Souba, Wiley MD, ScD, MBA; Way, David MEd; Lucey, Catherine MD; Sedmak, Daniel MD; Notestine, Mark PhD
Most everyone is familiar with the expression “the elephant in the room,” a phrase which is used to denote an obvious problem that people are unwilling to discuss or confront. The idea is simple: An elephant in a room would be impossible to miss; thus, the people in the room who are jointly and cooperatively pretending it is not there have made a collective choice to ignore it.
Morrison and Milliken1 popularized the term “organizational silence,” which refers to the collective-level phenomenon of doing or saying very little about the problems facing an organization. Organizational silence derives both from people's fears of negative feedback and from a set of behavioral cues adopted by supervisors that lead to structures, procedures, and processes that discourage speaking up. Two common structural features of organizations that foster organizational silence are centralized decision making and a lack of formal feedback mechanisms.1
Some organizations face an apparent dilemma in which employees know the truth about specific problems within the organization yet dare not speak that truth to their superiors. A key factor that fosters the creation of a climate of organizational silence is senior leaders' fears of receiving criticism, especially from subordinates.1 The unwritten message from the top is “No bad or unpleasant news.” Fearing retaliation or the label “not a team player” if they speak their minds, subordinates become silent; even if they do speak up, they may discover that their feedback is disregarded. A culture of silence becomes ingrained.
Ignoring elephants comes with a price.2 Maxfield and colleagues3 report on clinicians who infrequently questioned their colleagues even when they were aware of clinical decisions that could harm patients. Said otherwise, organizational silence impairs performance because the information that decision makers receive is distorted. When debate is absent, innovative solutions often do not emerge and groupthink becomes the norm.4 The organization is less likely to detect and correct errors. When organizations and their leaders do not make course corrections, or when they chart the wrong course, organizational performance suffers.5,6
Most studies of organizational silence have focused on the corporate world.1,7 “Elephants” have not been studied in any substantive way in academic health centers (AHCs). The purpose of the present study was to obtain the perspective of department chairs regarding elephants in their AHCs. We chose to poll surgery and medicine department chairs because they oversee large and complex research, education, and clinical enterprises and because they work closely with medical school and hospital leaders in carrying out their roles and responsibilities. We believe that an environment such as the medicine or surgery department of a medical school could be a fertile breeding ground for elephants.
We surveyed the chairs of departments of medicine and of surgery in U.S. medical schools that were accredited by the Liaison Committee on Medical Education (LCME). These chairs are individuals who have firsthand experience with the complexities of organizational behavior and dynamics. We identified the survey recipients through the membership rosters of the Society of Surgical Chairs and the Association of Chairs and Chiefs of Medicine. We verified the rosters and updated participants' contact information by reviewing medical school Web sites. When we sent out the survey in the spring of 2010, there were 127 LCME-accredited medical schools in the United States (including Puerto Rico). We identified any individual with the title of department chair or interim department chair for the departments of medicine and of surgery at each of the 127 medical schools for a total population of 254 department chairs.
We designed a survey instrument to assess the prevalence of organizational silence within the clinical departments of AHCs. The literature on organizational silence contributed to the formation of the survey questions.1,7
We designed the survey to save participants time and to maximize return rates. Accordingly, we used ranking tasks as alternatives to Likert scales. The survey included 15 items divided among three ranking tasks that answered the following questions: (1) What are the major elephants in your AHC? (2) What do you believe to be the most prevalent reasons people do not speak up? and (3) What are the consequences of remaining silent? In addition, we asked participants to answer, using Likert or Likert-type scales, questions about the prevalence of, locations of, reactions to, and possibility of dispelling elephants in their organizations or institutions. On the questionnaire, we defined elephants as “important problems within departments, the medical school, or the teaching hospitals that need to be confronted, but for various reasons are ignored, often for long periods of time.” We also asked how long the chairs had served in their position (see Supplemental Digital Appendix 1, http://links.lww.com/ACADMED/A67, for the final version of the survey).
We pilot-tested the survey instrument with the clinical department chairs of the sponsoring medical school. To check for validity, we compared the pilot test results and our personal assessments of the prevalence of elephants in the AHC. In addition to taking the survey, pilot test respondents also provided feedback on each of the questionnaire items. We made adjustments to the survey instrument on the basis of feedback from the pilot group.
We used both postal and electronic versions of the questionnaire to poll the department chairs. We disseminated the electronic surveys through the Survey Monkey electronic survey service (Palo Alto, California). A cover letter explained the purpose of the study and enlisted each recipient's participation. We did not offer any incentives for completing this survey. We explained in the cover letter that we would code the returned surveys for tracking purposes and that we would hold responses in confidence until the conclusion of data collection, at which time the responses would be deidentified and reported only in an aggregate manner. Tracking provided us with the ability to follow up with nonrespondents, to match medicine and surgery chairs from the same institution, and to assess how well survey participants represented the population at large. We timed the dissemination of the postal and electronic versions of the questionnaire so that the two versions would arrive simultaneously, allowing each recipient to choose how he or she wanted to respond. We mailed the first wave of questionnaires during the last week of April 2010. We sent a follow-up e-mail reminder to nonrespondents two weeks later. We sent a second wave of both electronic and postal questionnaires in early June 2010.
We merged electronic and paper survey responses into one data set and analyzed the results using SPSS for Windows 17.0 (Chicago, Illinois). We analyzed respondent and nonrespondent demographic data for representativeness using chi-square tests of proportions. The institutional demographic characteristics that we used to test for bias include (1) region of the country (Northeastern, Central, Southern, or Western), (2) type of institution (public or private), (3) faculty size by percentile rank (1st–25th percentile, 26th–50th, 51st–75th, or 76th–99th), and (D) institutional research ranking from U.S. News & World Report.8
The Ohio State University Behavioral Science Research Review Board reviewed and approved this study protocol.
The pilot survey return rate was 8 of 18 (44%). The pilot survey participants' assessment of organizational silence aligned with ours. We made some adjustments to the instrument on the basis of feedback from our home institutions' department chairs.
Demographics of survey respondents
The overall survey return rate was 55% (139 of 254). Returns were greater for chairs of surgery (68%; 86 of 127) than for chairs of medicine (42%; 53 of 127). Our chi-square test of proportion analyses (Table 1) indicated that survey participants represented the general population of department chairs of medicine and of surgery at U.S. LCME-accredited medical schools in the following characteristics: region of the country, type of medical school, size of faculty, and U.S. News & World Report8 ranking.
More of the 139 respondents worked at institutions in the Southern region of the country (as defined by the Association of American Medical Colleges) than in other regions; 53 (38%) worked in medical schools in the South, 40 (29%) in the Central region, 32 (23%) in the Northeast, and 14 (10%) in the West. About two-thirds of the respondents worked in public medical schools (89; 64%), whereas about a third worked in private schools (50; 36%). Respondents were equally distributed among medical schools of varying faculty sizes: below the 25th percentile (40; 29%), between the 25th and 50th percentile (35; 25%), between the 51st and 75th percentile (34; 24%), and above the 75th percentile (30; 22%). Finally, the majority of respondents (84; 60%) came from medical schools in the unranked tier (rank of 50+) of medical schools as reported by the U.S. News & World Report,8 whereas 29 (21%) came from tier 3 (ranks between 26 and 50), 15 (11%) came from tier 2 (ranks between 11 and 25), and 11 (8%) came from the top tier (ranks between 1 and 10). Finally, the average tenure in the position of department chair for all respondents was 7.5 years.
Types of elephants
The first survey task asked each chair to rank-order the five most common elephants in his or her own organization. The responses of the chairs of medicine and surgery were nearly identical. The most common elephant reported by chairs is a misalignment between goals and available resources (Table 2). Two other elephants perceived as major dilemmas are ignoring information that clearly indicates a performance problem and the unwillingness to give up on a failing strategy. These were ranked 2 and 3 by both medicine and surgery chairs.
Reasons for organizational silence
A second ranking task asked chairs to identify the major reasons people decline to speak up about elephants. The chairs were fairly consistent across departments in their rankings of reasons people decline to speak. Both medicine and surgery chairs ranked the “belief that speaking up will be ignored” as the number 1 reason people are silent (Table 3). Medicine chairs ranked the “fear of repercussions” as number 2, whereas surgery chairs ranked that item as number 3; surgery chairs ranked “reluctance to deal with a problem once exposed” as number 2, whereas medicine chairs ranked it third.
Consequences of organizational silence
The third and final ranking task asked chairs to identify the most common consequences of not speaking up about elephants. Surgery and medicine chairs felt differently about the effects of ignoring obvious problems. Medicine chairs felt that the most negative effect is low morale among faculty and staff. Surgery chairs, on the other hand, were more concerned about the organization not learning from its mistakes (Table 4). Both surgery and medicine chairs conveyed concerns about the poor decisions, a consequence of inadequate information sharing, that could result from elephant dodging.
Prevalence of, locations of, and reactions to elephants
Chairs provided their views regarding how prevalent elephants are in their own AHC and how the prevalence of elephants at their own AHC compares to the prevalence at other centers. Again, chairs from both medicine and surgery were very consistent in their responses. About two-thirds of chairs (95 of 137; 69%) said elephants are considerable or widespread in their own organization, and nearly another two-thirds (82 of 133; 62%) believed that the status of the elephants at their own organization is about the same as at other AHCs. Only 16% (21 of 133) said that elephants are more common in their home institution, whereas 23% (30 of 133) said that they are less common.
Both chairs of surgery and of medicine believed that elephants are more commonly ignored by deans and hospital leaders than by other department chairs or themselves. Surgery chairs were more likely to say that hospital leaders ignore elephants, whereas medicine chairs were more likely to say that deans disregard elephants (Figure 1).
All 139 of the respondents answered the questions regarding how openly faculty and leaders at their AHCs discuss and confront elephants. Only 52 of the chairs (37%) said that elephants are usually discussed in an appropriate venue, whereas 87 (63%) said that elephants are discussed in less constructive venues or not discussed at all. Less than a quarter of the chairs (32; 23%) reported that the top leaders at their institutions actually encourage people to call out and deal with elephants. More commonly, the chairs (77; 55%) reported that the top leaders of their institutions say they want people to be frank about elephants, but their actions or nonverbal cues indicate otherwise. A higher percentage of medicine chairs than of surgery chairs (16 of 53 [30%] versus 16 of 86 [19%]), said that top leaders pretend that elephants do not exist. Only 7 chairs (5%) reported that their institutions' top leaders are unaware that elephants actually exist at all.
Creating a culture that confronts organizational silence
Chairs answered questions about whether some issues should be ignored and about the ease of creating a culture more amenable to discussing problems. Most chairs (86 of 139; 62%) disagreed with the idea that some elephants at their institutions are better left alone, but nearly a fifth (26 of 139; 19%) agreed that, indeed, some issues are best left undiscussed. Even though most chairs felt elephants should be addressed, two-thirds (92 of 137; 67%) felt that creating a culture in which elephants are openly discussed would be very or moderately difficult. About 80% (112 of 139) of chairs believed that the best way to encourage people to call out elephants is for top leaders to set the example by acknowledging and confronting elephants. Setting dedicated time aside at meetings to discuss elephants was a distant second; only 22 chairs (16%) endorsed this solution.
Discussion and Conclusions
Organizational silence and its effects
In virtually every organization, forces exist that cause members to withhold information, resulting in a collective-level phenomenon known as “organizational silence.”1 Our study indicates that academic medicine is no exception; nearly 70% of our respondents reported that elephants are common or widespread in AHCs. We believe that AHCs are designed, often subconsciously, to keep the range of conversation limited to a few voices, usually the voices of those in power. The powerful silence the voices of others because they consider others' views to be either contrary to the status quo or of limited value.9 Weick points out that in trying to make sense of their environments, organizations' leaders routinely regulate, standardize, and simplify.10 These managerial activities, which are designed to create order and consistency, often contribute to organizational silence.10 Henriksen and Dayton agree; in their study on patient safety, they write, “The irony is that out of deference to existing authority gradients and a desire to maintain harmonious working relationships with colleagues, providers suppress their concerns about doing the right thing, and further distance themselves from having meaningful discussions about practices that will ensure safe and high quality care.”11
Zaffron's and Logan's views support those of Henriksen and Dayton. Zaffron and Logan12 stress that it is the unsaid and uncommunicated components of language that represent the greatest opportunity for improving organizational performance. They believe that this unspoken domain “determines and shapes which messages are possible, not possible, important, unimportant, relevant, not relevant, appropriate, not appropriate, and so on.”12 They have written that until organizations confront the silence, that is, “[u]ntil we find leverage on this part of language, the future is written and can't be altered.”12 In other words, without addressing our elephants, we live unknowingly in our same ol', same ol' future and nothing much changes.13 People cannot solve the problems that they do not talk about, yet the overriding choice made by employees in many organizations, including AHCs, is not to talk.1
This silence is curious and perplexing in academic medicine given that top AHC leaders are not only aware that elephants are common, but also cognizant of the importance of upward information flow for organizational performance.5,6 When an organization's culture impairs open and candid communication, or when people interpret silence as consent, understanding why problems remain unaddressed is easy. Indeed, the inability to manage agreement (e.g., everyone agrees to remain silent about a problem that everyone agrees exists) is a major form of organizational dysfunction, often posing a greater challenge than managing conflict.14 This inability to appropriately confront elephants was apparent in our findings. When elephants were exposed or named, they were discussed in an appropriate venue less than 40% of the time. Most commonly, members of the AHC discussed them in informal settings where the conversations were less constructive and where all the relevant stakeholders were not present.
According to the department chairs who responded to our survey, the primary deleterious consequence of ignoring elephants is the negative effect on organizational learning and change. If they do not learn of and from elephants, leaders and administrators fall short of mastering the “conversational domain” of leadership necessary to confront tough issues; consequently, they limit their range of possible responses to challenges that arise.15 Moreover, since lower-ranking employees (or junior faculty) tend to be silent about bad news, positive information is much more likely to flow up to the top of organizational hierarchies than negative information, resulting in a dearth of information about potential problems. This lack of balance between the knowledge of problems and successes at the top will likely lead to serious deficits in the information that leaders use to make key decisions.5
Silence in health care
Very few others have written about the effects of elephants or organizational silence in health care. Hafferty has addressed the familiar silence around the tension between medical professionalism and commercialism16 while Hart and Hazlegrove have used the term “cultural censorship” to describe a deceptive side of health service organizations—a side where problematic events are simultaneously recognized yet hidden.9 Their description aligns with that of Sherriff who points out that “one of the central features of cultural silence is that it tends to be, in rather paradoxical terms, simultaneously recognized and concealed.”17
Similarly, Henriksen and Dayton11 coined the phrase “consensual neglect” to refer to the tendency of decision makers in health care and other organizations to covertly ignore many of the upsetting issues or difficult problems that exist in a misguided effort to achieve unity of purpose and act in a unified manner. They believe that consensual neglect absolves people from taking on tough issues: “Maintaining the status quo is comfortable and requires no further action. Breaking away and taking a different course of action requires decision making, uncertainty, doubt, and renewed responsibility. As a consequence, it is easy to find reasons to do nothing.”11
Henricksen and David's consensual neglect may explain the phenomenon Hafler and colleagues18 note in their discussion of the hidden curriculum that medical school faculty often experience. These faculty encounter a variety of conflicting messages about their educational and professional undertakings that regularly go unaddressed.
Ending the silence
More than half of the chairs surveyed reported that top leaders in their AHCs said they wanted to hear about the elephants but their actions and non-verbal cues sent a different message. AHC leaders may want to hear about problems, but individually they must actively work at influencing the culture so that problems are more likely to come to them. As Edmondson writes, “Hospital cultures, in short, are patchwork quilts rather than uniform, smooth fabrics where learning culture, or what some have called patient safety culture, is concerned. The variation is primarily driven by local leadership behavior, which in both overt and subtle ways shapes the climate for learning”19 (emphasis added).
Changing the culture of silence may be difficult for leaders, as more than likely, they developed as leaders within—that is, as members of— the culture that does not address elephants. Kotter points out that culture is powerful (and difficult to change or even discuss, even from the top) because individuals are indoctrinated into the organization so well, because the culture exerts itself through the actions of thousands of people every day, and because culture-shaping takes place without much conscious thought.20 Our finding that top AHC leaders often fail to “walk the talk” highlights this conundrum, this cycle of organizational silence that is reinforced by the proverbial “that's just the way we do things around here.”21
Tackling organizational silence can become much more difficult, if not impossible, when one part of the organization—be it a leader or administrator, a faculty member, a department, or the board of trustees—is exempt from the espoused values, goals, and guiding principles. The resentment among the rank and file that develops is deep seated and often pervasive. When political decisions trump decisions made by process, transparency, rationality, and prudence, the decisions are arbitrary and ad hoc, and the outcome is consistently suboptimal.21 People quickly realize which elephants are not up for discussion, and they know that if they speak up, retribution may be lurking around the corner. Our findings regarding the reasons members of AHCs ignore elephants—the most common, as reported by our respondents, was the belief that speaking up would have little or no effect, followed by, among others, a fear of personal repercussions—align with Hart and Hazelgrove's9 understanding of the causes of organizational silence. They write, “Constrained by lack of consensus and motivated by a variety of political and psychological interests, people choose to forget what they know and withdraw into silence.”
The possibility of liberating the “unspoken” domain of the academic medicine culture represents an enormous opportunity for improving AHC performance.15 Leaders must not only create the psychological safety necessary to permit the unspoken to be said but must also model the courage to actually say the unspoken. As the unspoken is emancipated, mental maps relax and people begin having conservations that are more authentic and more open. Transparent information flows freely up and down organizational hierarchies. Barriers to organizational change become dismantled. Decision making becomes faster and better.
People's actions (or inactions) always correlate with the way in which the circumstances they are dealing with occur (show up) for them.22 Furthermore, the way a situation occurs for an individual is always colored and shaped by the context he or she brings to those circumstances. In other words, if a faculty member's elephant occurs to her as risky or pointless to address, her actions (or inactions) will match that occurring and she will remain silent. Leaders must create psychologically safe workplace contexts so that organizational members will speak up. Language permits leaders to convert events and issues into “talkable” objects. Without the ability to convert a problem into a “talkable” one, they cannot lead effectively.
Quality guru W. Edwards Deming23 used to say, “Drive out fear.” He was referring to the kind of fear that deters people from speaking up. When people are afraid of their bosses, afraid of being punished for making mistakes, or afraid of being labeled troublemakers if they say what is on their mind, performance suffers. More than 80% of chairs reported that having top leaders lead by example—that is, by acknowledging and tackling elephants themselves—would be the best way to attack organizational silence. To ensure that people will speak up when they discover problems, leaders who are serious about designing learning organizations must realize that the dominant tendency is for employees to regard speaking up about concerns as risky and, thus, to withhold information.1
The finding that both medicine and surgery chairs believe that the very large majority of organizational elephants reside outside their own departments is a curious one. It raises an important question: Do faculty and leaders in general believe that the unspoken problems in their AHCs lie with others? Future studies on organizational silence designed to obtain the perspectives of deans and hospital leaders (and junior faculty) would help address this question. Additional studies confirming that faculty and administrators believe that elephants are less of an issue in their own department or unit and more of a problem in others would suggest that the challenges around tackling organizational silence are even more complex. In other words, not only do the observers ignore the elephants, they suffer from myopia as well.
Our study is the first to explicitly examine the views of chairs of medicine and of surgery at U.S. LCME-accredited medical schools regarding organizational silence. The response rate, especially high among surgery department chairs, and the representativeness of our respondents of the general population of surgery and medicine department chairs are strengths of our study. Still, our study may not be generalizable to all chairs at all medical schools. Our ranking-task and Likert-like responses may have increased our response rate, and the choices we provided were based on the literature,1,2,7 but future qualitative research may uncover further types, causes, and consequences of unspoken issues in academic medicine. We examined chairs' perceptions only. Further studies may help validate our respondents' views, including those regarding deans' and hospital leaders' actions or inactions. Given the emerging field of neuroleadership,24–27 investigations into the ingrained mental maps and frames of reference that give rise to organizational silence would also likely provide important insights into this behavioral tendency.
The performance benefits of eradicating organizational silence are clear. Higgs28 maintains that the rationale for eliminating elephants goes beyond the business imperative: “[H]elping people to break their silence, or to find their voice, hitherto unheard or unacknowledged, is one of our major moral imperatives.” To begin the process of reducing organizational silence, institutional leaders must recognize and support individuals who ask penetrating, uncomfortable questions rather than viewing these individuals as rabble-rousers or nonteam players. Vigorous, healthy debate, where no topic is verboten, is a sign of organizational health, but few AHCs are so healthy.
The Ohio State University Behavioral Science Research Review Board reviewed and approved this study protocol.