Article Text

PDF

Bullying, workers’ health, and labour instability
  1. I M Godin
  1. School of Public Health, Universite Libre de Bruxelles, CP 596 808 Route de Lennik, Brussels 1070, Belgium; igodinulb.ac.be

    Statistics from Altmetric.com

    Violence and bullying in the workplace seem to be an increasing phenomenon in Europe, even though self reported bullying shows wide variations across nations.1 Bullying in the workplace is certainly not a recent phenomenon and is probably inherent to many human relations and organisations. However, its rising importance could be related to the global deterioration of working conditions.1,2 Thus the European Parliament,3 besides an increasing number of countries has legislated to cut down on and prevent these occurrences.4

    Notwithstanding the importance of these legal measures, we are faced with crucial problems. At the working population level, the crux of the matter is that we are facing two unresolved questions: how can we measure bullying behaviours and its deleterious effects on the victim? What are the dimensions of interest? Which methods can be used? The wide variations observed across surveys and/or places is probably the expression of the difficulty in measuring bullying, and moreover, of the different interpretations from one context to another one. The second question is what impact has bullying on the worker’s health and wellbeing?

    According to Einarsen, bullying is characterised by “repeated and enduring negative acts”.5 Its manifestations are described in the literature as forms of “intimidation”, “physical violence”, “discrimination”, “threats”, “social isolation”, “destabilisation”. As Hirigoyen states, bullying can be behaviours, words, acts, gestures, writings that can affect the personality, the dignity, or physical or psychological integrity.6

    Key points

    • Bullying in the workplace seems to be an increasing phenomenon in Belgium

    • In this survey bullying is a significant independent predictor of sick leave

    • More research is needed to validate instruments that measure bullying

    • Research is needed to evaluate the effects of the new legislation

    • More should be done at the workplace level to prevent it

    We conducted a longitudinal survey in Belgium (two measures, one year interval) on psychosocial factors at work and their influence on the worker’s health.7 With the working hypothesis that bullying, as other stressors, could have a negative impact on health. A second hypothesis was that in an unstable work environment, besides being exposed to higher stress, bullying could be an additional stressor.

    The objective was double: to assess the bullying prevalence—including its comparison in different work environments—and to study its eventual negative effects on the victims.

    Bullying is one of the negative environmental characteristics in the workplace that can have an impact on the worker’s health; we measured it through a self administered postal questionnaire submitted to all workers in two enterprises (one stable and one unstable). Work instability is understood as enterprise instability—that is, recent or future merge, downsizing, or restructuring. The two workplaces are from the tertiary sector (service industry). The asset of such design survey is the possibility to study bullying on large samples and to go beyond individual case report, which is often the case in this subject matter. Another advantage is the longitudinal protocol, which permits more interpretations than cross sectional design.

    Identical questionnaires were sent to all workers twice: in 2000 and in 2001. Global participation (completed questionnaires) was about 40%, and 1030 workers (two thirds women) participated to the two measures; 549 in the stable enterprise, 481 in the unstable one. Participants were representative of the total worker’s population for the available criteria: age, sex, department/service. Two among the five dimensions of Quine’s instrument were used to measure bullying at work: namely isolation and destabilisation.8 The upper quartile of a score based on all those seven items defined arbitrarily the category of “presently bullied”. Self reported repeated absenteeism was used as health indicator, defined as at least three sick leaves between the two measures.

    In the unstable work environment, there is a significant increase in the proportion of bullied workers between the two measures (from 27% to 33%, p<0.05%, Mc Nemar test for paired samples). The two most often cited items of bullying are “a constant undervaluing of my efforts” and “withholding necessary information”.

    Repeated absenteeism remains stable in the two firms, but its prevalence is higher in the unstable firm, for the two measures (around 11% in the stable, 16% in the unstable one).

    Cross sectional data (at the first measure and at the second one) show very important relations between bullying and different health outcomes (subjective health status, depression, anxiety, somatisation, neuroticism, or chronic fatigue). Bullied workers being always at higher risk of the mentioned health problems than non-bullied ones. No noticeable differences could be observed in terms of gender, nationality, education, professional qualification, or health related behaviours (smoking, alcohol consumption, and alcohol dependence).

    Prospective data analysis permits a better comprehension of the pathways and relations between risk factors and health outcomes. We have conducted logistic regressions on these prospective data (paired samples; each of the 1030 participant workers answered twice to the same questionnaire, in a one year interval). The goodness of fit of the model was tested with the Homer and Lemeshow test.

    After having controlled for sex, age, education, work instability, and two stress models—that is, the Karasek’ job demands control-social support (JDC-S) and the Siegrist’ efforts rewards imbalance (ERI) models,9,10 we found significant relations between bullying (measured at the first point) and repeated absenteeism (measured at the second point, one year later). Bullied workers have a double risk of repeated absences than non-bullied ones (OR:2.3; 95% CI 1.4 to 4.0). We can therefore conclude that repeated absenteeism and bullying are more frequent in an unstable work environment, as we can also observe that bullied workers have a much higher risk of absenteeism than the others.

    Undoubtedly, self reported measure of bullying and of health outcomes are subject to limitation and the validation of the first could be questioned. Other methodologies, like in depth interviews or focus groups could launch other figures and information. However, few studies have investigated this new psychosocial dimension in large scale populations of workers, and to our knowledge, none in Belgium. However, some studies like the ones of Kivimäki et al or Voss et al—even though they show large discrepancies in the bullying prevalence—have also found clear relations between workplace bullying and absenteeism.11,12

    It is essential to denounce bullying, which has negative consequences not only for the workers but also at the managerial level: absenteeism, turn over, loss of productivity. The consideration of this practice is not to be placed at the individual level only, but can be considered as a collective nuisance, and, in some cases, as a workplace strategy directed to individuals.

    Further researches are crucial to explore and validate the methodological aspects of bullying measurement as well as the influences of bullying on different health outcomes by way of well conducted prospective epidemiological studies. It is also too early to evaluate the effects of the new regulation on the prevention of bullying. Explanations and prevention could be sought in a more global, organisational, and economic environment to prevent this from happening, and not at the individual level solely, even if victims must be helped and supported. Unstable, fragile workplaces—nowadays such a common reality for so many workers—deserve special attention.

    References

    View Abstract

    Footnotes

    • Funding: this study is supported by the Federal Office for Scientific, Technical and Cultural Affairs, Belgium.

    Request permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Linked Articles