Search tips
Search criteria 


Logo of oenvmedOccupational and Environmental MedicineVisit this articleSubmit a manuscriptReceive email alertsContact usBMJ
Occup Environ Med. 2007 December; 64(12): e24.
PMCID: PMC2095345


113 The effects of workplace incivility on psychological distress and health

L. E. Langlois1, H. S. Shannon1, L. Griffith1, T. Haines1, L. M. Cortina2, S. Geldart3. 1McMaster University; 2University of Michigan; 3Wilfrid Laurier University

ObjectivesThis research examines the influence of incivility and other work‐related factors on psychological distress and health.

MethodsA nation‐wide cross‐sectional study of workers at a large Canadian organisation was completed as the second phase of a two‐part project. Phase I involved focus groups to identify work‐related health concerns (eg, lack of respect for others, burnout). In phase II, employees were randomly sampled from 12 workplaces stratified by job type. All sampled workers (n = 1968) were sent anonymous questionnaires that had previously been pilot tested among a small group of workers from the organisation. To maximise response, the Dillman method was used. After the third mailing, the response rate of completed questionnaires was 49.5% (n = 953). We used multivariable regression to examine the amount of variance explained by incivility after adjusting for demographic and psychosocial job factors for several work‐related outcomes. Incivility was measured using the Workplace Incivility Scale (Cortina et al. 2001).

ResultsThe respondents were 60% male, with a mean age of 47.2 years (SD 8.3). Employees primarily worked full‐time (75%), on day shifts (73%), and many interacted with customers (54%). Most workers had been with the organisation for some time, with a mean duration of employment of 17.9 years (SD 10.4). Incivility was common in the sample, with 84% of workers experiencing incivility to at least some degree. Demographic and job characteristics accounted for only modest amounts of the variation in the psychological and health outcomes, with adjusted R2 values ranging from 0.0 to 8.3%. The addition of supervisor and co‐worker incivility to the regression model improved the explained variance, with increases in adjusted R2 values as high as 28%. Even after controlling for demographic variables, job characteristics, job demand, job control and work social support, incivility increased the explanatory power of the regression model by up to 11%. Of the examined outcomes, the explanatory power of incivility was greatest for burnout, anxiety, depression and hostility.

ConclusionIncivility can explain a significant amount of variance in several health‐related outcomes even after adjusting for other workplace exposures. It is therefore worthwhile to consider incivility in the examination of the relationship between work, health and well‐being.

Key wordspsychosocial exposures; incivility; mental health

114 Victims' responses to different categories of workplace violence in a psychiatric hospital: a questionnaire survey

W‐C. Chen1, H‐J. Chiu2, H‐G. Hwu3, J‐D. Wang1. 1Institutes of Occupational Medicine and Industrial Hygiene, College of Public Health, National Taiwan University, Taiwan; 2Yu‐Li Hospital, Department of Health, Executive Yuan, Taiwan; 3Department of Psychiatry, National Taiwan University Hospital and College of Medicine, National Taiwan University, Taiwan

ObjectivesThis study explored victims' responses (primary care staff at a psychiatric hospital in Taiwan), how attackers (mostly chronic schizophrenics) were treated, and the reasons for not reporting all kinds of workplace violence, including physical violence (PV), verbal abuse (VA), bullying/mobbing (BM), sexual harassment (SH) and racial harassment (RH).

MethodsA questionnaire developed by ILO/ICN/WHO/PSI was used as a cross‐sectional survey. A total of 222 out of 231 staff completed valid surveys. Bonferroni multiple comparisons were used to determine whether there were significant differences between the different categories of violence.

ResultsOnly 31% of the PV incidents and less than 10% of the other categories of violence were formally reported. VA victims most frequently took no action and tried to pretend it never happened. More SH victims reported the incident to senior staff. More PV victims defended themselves, and their attackers were more likely to be injected with medication. Other treatments included physical restraint, seclusion, and even electronic convulsion therapy. As for reasons for not reporting, more VA victims considered the incident not to be important enough. Other reasons were “fear of negative consequences” especially for BM victims, and “shame” for SH victims.

ConclusionEstablishing a reliable reporting system for workplace violence and eliminating negative feelings about reporting is necessary. A careful examination of the mental status of patients and intensive treatment to ameliorate patients' psychotic symptoms is supposed to be necessary to reduce violence in the psychiatric field.

Key wordsworkplace violence; victim; psychiatry

115 Informal social status among coworkers and risk of acute injury and physical assault in a long‐term healthcare facility

D. J. M. Myers. Duke University

ObjectivesThe goal of this study was to explore the hypothesis that one's position in the informal social hierarchy (within job titles) in a long‐term care workplace is associated with the incidence of work‐related acute injury and physical assault. Previous research generated the hypotheses that, in this population of healthcare workers, social status would generally increase risk from the low to moderate levels then decrease risk only at the highest levels. This stands in contrast to most hypotheses that predict an inverse relationship between status and health effects.

MethodsSurvey data were used to measure social status, operationalised as the number of coworkers who would approach the individual for advice about work‐related matters. Seven months of self‐reported injury and assault records and administrative data were used. The data include 7324 person‐shifts recorded on 192 individuals (54 nurses and 138 CNAs); 67 injuries and 85 assaults were reported. Logistic regression was used to model the data and non‐parametric smoothing techniques were used to graphically represent associations. Although “advice ties” used to score social status could come from either nurses or CNAs, associations with outcomes were analysed separately by job title to account for mean differences in exposure levels and to examine the effect of informal social status within levels of the workplace's formal hierarchy.

ResultsRegression models and graphical representations showed the predicted curvilinear associations. The strongest effect of the social status variables was found among nurses with respect to injury outcomes (first‐order term OR 2.63, 95% CI 0.84 to 8.29, squared term OR 0.88, 95% CI 0.76 to 1.01). Results were adjusted for floor‐ and shift‐level physical demands (lifting and resident combativeness), sex and job tenure. Although the regression results in this small study were of marginal statistical significance, all coefficients were in the hypothesised directions.

ConclusionResults suggest that one's position in the informal social hierarchy, within job title, may be associated with acute injury and physical assault risk among healthcare workers. In addition, the socio‐cultural context must be considered in studies of the impact of social status on health, as it may determine non‐linear associations not usually hypothesised in these studies.

Key wordsworkplace injuries; social status; healthcare workers

116 Determinants of interpersonal conflicts at work: results from a prospective cohort study

L. De Raeve, N. W. H. Jansen, P. A. van den Brandt, R. M. Vasse, I. Kant. Maastricht University

ObjectivesThe aim of this study was to examine the influence of work and health related risk factors and personality in the aetiology of interpersonal conflicts at work.

MethodsLongitudinal data from the Maastricht Cohort Study on “Fatigue at Work” (n = 9655) were used. Gender stratified Cox regression analyses, adjusted for age, education and living alone were conducted to examine the role of several potential risk factors at baseline in the onset of a conflict with co‐workers or supervisors at 1‐year follow‐up.

ResultsAt baseline, the prevalence of a conflict was 7.4% for co‐worker conflict and 10.2% for conflict with supervisors. After excluding respondents experiencing a conflict at baseline, about 4% of the study population reported a co‐worker conflict, while 5.3% reported a supervisor conflict at 1‐year follow‐up. The onset of interpersonal conflicts with co‐workers in men was most strongly associated with high emotional demands (RR 2.21, 95% CI 1.72 to 2.86) and poor overall job satisfaction (RR 2.04, 95% CI 1.48 to 2.80). In women, poor overall job satisfaction (RR 2.40, 95% CI 1.33 to 4.31) and having a supervisory function (RR 1.84, 95% CI 1.03 to 3.28) showed the strongest associations with the onset of a co‐worker conflict. The onset of a supervisor conflict in men was most strongly related to poor overall job satisfaction (RR 2.44, 95% CI 1.88 to 3.18) and low supervisor social support (RR 2.44, 95% CI 1.96 to 3.05). In women, poor general health (RR 2.85, 95% CI 1.75 to 4.63), the presence of a long‐term disease (RR 2.38, 95% CI 1.51 to 3.75) and sickness absence (RR 2.26, 95% CI 1.28 to 3.97) were important risk factors for the onset of a supervisor conflict. Personality factors were only marginally related to the onset of interpersonal conflicts at work in both men and women.

ConclusionThis study showed that several work and health related factors are important determinants in the onset of interpersonal conflicts at work. Determinants were slightly different for conflicts with co‐workers and supervisors and varied among men and women. Given the rather serious consequences of interpersonal conflicts at work for health and well‐being, the results of this study might serve as a starting point for designing effective prevention and intervention strategies in the workplace.

Key wordsco‐worker conflict; supervisor conflict; prospective cohort study

117 Characterising workplace violence among healthcare workers in British Columbia

R. Kling1, A. Yassi2, C. Lovato2, E. Smailes2, M. Koehoorn2. 1School of Occupational and Environmental Hygiene, University of British Columbia; 2University of British Columbia

ObjectivesApproximately 40% of all violence‐related workers' compensation claims in British Columbia occur among health care workers, although these workers make up less than 5% of the workforce. The purpose of this study was to investigate which segments of the healthcare population in British Columbia are at a higher risk for workplace violence.

MethodsAll occupational incident reports of violence in British Columbia hospitals between 2003 and 2005 were identified through the use of the Workplace Health Indicator Tracking and Evaluation (WHITE) database. The WHITE database also provided data on total hours of work. Multivariable Poisson regression models, adjusted for age, gender, work status, years spent in current occupation and health authority, were used to calculate the relative risk of work characteristics (occupation, health care sector, facility size, and work unit‐entered in separate models) associated with violent incidents.

ResultsIn the final multivariable models, occupations found to be at higher risk for violence, compared to management and maintenance workers, included care aides (RR 10.05), licensed practical nurses (RR 8.64), registered nurses (RR 6.45), and health services workers (RR 3.76). Healthcare workers in the long‐term care sector (RR 3.02) and acute care sector (RR 1.77) were found to be at an increased risk for violence compared to the community and public health sector. The psychiatric (RR 6.29) and paediatric (RR 2.22) departments in acute care hospitals were found to be at an increased risk compared to human resources and finance department. Within healthcare facilities grouped by similar size and patient populations, very small hospitals (RR 6.58), extended and elder care facilities (RR 4.37) and mental health and addictions facilities (RR 4.14) were found to be at the highest increased risk for violence compared to health centres and health units.

ConclusionIf high‐risk segments of the healthcare population can be identified, prevention efforts can be specifically targeted at these workers. This study indicates that further research is needed to assess what the specific characteristics are that put care aides, paediatric departments, and very small hospitals at an increased risk for violence so that prevention efforts can be targeted towards these areas.

Key wordsviolence; healthcare; risk factors

Articles from Occupational and Environmental Medicine are provided here courtesy of BMJ Publishing Group