional status inﬂuences beliefs about how easy or appropriate it is to

speak up to offer ideas, raise concerns, or ask questions. At the same time, our results showed

signiﬁcant differences across groups in the strength of the association between status and psychological

safety, suggesting that this relationship need not be deterministic. In some teams, members with high

(i.e., physicians) and low (i.e., nurses and respiratory therapists) status reported more similar levels of

safety than in other teams, in which relatively large disparities in psychological safety between

professional groups were found. This ﬁnding provides reassurance that the effects of traditional status

differences in health care may be overcome, facilitating full participation in cross-disciplinary team

improvement efforts.

Unlike in other industries, where employees may advance in the hierarchy—moving from sales

assistant to sales associate or from analyst to consultant to principal, for example — medical

professionals cannot rely on professional mobility to confer greater status. Professions tend to be stable

over a career. Nurses and respiratory therapists generally do not become doctors. As a result, the

opportunity for natural status gains by lower status individuals can be rare. With history and industry

structure as perpetuators, profession-related status differences continue. And, the growing

interdependence among professions only heightens intergroup tensions. According to social

psychological (Alderfer, 1977; Alderfer & Smith, 1982; Messick & Mackie, 1989) and ethnographic

(Dougherty, 1992) research, this is natural.

Our results suggest leader inclusiveness—words and deeds by leaders that invite and

appreciate others’ contributions— can take nature off its course, helping to overcome status’

inhibiting effects on psychological safety. In cross-disciplinary teams with high leader

inclusiveness, the status–psychological safety relationship was weakened. We thus suggest

that active, inclusive behavior on the part of physician leaders may be an essential means of

facilitating others’ meaningful engagement in team-based quality improvement work. Thus, a

key contribution of this study is the identiﬁcation of an interpersonal strategy for improving the

climate for learning within cross-disciplinary teams in health care. This extends prior work that has

shown psychological safety consistently to be related to leader behavior (Edmondson, 1999, 2003;

Hult, Hurley, Guinipero, & Nichols, 2000; Lovelace,Shapiro,&Weingart,2001;Norrgren&

Schaller, 1999), by narrowing and sharpening the nature of leader behaviors that promote

psychological safety.

Recent studies of psychological safety and communication in the health care environment

have also highlighted the role of the leadership in cultivating a culture of safety, but have

not articulated the actual practices of leaders that are needed, other than training staff to speak

DOI: 10.1002/job

958 I. M. NEMBHARD AND A. C. EDMONDSON

up (Leonard et al., 2004; Maxﬁeld, Grenny, McMillan, Patterson, & Switzler, 2005). Our

research suggests that training leaders to invite team members’ comments and to appreciate

those comments overtly is as important. As we did not test any training methods, we cannot

draw conclusions about what training strategies will work. However, training that includes

instruction in the timing and phrasing of invitation and appreciation may be useful. Boyle and

Kochinda (2004) tested a training intervention to improve communication between nurse and

physician leaders,

and found that staff perceptions of leadership, communication, and problem-

solving between professional groups improved in the post-intervention period, afﬁrming that

leadership training is a promising route to improved team climate. Our results also suggest it is a

route to increasing engagement in quality improvement.

A more nuanced look at team learning for health care improvement efforts

The study also provides insight into a more focused type of team learning than has been investigated in

prior-related research: team learning for quality improvement. First, we found evidence that

psychological safety, in combination with leader inclusiveness, promotes team engagement in quality

improvement work. Second, the ﬁnding that psychological safety mediates between leader

inclusiveness and engagement in quality improvement extends prior work that showed psychological

safety to be a mediator between the broader construct of team leader coaching behavior and team

learning behavior, more generally (Edmondson, 1999). Finally, our model and ﬁndings shed light on

the important but rarely studied phenomenon of engagement in discretionary, effortful quality

improvement activities. Although our emphasis was on the effects of psychological safety, we also

acknowledge that there may be other processes that contribute to engagement (e.g., psychological

availability (Kahn, 1990)) that we have not addressed. Future research should explore additional

mechanisms by which teams become actively engaged in quality improvement work.

The challenge of engaging busy, overworked professionals and service workers —already burdened

by the heavy demands of their ‘‘real jobs’’— in quality improvement activities should not be

underestimated. In the course of a day, health care professionals are continuously called upon to

immediately address the many needs of patients and co-workers. The requests are so many that nurses

often sacriﬁce their personal breaks (e.g., lunchtime) and work overtime without compensation in order

to complete their daily clinical care tasks (Tucker & Edmondson, 2003). The time pressure to complete

tasks is so great that the most burdened frequently neglect to wash their hands, despite almost universal

awareness that hand hygiene is a cornerstone of good clinical practice (Institute of Medicine, 2003).

In this challenging work environment, competing demands could surely relegate the perceived extra-

role work of participating in quality improvement efforts to the background unless engagement ignites

a counter response. Our analyses —using staff-to-patient beds as a measure of workload—suggest that

the engagement needed to motivate such efforts is related more to psychological safety within the unit

team than the burden created by competing demands and workload. This is good news for health care

because it suggests that improvement efforts need not be delayed until greater slack is created; rather

engagement in improvement efforts may occur in unit teams once a climate of psychological safety for

The ‘‘collaborative communication intervention’’ employed by Boyle & Kochinda (2004) consisted of multiple learning

activities, small group skill practice and problem solving decisions; feedback and reinforcement of newly learned skills; a

planning assignment for on-the-job applications; and assessment and feedback after the intervention. Participating ICU leaders

spent a total of 23.5 hours learning communication and process skills (e.g., how to open a conversation, how to seek ideas from all

involved, etc.).

DOI: 10.1002/job

STATUS AND PSYCHOLOGICAL SAFETY 959

learning is cultivated. Thus, another contribution of the present study is to illuminate organizational

antecedents of engagement, which can be a difﬁcult state to elicit naturally.

Limitations

This study is not without limitations. First, our use of a sample consisting exclusively of NICUs and

their professionals may limit the generalizability of the ﬁndings. While NICUs are similar to other

health care units in exemplifying the trend toward cross-disciplinary teams mentioned at the outset

(associated with increasing knowledge, specialization, and interdependence), one could argue that

these trends are heightened in NICUs, making the setting unusual. Moreover, the fragility of the (infant)

patient and the importance of timely communication and action are also special characteristics of the

NICU. These contextual variables may foster an appreciation for teamwork and recognition of each

discipline’s contribution such that leader inclusiveness is more prevalent in NICUs than in other health

care settings, and more likely to elicit responsiveness. Other unit types (e.g., outpatient units) may be

slower to realize the unique contributions of traditionally lower status team members and the

interdependency among disciplines. Therefore, status differences may be greater in those unit teams

compared to those in the NICU environment, and leadership’s achievement of inclusiveness may be

more difﬁcult. Our data do not allow us to make a comparison based on unit type. However, the

presence of signiﬁcant variance between NICUs with respect to leader inclusiveness suggests that

leader inclusiveness is not a universal NICU attribute. Some NICU leaders experience signiﬁcant

difﬁculty behaving inclusively. Whether this relationship varies across unit types will have to be

investigated in future work.

Future work should also address imperfections of our data. We received completed surveys from

only 58% of the NICUs in the collaborative and only 46% of unit team members contacted (excluding

survey pilot sites). Although demographic comparison of participating and non-participating NICUs

showed no differences between the two groups and our response rates are similar to other studies in

hospitals (Edmondson, 1996; Kaissi, Kralewski, Curoe, Dowd, & Silversmith, 2004; Jacob &

Deshpande, 1997), we cannot ignore the possibility of non-response bias at the group and individual

levels. The latter is a most serious concern if individual non-respondents are those who did not feel

psychologically safe and included. The absence of their responses biases their unit data upward,

incorrectly giving the appearance of units that have managed to minimized status differences in

psychological safety and that are inclusive. Subsequent studies with higher response rates are needed to

address this concern. Additionally, those studies are advised to examine a relationship we were unable

to test with our dataset—whether specialization among nurses is associated with differences in

psychological safety. Nurse respondents did indicate their professional training (e.g., neonatal nurse

practitioner vs. Master’s prepared nurse) in their survey responses; however this was not enough

information to accurately identify different specialties within the nursing populations across units. We

urge scholars in this area to collect information about daily tasks and speciﬁc care delivery roles to

further assess the effects of specialization and status within a professional group on psychological

safety.

Finally, we must acknowledge that team engagement in quality improvement work is an

intermediary outcome. The ultimate goal of health professionals involved in quality improvement work

is to close the gap between current and evidence-based practice to produce the highest quality care

possible in a consistent manner (Shojania, McDonald, Wachter, & Owens, 2004). Although we lack

data in this study to test the relationship between engagement in quality improvement work and team

learning as evidenced by quality improvement outcomes, past research shows a positive relationship

between employee engagement and other organizational outcomes, including customer satisfaction,

DOI: 10.1002/job

960 I. M. NEMBHARD AND A. C. EDMONDSON

productivity, proﬁt, and safety (Buckingham & Coffman, 1999), and future research should further test

this relationship.

Practical implications

In sum, this study extends a stream of research on creating psychological safety for learning within

cross-disciplinary teams, and begins the discussion about team engagement in quality improvement

work. At a time when much attention is centered on how to advance quality improvement efforts in

health care, this paper offers initial theoretical and practical insight. Speciﬁcally, training leaders to be

inclusive to foster psychological safety may be a critical antecedent of effective quality improvement,

because it may create the engagement that is necessary for teams to participate in the speciﬁc learning

activities required for quality improvement. We ﬁnd, in related work, that psychological safety enables

the performance of ‘‘learn-how’’ activities— experimental activities such as pilot projects, dry runs,

and problem-solving cycles —which are interpersonally risky yet often required for the

implementation success of quality improvement projects in health care (Tucker et al., 2006).

Together, these ﬁndings imply a focus on team leader behavior to create the conditions (i.e.,

psychological safety) and then the opportunities (i.e., learn-how activities) for team quality

improvement efforts.

We do not intend to suggest that the development of inclusive leaders within NICUs would affect

psychological safety or engagement in other parts of hospital systems. NICU leaders rarely interact

with other clinical units, thus the opportunities for their inclusive nature to shape the climate of

other multidisciplinary care delivery teams directly is limited. Likewise, NICU team members

typically do not work on other care teams, so any habit of collaborative interactions is mostly NICU-

contained.

Although this research took place in the health care context, the ﬁndings may apply to other

organizational contexts as well. In particular, in other organizations with cross-disciplinary teams,

status diversity and a need for teams to continuously improve the services or products the organization

produces, the model developed in this paper may be highly relevant. A growing number of industries

are characterized by these features, as organizations realize that individuals in different functions,

locations, and stations of life possess specialized knowledge that can be valuable for problem-solving

and innovation, if shared and combined with the knowledge of other employees through teamwork.

Therefore, our model may have wider relevance than just health care, but health care, with its

prevalence of cross-disciplinary teams, salient status differences, and many quality improvement

efforts is particularly conducive for exploring and developing these ideas.

Acknowledgements

This research beneﬁted greatly from the expert help of health care management and operations

researchers Anita L. Tucker and Dr. Richard Bohmer (also a physician), as well as the superb research

assistance of Laura Feldman. We thank the participants in the Leadership and Groups seminar at

Harvard University, Denise Rousseau and two anonymous reviewers for constructive comments on

earlier versions of this manuscript. We also appreciate the invitation to conduct this research provided

by Dr. Jeffrey Horbar, Kathy Leahy, Paul Plsek and the Vermont Oxford Network, as well as the

willingness of nurses, physicians, respiratory therapists and other staff members of the participating

DOI: 10.1002/job

STATUS AND PSYCHOLOGICAL SAFETY 961

hospitals to engage in this research. Their time, effort, and honesty provided an invaluable contribution,

for which we are deeply appreciative. The Harvard Business School Division of Research provided

ﬁnancial support.

Author biographies

Ingrid M. Nembhard is a doctoral candidate in the Ph.D. Program in Health Policy (Management

Concentration) at Harvard University. She received her S.M. in Health Policy and Management from

the Harvard School of Public Health. Her research interests include intra- and inter-organizational

learning, collaborative teams, and quality improvement in health care.

Amy C. Edmondson is Novartis Professor of Leadership and Management at Harvard University. She

received her A.M. in Psychology and Ph.D. in Organizational Behavior from Harvard. Her research

focuses on understanding and improving processes through which organizations learn and innovate, in

health care and other industry settings.

References

Accreditation Council for Graduate Medical Education. (2004). 2002–2003 Annual Report. Retrieved December

14, 2004, from http://www.acgme.org/acWebsite/about/ab_2002–03AnnRep.pdf.

Aiken, L. H., & Sloane, D. M. (1997). Effects of specialization and client differentiation on the status of nurses:

The case of AIDS. Journal of Health and Social Behavior, 38, 203–222.

Alderfer, C. P. (1977). Group and intergroup relations. In J. R. Hackman, & J. L. Suttle (Eds.), Improving life at

work: Behavioral science approaches to organizational change (pp. 227–296). Santa Monica, CA: Goodyear.

Alderfer, C. P. (1987). An intergroup perspective on organizational behavior. In J. W. Lorsch (Ed.), Handbook of

organizational behavior (pp.190–222). Englewood Cliffs, NJ: Prentice-Hall.

Alderfer, C. P., & Smith, K. K. (1982). Studying intergroup relations embedded in organizations. Administrative

Science Quarterly, 27, 35–65.

Anderson, C., John, O. P., Keltner, D., & Kring, A. M. (2001). Who attains social status? Effects of personality and

physical attractiveness in social groups. Journal of Personality and Social Psychology, 81, 116–132.

Argyris, C. (1985). Strategy, change, and defensive routines. Boston, MA: Pitman.

Argyris, C., & Scho

¨n, D. A. (1978). Organizational learning: A theory of action perspective. Reading, MA:

Addison-Wesley.

Ashford, S. J., Rothbard, N. P., Piderit, S. K., & Dutton, J. E. (1998). Out on a limb: The role of context and

impression management in selling gender-equity issues. Administrative Science Quarterly, 43, 23–57.

Bacharach, S. B., Bamberger, P., & Mundell, B. (1993). Status inconsistency in organizations: From social

hierarchy to stress. Journal of Organizational Behavior, 14, 21–36.

Baggs, J. G., Schmitt, M. H., Mushlin, A. I., Mitchell, P. H., Eldredge, D. H., & Oakes, D., et al. (1999).

Association between nurse-physician collaboration and patient outcomes in three intensive care units. Critical

Care Medicine, 27, 1991–1998.

Baker, R. G., King, H., MacDonald, J. L., & Horbar, J. D. (2003). Using organizational assessment surveys for

improvement in neonatal intensive care. Pediatrics, 111, e419–e425.

Baker, R. G., Murray, M., & Tasa, K. (1995, December). Quality in action: An instrument for assessing

organizational culture for quality improvement. Paper presented at the First International Scientiﬁc Symposium

on Improving Quality and Value in Health Care, Orlando, FL.

Baron, R.A. (1990). Countering the effects of destructive criticism: The relative efﬁcacy of four interventions.

Journal of Applied Psychology, 73, 199–207.

DOI: 10.1002/job

962 I. M. NEMBHARD AND A. C. EDMONDSON

Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological

research: Conceptual, strategic and statistical considerations. Journal of Personality and Social Psychology, 51,

1173–1183.

Bass, B. M. (1990). Bass and Stogdill’s handbook of leadership. New York: Free Press.

Benoit-Smullyan, E. (1944). Status, status types, and status interrelations. American Sociological Review, 9, 151–

161.

Berger, J., Cohen, B. P., & Zelditch, M. (1972). Status characteristics and social interaction. American Sociological

Review, 8, 241–255.

Berger, J., Fisek, H., Norman, R. Z., & Zelditch, M. (1985). The formation of reward expectations in status

situations. In J. Berger, & M. Zelditch (Eds.), Status, rewards, and inﬂuence. San Francisco, CA: Jossey-Bass.

Berger, J., Rosenholtz, S. J., & Zelditch, M. (1980). Status organizing processes. In A. Inkeles, N. J. Smelser, & R.

H. Turner (Eds.), Annual review of sociology (Vol. 6, pp. 479–508). Palo Alto: Annual Reviews.

Berwick, D. M., Godfrey, A. B., & Roessner, J. (1990). Curing health care: New strategies for quality

improvement: A report on the National Demonstration Project on Quality Improvement in Health Care.

San Francisco, CA: Jossey-Bass.

Boyle, D. K., & Kochinda, C. (2004). Enhancing collaborative communication of nurse and physician leadership in

two intensive care units. Journal of Nursing Administration, 34, 60–70.

Brown, P., & Levinson, S. C. (1987). Politeness: Some universals in language usage. New York: Cambridge

University Press.

Brown, M. S., Ohlinger, J., Rusk, C., Delmore, P., & Ittmann, P., on behalf of the CARE Group. (2003).

Implementing potentially better practices for multidisciplinary team building: Creating a neonatal intensive care

unit culture of collaboration. Pediatrics, 111 (Supplement), e482–e488.

Buckingham, M., & Coffman, C. (1999). First, break all the rules. New York: Simon and Schuster.

Bunderson, J. S. (2003a). Recognizing and utilizing expertise in work groups: A status characteristics perspective.

Administrative Science Quarterly, 48, 557–591.

Bunderson, J. S. (2003b). Team member functional background and involvement in management teams: Direct

effects and the moderating role of power centralization. Academy of Management Journal, 46, 458–474.

Chassin, M. R. (1998). Is health care ready for six sigma quality? Milbank Quarterly, 76, 575–591.

Chassin, M. R., Galvin, R. W., & The National Roundtable on Health Care Quality. (1998). The urgent need to

improve health care quality. Journal of the American Medical Association, 280, 1000–1005.

Coburn, D. (1992). Friedson then and now: An ‘internalist’ critique of Friedson’s past and present views of the

medical profession. International Journal of Health Services, 22, 497–512.

DeSantis, G. (1980). Realms of expertise: A view from within the medical profession. In J. A. Roth (Ed.), Research

in the sociology of health (pp.179–236). Greenwich, CT: JAI.

Detert, J. R., & Edmondson, A. C. (2005). No exit, no voice: The bind of risky voice opportunities in organizations.

Harvard Business School Working Paper, 05–049.

Dingwall, R. W. (1974). Some sociological aspects of ‘nursing research’. The Sociological Review, 22,

45–55.

Donaldson, M. S., & Mohr, J. J. (2000). Exploring innovation and quality improvement in health care

microsystems: A cross case analysis. Washington, DC: Institute of Medicine, National Academy Press.

Dougherty, D. (1992). Interpretive barriers to successful product innovation in large ﬁrms. Organization Science, 3,

179–202.

Driskell, J. E., & Salas, E. (1991). Group decision making under stress. Journal of Applied Psychology, 76, 473–

478.

Dutton, J. E. (1993). The making of organizational opportunities: An interpretive pathway to organizational

change. In L. L. Cummings, & B. M. Staw (Eds.), Research in organizational behavior (pp. 195–226).

Greenwich, CT: JAI Press.

Edmondson, A. (1999). Psychological safety and learning behavior in work teams. Administrative Science

Quarterly, 44, 350–383.

Edmondson, A. C. (1996). Learning from mistakes is easier said than done: Group and organizational inﬂuences on

the detection and correction of human error. Journal of Applied Behavioral Science, 32, 5–32.

Edmondson, A. C. (2003). Speaking up in operating room: How team leaders promote learn interdisciplinary

action teams. Journal of Management Studies, 40, 1419–1452.

Edmondson, A. C., Roberto, M. A., & Watkins, M. D. (2003). A dynamic model of top management team

effectiveness: Managing unstructured task streams. Leadership Quarterly, 219, 1–29.

Feiger, S. M., & Schmitt, M. H. (1979). Collegiality in interdisciplinary health teams: Its measurement and its

effects. Social Science & Medicine, 13A, 217–229.

DOI: 10.1002/job

STATUS AND PSYCHOLOGICAL SAFETY 963

Foushee, H. C., & Helmreich, R. L. (1988). Group interaction and ﬂight crew performance. In E. L. Wiener, D. C.

Nagel, & M. P. Friedman (Eds.), Human factors in aviation (pp. 189–227). San Diego, CA: Academic Press.

Friedson, E. (1970a). The profession of medicine. New York: Dodd, Mead, and Co.

Friedson, E. (1970b). Professional dominance: The social structure of medical care. Chicago, IL: Aldine

Publishing Company.

Friedson, E. (1994). Professionalism reborn: Theory, prophecy, and policy. Chicago, IL: University of Chicago

Press.

Fuchs, V. (1974). Who shall live? Health, economics and social choice. New York: Basic Books.

Gibson, C. B., & Vermeulen, F. (2003). A health divide: Subgroups as stimulus for team learning behavior.

Administrative Science Quarterly, 48, 202–239.

Grifﬁn, A. (1997). PDMA research on new product development practices: Updating trends and benchmarking best

practices. The Journal of Product Innovation Management, 14, 429–458.

Hafferty, F. W., & Light, D. W. (1995). Professional dynamics and the changing nature of medical work. Journal of

Health and Social Behavior, 35, 132–153.

Hafferty, F. W., & Wolinsky, F. D. (1991). Conﬂicting characterizations of professional dominance. Current

Research on Occupations and Professions, 6, 225–249.

Helmreich, R. L. (2000). On error management: Lessons from aviation. British Journal of Management, 320, 781–

785.

Hermann, C. F. (1963). Some consequences of crisis which limit the viability of organizations. Administrative

Science Quarterly, 8, 61–82.

Horbar, J. D. (1999). The Vermont Oxford Network: Evidence-based quality improvement for neonatology.

Pediatrics, 103, 350–359.

Hult, G. T. M., Hurley, R. F., Guinipero, L. C., & Nichols, E. L. (2000). Organizational learning in

global purchasing: A model and test of internal users and corporate buyers. Decision Sciences, 31, 293–

325.

Institute of Medicine. (1999). In L. T. Kohn, J. M. Corrigan, & M. S. Donaldson (Eds.), To err is human: Building a

safer health system. Washington, DC: National Academy Press.

Institute of Medicine. (2001). Crossing the quality chasm: A new system for the 21st century. Washington, DC:

National Academy Press.

Institute of Medicine. (2003). Keeping patients safe: Transforming the work environment of nurses. Washington,

DC: National Academy Press.

Jacob, J., & Deshpande, S. P. (1997). The impact of ethical climate on job satisfaction of nurses. Health Care

Management Review, 22, 76–80.

Kahn, W. A. (1990). Psychological conditions of personal engagement and disengagement at work. Academy of

Management Journal, 33, 692–724.

Kaissi, A., Kralewski, J., Curoe, A., Dowd, B., & Silversmith, J. (2004). How does the culture of medical group

practices inﬂuence the types of programs used to assure quality of care? Health Care Management Review, 29,

129–138.

Kenny, D. K., & LaVoie, L. (1985). Separating individual and group effects. Journal of Personality and Social

Psychology, 48, 339–348.

Kilo, C. M. (1999). Improving care through collaboration. Pediatrics, 103 (Supplement), 384–393.

Kirchler, E., & Davis, J. H. (1986). The inﬂuence of member status differences and task type on group consensus

and member position change. Journal of Personality and Social Psychology, 51, 83–91.

Klein, A. L. (1976). Changes in leadership appraisal as a function of the stress of a simulated panic situation.

Journal of Personality and Social Psychology, 34, 1143–1154.

Leape, L. L., Bates, D. W., Cullen, D. J., Cooper, J., Demonaco, H. J., & Gallivan, T., et al. (1995). Systems

analysis of adverse drug events. Journal of the American Medical Association, 274, 35–43.

Leonard, M. S., Frankel, A., Simmonds, T., & Vega, K. B. (2004). Achieving safe and reliable healthcare:

Strategies and solutions. Ann Arbor, MI: Health Administration Press.

Littlepage, G., Robison, W., & Reddington, K. (1997). Effects of task experience and group experience on group

performance, member ability, and recognition of expertise. Organizational Behavior and Human Decision

Processes, 69, 133–147.

Lovelace, K., Shapiro, D. L., & Weingart, L. R. (2001). Maximizing cross-functional new product teams’

innovativeness and constraint adherence: A conﬂict communication perspective. Academy of Management

Journal, 44, 779–793.

MacDufﬁe, J. P. (1997). The road to ‘‘root cause’’: Shop-ﬂoor problem-solving at three auto assembly plants.

Management Science, 43, 479–502.

DOI: 10.1002/job

964 I. M. NEMBHARD AND A. C. EDMONDSON

Madhavan, R., & Grover, R. (1998). From embedded knowledge to embodied knowledge: New product

development as knowledge management. Journal of Marketing, 62, 1–12.

Maxﬁeld, D., Grenny, J., McMillan, R., Patterson, K., & Switzler, A. (2005). Silence kills: The seven crucial

conversations for healthcare. Provo, UT: VitalSmarts, LC.

McGregor, D. (1960). The human side of enterprise. New York: McGraw Hill.

Mechanic, D. (1991). Sources of countervailing power in medicine. Journal of Health Politics, Policy and Law, 16,

485–498.

Messick, D. M., & Mackie, D. M. (1989). Intergroup relations. Annual Review of Psychology, 40, 45–81.

Milliken, F. J., Morrison, E. W., & Hewlin, P. F. (2003). An exploratory study of employee silence: Issues that

employees don’t communicate upward and why. Journal of Management Studies, 40, 1453–1476.

Mittman, B. S. (2004). Creating the evidence base for quality improvement collaboratives. Annals of Internal

Medicine, 140, 897–901.

Morrison, E. W., & Phelps, C. C. (1999). Taking charge at work: Extrarole efforts to initiate workplace change.

Academy of Management Journal, 42, 403–419.

Nemeth, C. J. (1986). Differential contributions of majority and minority inﬂuence. Psychological Review, 93, 23–

32.

Norrgren, F., & Schaller, J. (1999). Leadership style: Its impact on cross-functional product development. The

Journal of Product Innovation Management, 16, 377–384.

Nunnally, J. (1978). Psychometric theory (2nd ed.). New York: McGraw-Hill.

Oliva, R. (2001). Tradeoffs in response to work pressure in the service industry. California Management Review,

43, 26–43.

Pagliari, C., & Grimshaw, J. (2002). Impact of group structure and process on multidisciplinary evidence-based

guideline development: An observational study. Journal of Evaluation in Clinical Practice, 8, 145–153.

Ryan, K. D., & Oestreich, D. K. (1991). Driving fear out of the workplace: How to overcome the invisible barriers

to quality, productivity, and innovation. San Francisco: Jossey-Bass.

Sarin, S., & Mahajan, V. (2001). The effect of reward structures on the performance of cross-functional product

development teams. Journal of Marketing, 65, 35–53.

Schaefer, H. G., Helmreich, R. L., & Scheideggar, D. (1994). Human factors and safety in emergency medicine.

Resuscitation, 28, 221–225.

Schein, E. H., & Bennis, W. (1965). Personal and organizational change through group methods: A laboratory

approach. New York: Wiley.

Schmitt, M. H. (1990). Medical malpractice and interdisciplinary team dynamics. In Proceedings of the 12th

Annual Interdisciplinary Health Care Team Conference (pp. 53–66). Indianapolis, IN: Indiana University.

Shamir, B., Zakay, E., Breinin, E., & Popper, M. (1998). Correlates of charismatic leader behavior in military units:

Subordinates’ attitudes, unit characteristics and superiors’ appraisals of leader performance. Academy of

Management Journal, 41, 387–409.

Shine, K. I. (2002). Health care quality and how to achieve it. Academic Medicine, 77, 91–99.

Shojania, K. G., McDonald, K. M., Wachter, R. M., & Owens, D. K. (2004). Closing the quality gap: A critical

analysis of quality improvement strategies. AHRQ Publication No. 04-0051-1.

Shortell, S. M. (1974). Occupational prestige differences within the medical and allied health professions. Social

Science and Medicine, 8, 1–9.

Shortell, S. M., Rousseau, D. M., Gillies, R. R., Devers, K. J., & Simons, T. L. (1991). Organizational assessment in

intensive care units (ICUs): Construct development, reliability, and validity of the ICU nurse-physician

questionnaire. Medical Care, 29, 709–726.

Stamper, C. L., & Van Dyne, L. (2001). Work status and organizational citizenship behavior: A ﬁeld study of

restaurant employees. Journal of Organizational Behavior, 22, 517–536.

Staw, B. M., Sandelands, L. E., & Dutton, J. E. (1981). Threat-rigidity effects in organizational behavior: A

multilevel analysis. Administrative Science Quarterly, 26, 501–524.

Tabachnick, B. G., & Fidell, L. S. (2001). Computer-assisted research and design. Boston: Allyn and Bacon.

Tajfel, H., & Turner, J. C. (1986). The social identity theory of intergroup behavior. In S. Worchel, & W. G. Austin

(Eds.), Psychology of intergroup relations (pp. 7–24). Chicago: Nelson-Hall.

Tucker, A. L., & Edmondson, A. C. (2003). Why hospitals don’t learn from failures: Organizational and

psychological dynamics that inhibit system change. California Management Review, 45, 55–72.

Tucker, A. L., Nembhard, I. M., & Edmondson, A. C. (2006). The Effects of Learn-What and Learn-How on the

Implementation Success of Improvement Projects. HBS Working Paper No. 06-049.

Tyler, T. R., & Lind, E. A. (1992). A relational model of authority in groups. In M. P. Zanna (Ed.), Advances in

experimental social psychology (Vol. 25, pp. 115–191). New York: Academic Press.

DOI: 10.1002/job

STATUS AND PSYCHOLOGICAL SAFETY 965

Van de Ven, A. H., Delbecq, A. L., & Koenig, R., Jr. (1976). Determinants of coordination modes within

organizations. American Sociological Review, 41, 322–338.

Vinokur, A., Burnstein, E., Sechrest, L., & Wortman, P. M. (1985). Group decision-making by experts: Field study

of panels evaluating medical technologies. Journal of Personality and Social Psychology, 49, 70–84.

Wachter, R. M. (2004). The end of the beginning: Patient safety ﬁve years after ‘To Err is Human’. Health Affairs,

W4, 534–545.

Wageman, R. (1995). Interdependence and group effectiveness. Administrative Science Quarterly, 40, 145–180.

Webster, M., & Foschi, M. (1988). Status generalizations. Stanford, CA: Stanford University Press.

Weick, K. E. (2002). The reduction of medical errors through mindful interdependence. In M. M. Rosenthal, &

K. M. Sutcliffe (Eds.), Medical error: What do we know? What do we do? (pp. 177–199). San Francisco, CA:

Jossey-Bass.

Weisband, S. P., Schneider, S. K., & Connolly, T. (1995). Computer-mediated communication and social

information: Status salience and status differences. Academy of Management Journal, 38, 1124–1151.

Wind, J., & Mahajan, V. (1997). Issues and opportunities in new product development: An introduction to the

special issue. Journal of Marketing Research, 34 (February), 1–12.

Wolinsky, F. D. (1988). The professional dominance perspective, revisited. The Milbank Quarterly, 66, 33–47.

Yukl, G. (1994). Leadership in organizations (3rd ed.). Englewood Cliffs, NJ: Prentice Hall.

Zimmerman, J. E., Shortell, S. M., Rousseau, D. M., Duffy, J., Gillies, R. R., & Knaus, W. A., Devers, K., Wagner,

D. P., & Draper, E. A. (1993). Improving intensive care: Observations based on organizational case studies in

nine intensive care units: A prospective, multicenter study. Critical Care Medicine, 21, 1443–1451.

DOI: 10.1002/job

966 I. M. NEMBHARD AND A. C. EDMONDSON

Empowering nurses through inclusive leadership to promote research capacity building: A James Lind Alliance priority setting Partnership in Community Nursing

Article

Full-text available

Jun 2022
J ADV NURS

Aims This paper focuses on the benefits of inclusive leadership when undertaking a priority setting partnership in community nursing, through providing a collaborative and committed nurse‐led forum for initiating impactful changes, identifying evidence uncertainties and driving research capacity‐building initiatives. Design This is a Discussion paper. The project was undertaken between 2020 and 2021. Data sources This paper is based on shared reflections as 70@70 Senior Nurse Research Leaders and is supported by literature and theory. It draws on issues relating to collective leadership, stakeholder engagement, diversity, inclusivity and COVID‐19. Implications for nursing The James Lind Alliance Priority Setting Partnership catalysed the development of a rigorous evidence‐base in community nursing. The collaborative opportunities, networks and connections developed with patients, carers, nursing leaders, policy makers and healthcare colleagues raised the profile of community nursing research. This will benefit nursing research, practice, education and patients in receipt of community nursing care. Collective buy in from national leaders in policy, education, funding and commissioning has secured a commitment that the evidence uncertainties will be funded. Conclusion Four key learnings emerged: collective leadership can ensure learning is embedded and sustained; developing an engaged stakeholder community to promote community nursing research is essential; a diverse membership ensures inclusivity and representation; and insights into the impact of COVID‐19 aid progress. The process increased research engagement and created capacity and capability‐building initiatives. This will help community nurses feel empowered to lead changes to practice. Sustained engagement and commitment are required to integrate research priorities into community nursing research, education and practice and to drive forward changes to commissioning and service delivery. Impact The study promoted research capacity building through inclusive leadership. This can increase community nurses' research engagement and career development and patient care quality and safety; this can incentivize funders and policy makers to prioritize community nursing research.

How Does Inclusive Leadership Curb Workers’ Emotional Exhaustion? The Mediation of Caring Ethical Climate and Psychological Safety

Article

Full-text available

Jul 2022

The COVID-19 pandemic has transformed the politics, economy, and society of the world, which has dealt the most severe blow to the hospitality industry. Meanwhile, the pandemic and government control policies have brought high psychological pressure to hospitality front-line employees, resulting in emotional exhaustion. As a part of burnout syndrome, emotional exhaustion poses a threat to employees’ mental health, career sustainability, and well-being. Therefore, the purpose of this paper was to investigate the curb effectiveness of inclusive leadership on emotional exhaustion and to explore the mediation roles of ethical climate and psychological safety between them. Time-lagged data were collected from 65 teams and 358 hospitality front-line employees working in Chinese hotels in two stages with a one-month gap. This research verified that inclusive leadership has a negative impact on emotional exhaustion both indirectly through the mediation roles of ethical climate and psychological safety. And the ethical climate and psychological safety played partial mediation roles between inclusive leadership and emotional exhaustion. In theory, the findings explored the dual mediation mechanism of the inhibitory effect of inclusive leadership on emotional exhaustion. In practice, we provided the training and correct guidance to develop inclusive leadership for hotel enterprises and to resolve the emotional exhaustion of employees, which can enhance sustainability in careers.

Employee mindfulness, innovative work behaviour, and IT project success: the role of inclusive leadership

Article

Full-text available

Jul 2022
INFORM TECHNOL MANAG

This study investigates the relationship between project employee mindfulness and project success using innovative work behaviour as a mediator and the project manager’s inclusive leadership style as a moderator. Project Manager with high inclusive behaviour will strengthen the relationship of employee’s mindfulness and innovative work behaviour. The data were collected in three-time intervals from a total of 347 information technology project employees. The study findings validated the proposed model wherein employee personality traits, such as mindfulness, have a key impact on the initiation of project employees’ innovative work behaviour. Information Technology projects require innovation due to rapid technical improvements. The study confirms that innovative work behaviour adds to the project’s success. Furthermore, inclusive leadership helps mindful employees become innovative. Thus, the leadership roles should also be emphasised in IT projects.

Team and leadership factors and their relationship to burnout in emergency medicine during COVID‐19: A 3‐wave cross‐sectional study

Article

Full-text available

Aug 2022

Objective: We examined the relationship of team and leadership attributes with clinician feelings of burnout over time during the corona virus disease 2019 (COVID-19) pandemic. Methods: We surveyed emergency medicine personnel at 2 California hospitals at 3 time points: July 2020, December 2020, and November 2021. We assessed 3 team and leadership attributes using previously validated psychological scales (joint problem-solving, process clarity, and leader inclusiveness) and burnout using a validated scale. Using logistic regression models we determined the associations between team and leadership attributes and burnout, controlling for covariates. Results: We obtained responses from 328, 356, and 260 respondents in waves 1, 2, and 3, respectively (mean response rate = 49.52%). The median response for feelings of burnout increased over time (2.0, interquartile range [IQR] = 2.0-3.0 in wave 1 to 3.0, IQR = 2.0-3.0 in wave 3). At all time points, greater process clarity was associated with lower odds of feeling burnout (odds ratio [OR] [95% confidence interval (CI) = 0.36 [0.19, 0.66] in wave 1 to 0.24 [0.10, 0.61] in wave 3). In waves 2 and 3, greater joint problem-solving was associated with lower odds of feeling burnout (OR [95% CI] = 0.61 [0.42, 0.89], 0.54 [0.33, 0.88]). Leader inclusiveness was also associated with lower odds of feeling burnout (OR [95% CI] = 0.45 [0.27, 0.74] in wave 1 to 0.41 [0.24, 0.69] in wave 3). Conclusions: Process clarity, joint problem-solving, and leader inclusiveness are associated with less clinician burnout during the COVID-19 pandemic, pointing to potential benefits of focusing on team and leadership factors during crisis. Leader inclusiveness may wane over time, requiring effort to sustain.

How Leaders Inspire Voice: The Role of Voice Climate and Team Implicit Voice Theories

Article

Full-text available

Jun 2022
J BUS PSYCHOL

That individuals hold personal, implicit beliefs about the consequences of voice (i.e., implicit voice theories (IVTs)), and thus are inherently reluctant to speak up, is one of the most ubiquitous concepts in the voice literature. An extensive body of research also shows how situational factors (e.g., voice climate) profoundly affect voice. Unfortunately, few studies explore how dispositional beliefs combine with situational factors to influence voicing in teams; thus, the literature is unclear about whether and how leaders can encourage voice when their team members implicitly believe that voice is unsafe and inappropriate (i.e., high IVTs). We address this question by proposing a moderated-mediation model linking leaders’ prior reactions to voice (i.e., voice acceptance versus rejection) to team voice intentions via voice climate, such that encouraging leader behaviors (acceptance), unlike discouraging leader behaviors (rejection), creates positive voice climates that increase team voice intentions. Furthermore, drawing from the trait activation theory, we propose that team IVTs composition (i.e., the average IVTs held by each team member) moderates the mediated effect of voice climate on team voice intentions, such that this relationship is strongest for teams with high IVTs because, for these teams, positive voice climates are both highly salient to, and disconfirming of, implicit voice beliefs. Results of a multi-task team experiment support this model. We discuss the theoretical implications of considering person and situation factors in tandem, the potential influence of team IVTs composition, and practical implications for how leaders can inspire voice in teams.

Inclusive Leadership and Innovative Performance: A Multi-Level Mediation Model of Psychological Safety

Article

Full-text available

Jun 2022

Taking both individual and team levels into consideration has been called for years in terms of research on leadership. Inclusive leadership, a trending leadership style emerging from the global needs of managing the increasingly diversified workplace nowadays, has yet been rarely studied at both levels. To answer these calls, we specifically analyzed the relationship between inclusive leadership, team psychological safety, and innovative performance via a multilevel analysis. The results are based on a study of 356 employees from 90 working teams. Individual perceptions of inclusive leadership are positively related to the individual innovative performance through the mediation of individual psychological safety. Team perceptions of inclusive leadership are positively related to the team innovative performance through the mediation of team psychological safety. Moreover, team perceptions of inclusive leadership are positively related to the individual innovative performance through the cross-level mediation of individual psychological safety. Implications for both theory and practice are discussed.

Knowledge mobilization in critical care and the evolving communication role of nurses

Article

Jul 2022

The importance of appropriate communication skills within a health care setting rests upon the need for effective information sharing. When successful, this provides a supportive working environment for staff and has a positive impact on patient care and outcomes. The purpose of this study was to explore how knowledge/evidence is acquired, shared, and applied in the Critical Care (CC) environment for staff and patients/family members. A qualitative approach was used, consisting of semi‐structured interviews and focus groups. Data analysis was conducted using an iterative thematic approach. Data collected prior to the COVID‐19 pandemic from United Kingdom (UK) critical care workers (N = 46), patients, and family members (N = 21) identified four communication roles performed by the nursing staff: team member; diplomat; translator and friend. It was evident that without suitable training and support, the stresses and demands placed upon the nurses could lead to disenfranchisement and burnout. These findings are relevant and timely given the impact of the pandemic, highlighting the need for accessible and alternate communication strategies to support nurses by reducing stress, moral distress and increasing psychological safety. Improved communication can provide tailored information for staff and patients/family improving the CC experience for all.

Organizational factors associated with community therapists’ self-efficacy in EBP delivery: The interplay between sustainment leadership, sustainment climate, and psychological safety

Article

Jun 2022

Background Inner context organizational factors proximally shape therapist experiences with evidence-based practice (EBP) implementation and may influence therapist self-efficacy, which has been linked to sustained use of EBPs in community mental health settings. Research has primarily focused on constructs such as implementation leadership and climate. However, the effects of such factors may depend upon other inner context dimensions, such as psychological safety. Psychologically safe environments are conducive to taking risks, speaking up about problems, and requesting feedback and may promote therapist self-efficacy during implementation. This study examines whether organizational sustainment leadership and sustainment climate relate to therapist EBP self-efficacy only under conditions of psychological safety. Methods Data were collected from 410 clinicians in 85 programs during the sustainment phase of a system-driven implementation of multiple EBPs in children's mental health services. Therapists reported on their organization's sustainment leadership, sustainment climate, psychological safety, and their own self-efficacy in delivering specific EBPs. Multilevel regression analyses were conducted to account for nested data structure. Results Among program-level variables, sustainment leadership and psychological safety both significantly predicted therapist self-efficacy. However, there were no significant interactions between program-level sustainment climate and psychological safety. Exploratory post-hoc analyses revealed a significant interaction between program-level sustainment leadership and therapist-level perceptions of psychological safety such that that the conditional effect of psychological safety on EBP self-efficacy was significant at high levels of sustainment leadership, but not at low or average levels. Conclusion We noted independent links between sustainment leadership, organizational psychological safety and therapists feelings of confidence and mastery with EBPs. Therapists’ individual perceptions of psychological safety were linked to self-efficacy only in programs with high sustainment leadership. Thus, sustainment leadership and psychological safety may both represent implementation intervention targets, but it may not be critical to assess for perceptions of psychological safety before deploying organizational leadership strategies. Plain language abstract Therapist self-efficacy is a therapist's belief that they are capable, knowledgeable, and skilled enough to deliver evidence-based practices (EBPs), and is thought to promote improved clinical and implementation outcomes, such as therapists’ sustained use of EBPs. Conditions within community mental health organizations may influence therapists’ sense of EBP self-efficacy. Leaders’ support and expectations for EBP implementation, and collective staff perceptions about the organization's climate to support EBPs are linked to positive therapist attitudes and EBP adoption. However, less is known about how these implementation-specific organizational factors associated with therapist EBP self-efficacy in the long-term, and how this may depend on general workplace conditions. Specifically, psychologically safe environments – where therapists feel safe taking risks such as asking questions, admitting mistakes, and trying new skills – may be needed to promote self-efficacy when therapists are tasked with learning and using complex multi-component EBP innovations. The current study tested the prediction that leader-driven and program-wide focus on EBP sustainment may promote therapist EBP self-efficacy only in organizations where conditions for learning are psychologically safe. Our findings confirmed that fostering strong sustainment focused leadership and psychologically safe environments may each be important for increasing therapists’ EBP self-efficacy. The model results suggested that individual therapist perceptions of psychological safety were more strongly related to EBP self-efficacy in programs with greater implementation leadership. Findings suggest the importance of increasing EBP leadership behavior to fully potentiate other facilitating conditions for therapist learning in the sustainment phase of EBP implementation initiatives.

To Warn or Not to Warn: Factors Influencing National Weather Service Warning Meteorologists’ Tornado Warning Decisions

Article

Jul 2022

Weather warnings are critical risk communication messages because they have the potential to save lives and property during emergencies. However, making warning decisions is challenging. While there have been significant advances in technological weather forecasting, recent research suggests that social factors, including communication, influence warning meteorologists’ decisions to warn. We examine the roles of both scientific and social factors in predicting warning meteorologists’ decisions to warn on tornadoes. To do so, we conducted a cross-sectional survey of National Weather Service forecasters and members of management in the southern and the central regions of the United States, as well as conducted a retrospective data analysis of cross-sectional survey data from the central region Tornado Warning Improvement Project. Results reveal that dependency on radar velocity couplet and a variety of social factors predicted decisions to warn.

Sustainable knowledge management during crisis: Focus on Covid-19 pandemic

Article

Jun 2022
Bus Inform Rev

This paper consolidates literature that justifies effective knowledge management as a precursor for mitigating the effects of a crisis, Covid-19 pandemic in particular, through key antecedents of leadership, culture, and information and communication technology (ICT). A thorough review of retrieved literature relevant to the topic was conducted. The study materials were rigorously screened to limit any potential biases regarding their selection. Through the study, the paper concludes that the fight against Covid-19 crisis indeed requires knowledge to, among other things, find a lasting solution, mitigate the impacts, limit misinformation, revert to normalcy, and plan for similar crises in future. Further, the paper concludes that sustainable knowledge management during the Covid-19 crisis largely depends on a decisive leadership style that puts employees at the centre; a culture that embraces knowledge as a core asset, and supportive ICT infrastructure. Furthermore, the study reveals that relevance of ICT in the process of managing knowledge, largerly depends on a culture that accepts knowledge as a critical resource. The study establishes some challenges associated with ICT where a way forward for migrating from knowledge capture to knowledge creation and sharing has been re-affirmed. The present paper has led to the development of a model that further explains the relationships between the determinants of leadership, culture and ICT against effective management of a crisis using knowledge as a strategic resource. Further, six propositions have been put forward to provide clarity on the relationships.

Learning from Mistakes is Easier Said than Done: Group and Organizational Influences on the Detection and Correction of Human Error

Article

Full-text available

Jan 2004

Amy C. Edmondson

This research explores how group- and organizational-level factors affect errors in administering drugs to hospitalized patients. Findings from patient care groups in two hospitals show systematic differences not just in the frequency of errors, but also in the likelihood that errors will be detected and learned from by group members. Implications for learning in and by work teams in general are discussed.

Achieving Safe and Reliable Healthcare: Strategies and Solutions

Article

May 2005

From Embedded Knowledge to Embodied Knowledge: New Product Development as Knowledge Management

Article

Oct 1998
J MARKETING

Because new product development (NPD) teams are engaged in knowledge creation, NPD management should emphasize cognitive team processes rather than purely social processes. Using the notions of tacit knowledge and distributed cognition as a basis, the authors propose that the T-shaped skills, shared mental models, and NPD routines of team members, as well as the A-shaped skills of the team leader, are key design variables when creating NPD teams. The authors propose that trust in team orientation, trust in technical competence, information redundancy, and rich personal interaction are important process variables for the effective and efficient creation of new knowledge.

Politeness: Some Universals in Language Usage

Book