Predictors of moral harassment in nursing work in critical care units

Rev Bras Enferm. 2021;74(3): e20200442 https://doi.org/10.1590/0034-7167-2020-0442 8 of ABSTRACT Objectives: to analyze the predictors of moral harassment in nursing work in critical care units. Methods: a cross-sectional study conducted in a public hospital in Fortaleza, Ceará, with 167 nursing professionals in 2016. Sociodemographic/occupational questionnaire and Negative Acts Questionnaire Revised were applied. The analysis included descriptive statistics, measures of central tendency and dispersion, as well as Mann-Whitney, Kruskal-Wallis and Conover Inman U-tests for multiple comparisons. Results: there was a 33% prevalence of self-perception of moral harassment, highlighting personal/professional disqualification and work-related harassment. The predictors of moral harassment included age, time working in the job and time in the unit, employment relationship and sector. Conclusions: young professionals (< 30 years), cooperative, crowded in intensive care or emergency units, with less time working in the job (< 5 years) or greater time in the unit (above 10 years) are the biggest victims of moral harassment in the work of nursing in critical environments. Descriptors: Nursing; Work; Non-Sexual Harassment; Incivility; Critical Care.


INTRODUCTION
Around the world, the working context of nursing and its unfolding in the team's results have been the subject of investigation in different scenarios. A greater emphasis is directed to the study of disruptive behaviors in healthcare work, which are common, measurable and associated with the culture of safety and professional welfare (1)(2)(3)(4) .
Disruptive behaviors in healthcare work involve incivility and psychological violence, in addition to physical or sexual violence (2,5) ; and, many times, are treated as synonyms of bullying at work. Researchers reviewed the attributes of this phenomenon in nursing and found characteristic behaviors of incivility, psychological aggression and moral harassment (6)(7) .
Specifically, moral harassment is any abusive conduct, performed with words, acts or behaviors that may damage the physical or psychic integrity of the worker. In this aggression, the harasser uses the weak points of the victim and makes him/her doubt himself/ herself, aiming at annihilating his/her defenses and, in such a way, progressively shaking his/her self-confidence (8) .
It is, therefore, a phenomenon that has been present for many years in nursing work environments and that has plagued the health of professionals worldwide, often veiled or remaining only as an index for situational diagnosis (1,9) .
Studies point out the need to investigate this phenomenon better in different practice scenarios, as it directly affects patient care. This happens because the harassed professionals feel incompetent and incapable of carrying out their work, leading to a higher risk of errors and adverse events, because they become unable to think clearly or to concentrate (1,(10)(11) .
Despite the real and complex situation of moral harassment in the hospital environment, studies on this phenomenon in nursing should diversify the public and the places of performance, which contributes to the expansion of knowledge about this problem in different circumstances (7) .
In this perspective, although international evidence indicates a higher frequency of moral harassment in critical or high stress units, such as intensive care, operating room and emergency (12)(13) , there is still no concrete evidence on the predictors of this type of violence in such practice scenarios in Brazil.

OBJECTIVES
To analyze the predictors of moral harassment in nursing work in critical care units.

Ethical aspects
The study respected all formal requirements for research with human beings, with the approval of the Ethics in Research Committee of the hospital chosen as the scenario of the research. All participants read and signed the Free Informed Consent Term, in two copies.

Design, time and place of study
Cross-sectional study guided by the STROBE tool and developed in a public teaching hospital, located in Fortaleza, Ceará, Brazil, in 2016. It is a reference institution for the North and Northeast regions of the country, with 525 beds in several units and specialties. It was specifically carried out in intensive care units (ICU), emergency and surgical center, as they are sectors of critical care, where there is greater possibility of evidence of moral harassment according to previous studies (1,(12)(13) .

Population or sample; inclusion and exclusion criteria
According to data from the hospital, in 2016, approximately 600 workers were part of the nursing team of the emergency units, ICU and operating room, among coop members and public servants. The sample was calculated based on the formula for finite population, considering if sample error of 5%, confidence interval of 95% and prevalence of 50%. With this, there was an estimated total of 234 participants.
The sample was formed by convenience, being included nursing professionals who met the following criteria: to occupy only a care function and to have, at least, one year of activity in the sector. The professionals who were on vacation or distant in the period of the data collection were excluded. At the end, the total of 167 participants was reached, corresponding to 70% of the estimated sample.

Study protocol
Two instruments were used to collect the data: a socio-demographic and occupational questionnaire, elaborated by the authors, contemplating the variables gender, age, time working in the job, time in the unit, post-graduation, number of jobs, weekly workload, sector/unit and type of employment relationship; and the Negative Acts Questionnaire Revised (NAQ-R), adapted and validated in Brazil (14) .
The good quality of NAQ-R in determining variables related to bullying is endorsed in several studies (15)(16)(17) , as it allows the identification of targets and the magnitude of exposure, and may favor the construction of significant intervention programs. It is composed of 22 items, each in the form of behavior, without reference to the term "harassment"; it measures how many times the interviewee, during the last six months, was subjected to a series of negative acts and potentially offensive behaviors; and it differentiates subjects who were harassed from those who were not, using a Likert scale from 1 to 5 (1 -never, 2 -from time to time, 3 -monthly, 4 -weekly, 5 -daily). Cases of harassment are those in which professionals have experienced or have experienced at least one negative act per week in the past six months (14) .
It is divided into four factors: Factor 1 -Work-Related Harassment: the person suffers pressure, negative criticism and boycotts, in such a way that his/her performance is impaired at work; Factor 2 -Personal Harassment: the person feels attacked by hostile behavior and exclusion; Factor 3 -Personal and professional disqualification: the person feels that both his/her work and opinion are not taken into account; and Factor 4 -Physical Intimidation: the person finds himself/herself involved in a situation of physical intimidation or even is a victim of physical violence (14) .
The instruments were distributed to nursing professionals in the different shifts (morning, afternoon and night), in a closed envelope and with a deadline of one week for return, without participant identification.

Statistical analysis of results
The data was processed in the Statistical Package for the Social Sciences (SPSS) version 20.0, license number 10101131007.
The continuous quantitative variables were analyzed for the mean and standard deviation (SD); and the qualitative variables for the simple and percentage frequency. For the comparison between the means of the scale factors according to the predictor variables (demographic and occupational), it was initially verified the normality of the distribution of the variables through the Kolmogorov Smirnov test, finding a non-normal distribution (p < 0.05). Then, the non-parametric tests U of Mann-Whitney (for variables with two categories) and Kruskal-Wallis (for variables with more than two categories) were applied, and multiple comparisons were made by the Conover Inman test. For all analyses, p < 0.05 was fixed.

RESULTS
Participated in the study, 167 nursing professionals, with average age of 37. With the application of NAQ-R, it was found that 33% of the participants reported being harassed at work, and 3.0% felt harassed several times a week or almost daily. Table 1 gathers the averages of the NAQ-R factors and its general average.
It is noted that the highest averages make up Factors 3 (Personal and Professional Disqualification) and 1 (Work-Related Harassment), respectively, and these are pointed out as the most common forms of moral harassment perceived among study participants.
As for the mean comparison test, it was found that all showed significant differences between them (p = 0.000). Table 2 shows the frequency of professionals who reported weekly or daily harassment in each item of the NAQ-R.
The most mentioned forms of harassment involved: being forced to perform activities at a level below professional capacity (16.2%), receiving excessive monitoring of work (15.6%), being pressured not to claim rights (13.2%), receiving tasks with impossible or irrational goals or deadlines (11.4%) and having gossip or rumors spread about the professional (11.4%). Table 3 shows comparisons of the averages of the predictor variables and of the NAQ-R factors.
The following predictors were associated with moral harassment: age group, time working in the job, time in the unit, type of employment relationship and sector. As for age, there was a significant statistical difference in the averages for Factors 1, 2 and 3, demonstrating that younger professionals (up to 30 years of age) are more likely to be targeted by work-related harassment, personal harassment and personal and professional disqualification than older professionals. In the same way, professionals with less time working in the job (up to 4 years) were more harassed than those trained longer ago. As for the time spent working in the unit, those with 11 to 20 years of service had a higher average than those with less time in the work sector. Finally, the type of unit also showed significant association with moral harassment, being this more experienced in the emergency services and ICU. It also draws attention to the cooperative members (professionals on temporary contracts) being more victims of work-related harassment (Factor 1) than public servants, to the detriment of personal harassment (Factor 2), more referred to by the latter.

DISCUSSION
The profile of the participants is similar to that already described in other researches carried out with nursing professionals in different clinical contexts. In general, female workers, younger and with little time of service feel more harassed (15,(18)(19)(20) .
Regarding the employment relationship, the fact that more than half of the sample works through cooperatives reflects the problem of precarious ways of working in Brazilian health services. This has the consequence of the lack of job security and the compromising of work relations and patient safety (21) .
As for the employment relationship, the fact that more than half of the sample works through cooperatives reflects the problem of precarious ways of working in Brazilian health services. This has the consequence of the lack of job security and the compromise of work relationships and patient safety.
It is important to emphasize that the dimensions of the precariousness of nursing work involve working conditions, the intensity of organization of the work process and the management of this process. In the latter, the existence of conflicts, embarrassment,  discrimination and workplace violence are discussed. This happens in a context that points out omissions of public power in the maintenance of health services, such as: bad remuneration, absence of public contests and increase of outsourcing of labor force (22) .
Considering the results of the application of NAQ-R (Table  1), the predominant forms of moral harassment (personal and professional disqualification and work-related harassment) show the most characteristic behaviors of this phenomenon in critical units, as already observed in other studies: professionals suffering pressure, negative criticism, in such a way that their performance is impaired at work; and, moreover, feeling that both their work and their opinion are not taken into account (2,8,10,(15)(16) .
Researchers emphasize the clear prevalence of descending moral harassment types among nurses and nursing technicians, being common the humiliations, embarrassment and persecutions in a repetitive way in the work environment, which destabilize the physical and, mainly, emotional balance of the victim, besides altering the relationship with co-workers and with other individuals (8) .
The perpetrators of moral harassment are often professionals with more time on the job, job stability and recognized experience. In general, they adopt attitudes such as delegating excessive activities to subordinates, acting rudely, using abusive language, throwing objects or instruments towards others, generating gossip or conflicts among team members, demonstrating discriminatory attitudes, retaining information about patients or institutional decisions to harm others, to threaten, among other disruptive behaviors (2) . The impact on the time offered to direct nursing care is remarkable, harming the patient, who has no involvement with the situation involving the professionals.
Regarding the abuse of power, the pressure of leadership and the ambiguity of roles, it must be considered that such rigid and traditional hierarchical structures, associated with the indetermination of work processes, favor the emergence of increasingly authoritarian nurses and managers with little or no recognition and appreciation by workers and team members (2) .
Another example of psychological aggression perceived in health work environments concerns gossip, whose negative side constitutes a dark side of interpersonal relationships, with the capacity to produce elitism, exclusion of individuals, personal and institutional defamation, in addition to systemic abuses, negligence and damage to patients. This type of communication is considered a reflection of several organizational problems that are often silenced and unmanaged (23) .
In the UK and elsewhere, bullying involves intentional, personal, repeated, humiliating and isolating psychological violence by co-workers. The main perpetrators are nurses in higher positions than those who are bullied and colleagues who are fixed team members. Those who are likely to be bullied are students and new staff (10) , as we also found in this study.
Despite the real evidence of moral harassment in the Nursing practice environment, there is still a wide recognition that, although these mistreatments and disrespects are perceived and named as violence, they are often considered necessary conduct for the work at some times, as part of the exercise of professional authority (24) . Therefore, a systemic and responsible approach to these negative acts is necessary, especially on the part of the area's leaders, taking care not to allow them to become impregnated in the organizational culture.
Workplace violence can cause anguish and depression, causing up to 25% of intimidated people to leave their jobs or profession, in addition to having a direct impact on patient care. The silence together with the lack of attitudes and management by managers and colleagues allow this behavior to continue. Zero tolerance and the treatment of this behavior in a clear and quick way by the managers should be instigated (10) .
Thus, it is believed that hostility in the work environment is perceived in a space where inadequate working conditions and low job satisfaction are frequent, which reflects in losses in interpersonal relationships, in conflict resolution, in the communication process and, consequently, on the rise of authoritarian and excluding professionals (25) .
Finally, analyzing the predictors of moral harassment identified in this study (Table 3) (young professionals, recent graduates, or those with longer working time in the unit and working in ICU and emergency), it is possible to perceive the magnitude of the impact of such problem on the career and emotional health of these professionals, even leading them to leave the profession.
Systematic review study on the use of NAQ-R to evaluate bullying among nurses corroborates the findings of this study (15) . The data indicated that participants reported bullying at prevalence of 17% to 94% (19 studies), with age, educational level and years of work experience as the most frequently associated variables, confirming the results of this research.
In New Zealand, there was a general prevalence of 38% of moral harassment at work by 1,759 health professionals (doctors and dentists), measured by NAQ-R (at least one negative act weekly or daily), 37.2% self-reported and 67.5% witnessed. There were significant differences in the harassment rates by specialty (p=0.001), with emergency medicine reporting the highest prevalence (47.9%). Harassment was significantly associated with increased demand for work and less support from colleagues and managers (p=0.001). The consequences were varied, affecting work environments, personal well-being and the subjective quality of patient care (16) .
In another study that evaluated the intentions of rotation, as well as the prevalence and frequency of incivility behaviors in the perspective of 170 health professionals, also with the NAQ-R, it was evidenced the exposure to incivility more frequent in the nursing team than among other health professionals. However, no significant correlations were found between exposure to uncivil behavior and selected demographic variables, suggesting that exposure does not depend on age, race, unit type or educational level. These data differ from those found in our study (17) .
Moreover, it is important to emphasize that the nursing workers, in comparison to the other health workers, can suffer more moral harassment, independently of age, race, place of performance, time of formation and performance. The history of the category explains this reality and shows a profession that has always experienced internal and external conflicts related to power and authority.
Researching the past of nursing, it is understood that the difference does not accept and the unresolved conflict is what 8 of Predictors of moral harassment in nursing work in critical care units Sousa LS, Oliveira RM, Santiago JCS, Bandeira ES, Brito YCF, Alves HFA, et al. most corrupts relationships within the category. The thesis is that the first nurses were oppressed by a dominant patriarchal system, resentful army medical teams and internal struggles for power, payment, influence and resources (26) . Such struggles were propagated and perpetuated in the different scenarios of training and performance of nurses over the years, reflecting negatively on their work processes until today.
In Brazil, when investigating practices of intra-team psychological violence, in the relationships between patients, escorts and other professionals with the nursing workers of the public hospital network, a study indicated that they were relatively young, predominantly female and with considerable professional experience. Verbal aggression was the most frequent subtype of psychological violence (95%; 84), followed by moral harassment (27%; 24). The emergency (51%; 45) was the sector of greatest occurrence; patients (60%; 53), the main aggressors; nurses (76%; 19) suffered more violence, being the majority female, young and little experienced (18) . This association was also tested in the present research, having been observed the inversely proportional relationship between training time and being harassed; and directly proportional between experience time and being harassed.
Other researchers have proven most of the intensive care unit nursing staff to be victims and witnesses of moral harassment behavior, with negative repercussions on job satisfaction and performance, and is also the cause of constant employee turnover (27) .
In a Brazilian study with 112 nurses who work in Primary Health Care, workplace violence comprised verbal aggression (67%) and psychological harassment (bullying -27.1%). Team leaders (78.3%) and other health professionals (41.7%) were perpetrators, besides patients. Occupational violence was not statistically associated with gender, professional experience, experience in Primary Health Care, weekly work schedule or work shift. The authors concluded that occupational violence is highly prevalent among Brazilian nurses working in Primary Health Care and that the lack of safety at the workplace is the main risk factor associated with (20) .
At the institutional level, it is necessary to conceive a culture of health and safety through a shared and sustained commitment of employers and employees, in three levels of action, in order to reduce the vulnerability of the work environment. These three levels include primary, secondary and tertiary prevention strategies. Primary prevention involves education and other measures to identify and reduce vulnerabilities in order to prevent the occurrence of workplace violence. Primary prevention initiatives also aim at improving interpersonal and interprofessional relations. Secondary intervention strategies are designed to reduce harm once an incident of workplace violence has begun. Tertiary strategies, on the other hand, aim to reduce the consequences associated with such an event. Although specified for nurses and employers, recommendations may also be relevant to other health care professionals and stakeholders who collaborate to create and maintain a safe and healthy interprofessional work environment (28) .
Managers have important roles in mitigating the problem. Aspects that are determinant in the occurrence of the diseases, such as organizational characteristics and the health work process, should be included in the investigations, thus contributing to a greater understanding of violence in health work and to the search for strategies to combat it (5) .
Considering the work in ICU and emergency units, these sectors have exhaustive workloads, either due to the need for agility in the care or the critical character of the patients. These are sectors that exploit great technical and scientific capacity of the professional, counting on protocols and, many times, an exacerbated supervision. Still, in face of the imminent needs, systemic problems leap to the eye.
This context is also part of the work of the nurse in the emergency units (29)(30) . These factors contribute to harassment behaviors, since the dichotomy between need and reality generates great disquiet, dissatisfaction and often retaliatory attitudes.
Failure to address the problem of harassment also limits the contribution that nurses can make to the health service and prevents future nurses from entering or remaining in the profession, as well as harming and belittling countless good people who want to make a difference and care for others (26) .
Thus, early identification in the professional practice environment can offer subsidies to health service managers for the development of prevention and control programs of this behavior in health institutions (6) . Efforts to combat bullying at work among nurses and staff need to include training in legitimate methods of performance analysis, workshops on how to interact with various co-workers, and examining how nursing teaching practices contribute to the perpetuation of bullying in clinical settings (9) .
It is known that the positive senses attributed to work favor the confrontation of degrading circumstances originated by moral harassment (31) . Therefore, it is necessary to stimulate the construction of healthy relationships among co-workers in critical units, favoring positive and resilient practice environments.
Promoting strategies for coping with moral harassment in Nursing work, which focus on the needs of the institution and the professionals, depends on the cohesion between the workers and the organization as a whole (25) .

Study limitations
Some study limitations were the scope of the total estimated sample and the fear that workers showed in participating, for fear of retaliation from colleagues or management, although it was explained about ensuring anonymity and confidentiality of data. A larger sample could highlight more critical and relevant points for a future intervention.

Contributions to the Nursing, Health or Public Health Area
Assessing moral harassment in the daily work of the nursing team in critical units encourages nurses, managers, researchers, students and health leaders to learn more about the related factors, the most frequent behaviors and the consequences of this phenomenon for everyone involved. Thus, there are subsidies to plan strategies for permanent education and performance evaluation more directed to the effective management of each case, leading professionals to review their conflict resolution skills, as well as their roles and professional values.

CONCLUSIONS
There was a 33% prevalence of self-perception of moral harassment among nursing professionals in critical care units, 3% weekly or daily and 30% rarely or once in a while. The predictors of the moral harassment were age (under 30 years), time working in the job (under 5 years) and time in the unit (more than 10 years), type of employment relationship (cooperative) and work unit (ICU and emergency). Thus, the professionals who most feel morally harassed at work in critical care units are the newly graduated, those with longer working time in the unit, those with temporary contracts (cooperative) and those working in emergency and ICU.