Walrath, Jo M. PhD, RN; Dang, Deborah PhD, RN, NEA-BC; Nyberg, Dorothy MS, RN
DISRUPTIVE behavior between clinicians is an entrenched and intractable problem that undermines the culture of safety for patients and clinicians. These behaviors reduce teamwork and communication and adversely affect patients, staff, and the organization by eroding trust, mutual respect, and collegiality.1–6 The complexity of the health care environment contributes to the persistence and pervasiveness of these behaviors.
Numerous factors in the health care environment set the stage for the occurrence of disruptive behavior. Hierarchical structures create steep authority gradients between clinicians and barriers to effectively addressing disruptive behavior. Although this is well documented in the literature, we know little about the similarity and differences between registered nurses (RNs) and physicians (MDs) and their personal experiences with disruptive behavior.
The production pressure to keep patients moving through the system of care adds to the stress of the clinician and the system. Coupled with this complex work environment are clinicians' personality characteristics, socialization during professional education, role differences, and constant change in team members. Disruptive behavior persists when there is a high level of tolerance and indifference,7 an unwillingness to address it, and a lack of awareness of its prevalence and impact on staff, patients, and the organization. Egregious events usually rise to the top of the organizational hierarchy; lower-intensity behaviors fly under the radar of leadership because they are embedded in the culture and are not recognized.
Disruptive behavior is a multidimensional construct. Current literature presents a fragmented perspective of this complex problem. Researchers report numerous types of disruptive behaviors in health care, ranging from incivility to psychological aggression and physical violence,4,8–10 probable causes of disruptive behavior,8,11–13 the effects of these behaviors on individuals4,10–18 and organizations,4,13,14,17,19 and the perceptions of harm to patients.4,9,16,18,20 In a previous study, we presented a conceptual model for disruptive behavior, Johns Hopkins Model for Disruptive Clinician Behavior, which brings coherence to the vast literature base on this topic and aids in understanding its many dimensions. In this model, disruptive behavior is explained by 4 concepts: triggers, disruptive behaviors, responses, and impacts5 (Figure).
Figure. Johns Hopkin...Image Tools
A trigger is a trait, preceding event, or condition that contributes to the occurrence of a disruptive behavior event. Triggers occur at 3 levels: intrapersonal (traits or conditions occurring within the individual such as personal characteristics, lack of competency, or fatigue); interpersonal (relationships between persons); and organizational (systems, processes, or cultures that inhibit interaction or work).
Disruptive behavior is “personal conduct, whether verbal or physical, that negatively affects or potentially may affect patient care including, but not limited to, conduct that interferes with one's ability to work with the other members of the health care team.”21(p3) Disruptive behaviors exist on a continuum based on levels of intensity: low (incivility), moderate (psychological aggression), and high (physical violence). Workplace incivility is defined as low-intensity deviant behavior that violates workplace norms for mutual respect, may or may not be intended to harm the target, does not physically threaten the target, and may transcend organizational hierarchy.11 Psychological aggression is defined as active or passive behaviors that intentionally inflict psychological injury to the target.8 Violence is defined as physical, active, and direct forms of aggressive behavior.5 Implicit in this categorization is that the cumulative effects of lower-intensity behaviors can escalate to higher-intensity behaviors over time.15
Positive responses are those that constructively address disruptive behavior, whereas negative responses are ineffective actions taken to avoid addressing the person with disruptive behavior. Reasons for not addressing the behavior, or not speaking up, provide insight on the barriers staff experience in addressing disruptive behavior.
The impacts of disruptive behavior refer to detrimental consequences for the safety and well-being of patients, staff, and the organization. Consequences patients may face include harm, dissatisfaction, and loss of trust in the provider. For the staff, the consequences may involve physical, emotional, or psychological reactions, which, if traumatic or persistent, may lead to transfers within or resignations from the organization. Organizational impacts could include ineffective teamwork, staff turnover, or interruptions in the delivery of quality patient care.
Guided by this model, an instrument, Disruptive Clinician Behavior Survey for Hospital Settings, was developed to assess disruptive behavior among health care professionals in the hospital practice environment. The purpose of this study was to understand the complexities of disruptive behavior in a large academic medical center. The aims were to (1) determine the type, frequency, and scope of clinicians' personal experience with disruptive behavior and its impact on staff, their patients, and the organization; (2) compare the disruptive behavior experiences of RNs and MDs; and (3) customize interventions for RNs, MDs, interprofessional teams, and the organization to foster culture change and enhance patient safety. The focus of this article is to report a comparison between RNs' and MDs' experiences with disruptive behavior and its impact on them and their patients.
Generation of the survey items was based on the literature and the voice of the clinicians. Registered nurse focus groups were conducted to elicit their experiences with disruptive clinician behavior.5 On the basis of the narrative analyses, a 105-item survey was constructed and composed of 6 subscales that reflected the concepts in the model. After 3 rounds by 7 external RN/MD content experts, a content validity index average of 0.97 was established.22 The survey items were further verified with RN and MD focus groups for applicability to their practice, readability, and language comprehension.
This 105-item survey was administered to RNs in a 325-bed community teaching hospital. The results were used both to conduct an exploratory factor analysis for item reduction and to establish construct validity (K. Miyong, unpublished data, May 2010). The survey was reduced from 105 to 62 items. The exploratory factor analysis resulted in a 1-factor solution for each of the 6 subscales. Reliability of the survey was α = 0.93 (Cronbach α), with subscales ranging from α value of 0.72 to 0.92 (see Table 1 for the survey subscales, levels, and items).
Table 1-a. Disruptiv...Image Tools
The 62-item survey was then administered to the study population in the academic medical center. Survey items included dichotomous, 4- and 5-point Likert scales, and multiple-choice response patterns. Respondents answered the items on the basis of their personal experience(s) with disruptive behavior “during the past year.” An optional open-ended item invited respondents to share a personal experience with disruptive behavior that had a significant impact on them.
Table 1-b. Disruptiv...Image Tools
Design and setting
A descriptive survey design was used to conduct the organizational assessment in a 1013-bed urban academic medical center in the mid-Atlantic region of the United States. This hospital serves community, regional, national, and international patients.
The study population (N = 5710) included 3 clinician groups: (1) all levels of clinical and administrative RNs (n = 2759); (2) nurse practitioners, certified nurse midwives, certified RN anesthetists, and physician assistants (affiliates) (n = 470); and (3) full-time School of Medicine clinical faculty, fellows, and house staff (MDs) (n = 2481). Nurses and MDs working in all departments and practice settings were included. Excluded from the population were agency RNs, RNs not employed by the department of nursing, and School of Medicine clinical faculty and fellows currently not practicing in the study hospital. The study population was limited to RNs, MDs, and affiliates because of the funder's restrictions.
On receiving institutional review board approval, recruitment e-mails with a link to the Web-based survey were sent to the work address of the study population, inviting their participation. Completion of the survey implied informed consent, and staff had the option to complete this survey at work or on their own time. The survey was available from October 11 to November 1, 2010, on a secured Web-based system. Three e-mail reminders for volunteers were sent over the 3-week study period.
Descriptive, univariate, and bivariate analyses (χ2, independent t tests) were conducted using IBM SPSS 19 (Chicago, Illinois).23 The Levine test for equal variance was nonsignificant. Items with missing values were deleted from the analyses.
Overall, 1559 clinicians completed the survey for a response rate of 27.3%. The respondents included 35.8% (n = 987) RNs, 20.0% (n = 496) MDs, and 16.2% (n = 76) affiliates. Most RNs (n = 657/713, 92.1%) and affiliates (n = 67/75, 89.3%) were female; males composed the largest group of MD respondents (n = 206/338, 60.9%). While respondents' ages ranged from 21 to more than 70 years, slightly more than half of the respondents (51.9%) were between the ages of 32 and 51 years. The majority of the respondents (n = 1074) were white (n = 842, 78.4%), Asian (n = 112, 10.4%), and black or African American (n = 61, 5.7%). The average years of practice experience in the study hospital were 10.9 ± 9.7 for RNs (n = 689), 11.4 ± 9.8 for MDs (n = 327), and 9.4 ± 9.7 for affiliates (n = 68). The practice settings of the respondents included inpatient, intermediate, and intensive care units; operating rooms; postanesthesia care units; specialty procedure areas; and outpatient settings.
Scope of disruptive behavior
More than 84% (n = 1322) of respondents reported personally experiencing disruptive behavior during the past year. These respondents experienced disruptive behavior daily/weekly (n = 349, 26.4%) or monthly (n = 389, 29.4%), and 67% (n = 774/1160) indicated that the behavior had been going on for a year or more. In addition, 73% (n = 1097/1503) of the respondents reported observing a coworker who was a target of this behavior.
Respondents were asked to identify the role of the person whose disruptive behavior had the most negative impact on them. Both RNs (n = 250/885) and MDs (n = 133/295) responded that the person who had the most negative impact on them in the past 12 months was a member of their own discipline. In addition, disruptive behavior was reported more frequently within the MD staff (45.1%) (χ2 = 30.594; df = 4; P < .001) than the RN staff (36.5%). The affiliates (n = 39/70, 55.7%) reported MDs as having the most negative impact.
Disruptive behaviors and triggers
An independent-sample t test was performed to compare RNs' and MDs' means for each item in the survey. Affiliates were excluded from the analyses because of the small sample size. While RNs reported experiencing a significantly higher (P < .001 to P = .03) frequency of all disruptive behavior items than MDs, Table 2 presents the 5 highest rank-ordered items in the Disruptive Behavior subscale. Passive aggressive behavior was experienced most frequently by both RNs and MDs and occurred monthly. Unlike MDs, RNs experienced almost monthly occurrence of malicious gossip, self-centeredness, and inappropriate use of communication technology. The highest-intensity behavior, physical violence, was reported by both RNs and MDs as occurring rarely or never.
For the Trigger subscale items, RNs reported a higher frequency of occurrence for all triggers of disruptive behavior. All subscale items were significant for RNs compared with MDs, except for the item lack of competency. The 5 highest rank-ordered triggers are presented in Table 2. Four of these items were organizational triggers: pressure from high census, volume, and patient flow; environmental overload; chronic, unresolved systems issues; and unit/organizational culture.
Responses to disruptive behavior
Responses to disruptive behavior were captured in 3 subscales: Positive, Negative, and Reasons for Not Addressing Disruptive Behavior. The significant findings are presented for each of the subscales in Table 3. For the Positive Response subscale items, RNs reported significantly higher levels of agreement with seeking support from their manager and peers, and using the chain of command, than MDs (P < .001). Both RNs and MDs generally disagreed with the item, I report disruptive behavior in the hospital-defined reporting system.
For the Negative Response subscale, there were 3 significant items. Registered nurses generally disagreed that they accept disruptive behavior as part of my job compared with MDs (P = .001). Both groups also disagreed with the 2 items related directly to patient safety. Nurses reported higher levels of disagreement than MDs in the following items: (1) I do not speak up when I observe behavior that negatively affects patients or employees (P = .04), and (2) I do not report deteriorating patient conditions to the disruptive provider (P < .001).
For the Reasons for Not Addressing Disruptive Behavior subscale items, MDs reported a significantly higher level of agreement that they do not have time to address disruptive behavior than RNs (P < .001). Nurses reported greater disagreement with the statement, I do not have the support of my manager when I address the disruptive person, than MDs (P = .007). Both RNs and MDs indicated that their culture of origin did not inhibit them in addressing the disruptive behavior.
Impact of disruptive behavior
Both RNs and MDs rank-ordered all of the Impacts subscale items the same; there was general agreement that disruptive behavior had a negative impact on them except for 1 item. Both groups generally disagreed that disruptive behavior causes physical symptoms. Nurses reported higher levels of agreement on each of the items (P < .001 to P = .01). Table 4 presents the 6 items with significant differences in the Impacts subscale.
To assess the impact of disruptive behavior on patients, we asked participants whether in the past year, disruptive behavior resulted in harm to my patient. Harms are defined as temporary (eg, requiring treatment or intervention or prolonged hospitalization), permanent (eg, wrong procedure or wrong site surgery), or require life-sustaining interventions (eg, intubation or emergency surgery). Of the 1131 respondents, 10.1% (n = 114) responded that harm had occurred and 19.6% (n = 222) reported that they were unsure whether the event had resulted in harm. For those reporting harm to their patients, they were then asked to select the level of harm that occurred. Although not mutually exclusive, respondents reported 189 incidences of harms to patients as a result of disruptive behavior: 77.2% (n = 146) temporary, 10% (n = 19) permanent, and 12.7% (n = 24) requiring life-sustaining interventions. The majority of the permanent harms and harms requiring life-sustaining interventions were reported by MDs.
We also asked participants whether they considered changing or leaving their current practice setting as a result of disruptive behavior. Of the 692 RNs who responded to this item, 35% (n = 243) agreed they had considered an intrahospital transfer and 10% (n = 68) indicated they were planning to resign. Approximately 20% (n = 67 of 326) of MDs reported they had considered seeking a residency program or position in another organization, and 5% (n = 16) of the clinical faculty indicated they planned to resign as a result of disruptive behavior. Reponses to the transfer and resignation questions were not mutually exclusive.
The study findings expose disruptive behavior as an organizational problem and add to the cumulative body of knowledge that this behavior undermines a culture of safety. While 4 of 5 respondents personally experienced disruptive behavior, 73% (n = 1097) also observed a coworker who was a target of this behavior. Evidence indicates that observing a disruptive behavior event can be as detrimental to the observer as it is to the target of the behavior.16,19 These results challenged this organization to examine the health of the practice environment, clinician well-being, and the impact on patient safety.
Both RNs and MDs reported that the person who had the most negative impact on them was a member of their own discipline. While disruptive behavior between RNs has been reported in the literature, the negative impact of disruptive behavior between MDs is a new finding that can likely be explained by the fact that RNs and MDs spend more work time interacting with members of their own discipline. Disruptive behavior is also an interprofessional problem. This may be attributed to the fact that RNs and MDs are educated in silos and have little to no opportunity to develop collaborative working relationships during their education. The current movement toward interprofessional education may reduce disruptive behavior because RNs and MDs increase their interactions throughout their professional education. In addition, variable team membership, RNs' shift rotations, long work weeks and frequent clinical rotations for house staff, and the process for consultation services may further interfere with collegiality and collaborative working relationships.
Nurses experienced a higher frequency of occurrence of all disruptive behaviors than MDs. Several factors may contribute to this finding. For example, the organizational structures in hospitals have a hierarchical chain of command for both RNs and MDs. In a study of public sector employees, instigators of disruptive behavior often held higher authority positions in the hierarchy than their targets, and women experienced higher frequencies of disruptive behavior than their male coworkers.9 In hospital settings, RNs are often perceived as holding lower hierarchical status than MDs and are predominately women; further study is required to establish whether gender and hierarchy play a significant role in the health care setting.
Disruptive behavior exists on a continuum based on intensity. These study findings identified a range of the intensity of disruptive behaviors from incivility to violence. Moderately intense behaviors (passive aggression and conflict) were the most frequently experienced by both RNs and MDs. The frequency of these behaviors may be a by-product of chronic exposure to low level intensity behaviors such as incivility, which are difficult to observe, frequently “fly under the radar” of leadership, often go unaddressed and have escalated over time.8 Physical violence, the highest intensity behavior, was rarely or never reported by RNs and MDs. This behavior is less subjective and easier to observe than incivility and psychological aggression. In addition, there are long-standing codified workplace violence policies in this organization, as in others, with clearly defined consequences for the instigator of physical violence.
For every occurrence of disruptive behavior, there is a triggering event that can be an intrapersonal, interpersonal, or organizational issue. Preventing triggers from occurring removes or minimizes the catalysts for disruptive behavior. In this study, both RNs and MDs identified pressure from high census, volume, and patient flow; environmental overload; and unresolved system issues, all organizational triggers, as the most frequently occurring. This information provides a focus for performance improvement initiatives. When RNs and MDs are involved with key decision makers to uncover the root cause of the triggers to disruptive behavior, this strengthens teamwork and builds relationships as clinicians become collectively invested in ensuring an effective and sustainable solution.
While both RNs and MDs generally agreed that they seek support from their peers and managers to address the disruptive behavior, they do not report disruptive behavior events through the hospital-defined reporting system. Barriers to reporting may include burdensome reporting processes, lack of specificity of what constitutes disruptive behaviors, staff's fear of retaliation, and lack of feedback or follow-up on actions when an event is reported. Reporting disruptive behavior is key to identifying trends, designing interventions, and monitoring their effectiveness.
Despite the provider's disruptive behavior, MDs and RNs speak up when observing behavior that negatively affects patients or employees and report a patient's deteriorating condition. They do not let the person with disruptive behavior get in the way of patient care. This finding is in contrast to previous research findings where intimidation was a barrier to reporting.6,15 However, it is also possible that this was a sensitive item and participants sacrificed the accuracy of their responses for one that would be perceived as more socially acceptable.24
These study findings are consistent with the findings of previous research on the impact of disruptive behavior on staff. It decreases job satisfaction and morale25,26 and hinders working relationships,4,5,15,27 all of which can place patients at risk for harm. While previous researchers quantified perceptions of harm to patients, this study sought to capture the actual knowledge of harm related to disruptive behavior. A major finding from this survey was that staff reported direct knowledge of actual harm to patients as a result of disruptive behavior. Clinicians espouse to do no harm. With evidence that disruptive behavior events resulted in actual harm to patients, clinicians and leaders have a moral imperative to address and minimize these events.
Disruptive behavior can put patients at risk by temporarily destabilizing the work environment as a result of turnover. Similar to other studies, RNs reported their intention to transfer to another unit or plans to leave the organization as a result of disruptive behaviors.26–29 Unexpectedly, we found that clinical faculty also reported their intention to leave the organization as a result of disruptive behavior. This turnover, if acted on, carries significant financial costs and loss of intellectual capacity for the organization. In this study, if the reported intent to transfer or leave the organization occurred, the projected incremental financial costs for replacement of RNs who transfer or resign would be $10.2 million, calculated using orientation, training, and vacancy backfill costs. Projected incremental costs for replacement of clinical faculty positions would be $5.3 million. Faculty recruitment costs, decrease in revenue due to loss of patient admissions and referrals, and disruption in research are not included in these calculations.29
There are several limitations of this study. The study findings represent one academic medical center, which may limit generalizability to other hospital settings. However, similar hierarchical structures and common processes are inherent in all hospital organizations, which could allow for generalization outside an academic medical center. This assumption was supported by similar survey findings in nonacademic health care settings (D. Nyberg and D. Dang, unpublished data, May 2010, S. Strobel and D. Dang, unpublished data, November 2011).
A response rate of 27.3%, a lack of data on nonresponders, and the potential that responders were victims of disruptive behavior raise the issues of selection bias and representativeness of the results. Despite the response rate, the results from more than a quarter of the practicing clinicians provide compelling evidence of an unhealthy work environment and an organizational imperative to address the harm to patients, staff, and the organization. The response rates varied for all subscales and the demographic items. This is likely due to several factors: the length of the survey, fatigue of the respondents, interruptions during completion of the survey, or demographic items that were too specific such that the respondents felt they may be identified. To address these factors that may have contributed to the variability of the response rate, the survey items have been reduced from 62 to 42, using item response theory,30 and limited the specificity of the demographic items.
Finally, the survey title terms, disruptive behavior, may have connotations that predispose how the respondents answer the items. Subsequently, the title has been revised using neutral words, Survey of Unprofessional Behaviors: Triggers, Responses, Impacts.
Disruptive behavior is a complex phenomenon with significant implications for patients, clinicians, and the organization if not addressed. An organizational assessment that identifies the full scope of disruptive behavior is the first step to achieving and sustaining a culture of trust, mutual respect, and collegiality, the hallmarks of a culture of safety. The model serves as a road map for such an assessment. The survey provides actionable information to address disruptive behavior as a systems issue and can guide the development of individual, team, and organizational interventions.