258 J Epidemiol Community Health 2004;58:258–259 J Epidemiol Community Health: first published as on 13 February 2004. Downloaded from PostScript......

‘‘social isolation’’, ‘‘destabilisation’’. As being always at higher risk of the mentioned LETTER Hirigoyen states, bullying can be behaviours, health problems than non-bullied ones. No words, acts, gestures, writings that can affect noticeable differences could be observed in the personality, the dignity, or physical or terms of gender, nationality, education, pro- psychological integrity.6 If you have a burning desire to respond to a fessional qualification, or health related We conducted a longitudinal survey in behaviours (smoking, alcohol consumption, paper published in JECH, why not make use Belgium (two measures, one year interval) and alcohol dependence). of our ‘‘rapid response’’ option? on psychosocial factors at work and their Prospective data analysis permits a better Log on to our web site (www.jech.com), 7 influence on the worker’s health. With the comprehension of the pathways and relations find the paper that interests you, and send working hypothesis that bullying, as other between risk factors and health outcomes. your response via email by clicking on the stressors, could have a negative impact on We have conducted logistic regressions on ‘‘eLetters’’ option in the box at the top right health. A second hypothesis was that in an these prospective data (paired samples; each hand corner. unstable work environment, besides being of the 1030 participant workers answered Providing it isn’t libellous or obscene, it exposed to higher stress, bullying could be an twice to the same questionnaire, in a one year will be posted within seven days. You can additional stressor. interval). The goodness of fit of the model retrieve it by clicking on ‘‘read eLetters’’ on The objective was double: to assess the was tested with the Homer and Lemeshow our homepage. bullying prevalence—including its compari- test. The editors will decide as before whether son in different work environments—and to After having controlled for sex, age, educa- to also publish it in a future paper issue. study its eventual negative effects on the tion, work instability, and two stress mod- victims. els—that is, the Karasek’ job demands Bullying is one of the negative environ- control-social support (JDC-S) and the mental characteristics in the workplace that Siegrist’ efforts rewards imbalance (ERI) Bullying, workers’ health, and can have an impact on the worker’s health; models,910 we found significant relations labour instability we measured it through a self administered between bullying (measured at the first postal questionnaire submitted to all workers point) and repeated absenteeism (measured Violence and bullying in the workplace seem in two enterprises (one stable and one at the second point, one year later). Bullied to be an increasing phenomenon in Europe, unstable). Work instability is understood as workers have a double risk of repeated even though self reported bullying shows enterprise instability—that is, recent or 1 absences than non-bullied ones (OR:2.3; wide variations across nations. Bullying in future merge, downsizing, or restructuring. 95% CI 1.4 to 4.0). We can therefore conclude the workplace is certainly not a recent The two workplaces are from the tertiary that repeated absenteeism and bullying are phenomenon and is probably inherent to sector (service industry). The asset of such more frequent in an unstable work environ- many human relations and organisations. design survey is the possibility to study ment, as we can also observe that bullied However, its rising importance could be bullying on large samples and to go beyond workers have a much higher risk of absentee- related to the global deterioration of working individual case report, which is often the case ism than the others. conditions.12Thus the European Parliament,3 in this subject matter. Another advantage is Undoubtedly, self reported measure of besides an increasing number of countries the longitudinal protocol, which permits bullying and of health outcomes are subject has legislated to cut down on and prevent more interpretations than cross sectional 4 to limitation and the validation of the first these occurrences. design. could be questioned. Other methodologies, http://jech.bmj.com/ Notwithstanding the importance of these Identical questionnaires were sent to all like in depth interviews or focus groups could legal measures, we are faced with crucial workers twice: in 2000 and in 2001. Global launch other figures and information. problems. At the working population level, participation (completed questionnaires) was However, few studies have investigated this the crux of the matter is that we are facing about 40%, and 1030 workers (two thirds new psychosocial dimension in large scale two unresolved questions: how can we women) participated to the two measures; populations of workers, and to our knowl- measure bullying behaviours and its deleter- 549 in the stable enterprise, 481 in the edge, none in Belgium. However, some ious effects on the victim? What are the unstable one. Participants were representa- studies like the ones of Kivima¨ki et al or dimensions of interest? Which methods can tive of the total worker’s population for the be used? The wide variations observed across available criteria: age, sex, department/ser- Voss et al—even though they show large surveys and/or places is probably the expres- vice. Two among the five dimensions of discrepancies in the bullying prevalence— on September 26, 2021 by guest. Protected copyright. have also found clear relations between sion of the difficulty in measuring bullying, Quine’s instrument were used to measure 11 12 and moreover, of the different interpretations bullying at work: namely isolation and workplace bullying and absenteeism. from one context to another one. The second destabilisation.8 The upper quartile of a score It is essential to denounce bullying, which question is what impact has bullying on the based on all those seven items defined has negative consequences not only for the worker’s health and wellbeing? arbitrarily the category of ‘‘presently bullied’’. workers but also at the managerial level: According to Einarsen, bullying is charac- Self reported repeated absenteeism was used absenteeism, turn over, loss of productivity. terised by ‘‘repeated and enduring negative as health indicator, defined as at least three The consideration of this practice is not to be acts’’.5 Its manifestations are described in the sick leaves between the two measures. placed at the individual level only, but can be literature as forms of ‘‘intimidation’’, ‘‘phy- In the unstable work environment, there is considered as a collective nuisance, and, in sical violence’’, ‘‘discrimination’’, ‘‘threats’’, a significant increase in the proportion of some cases, as a workplace strategy directed bullied workers between the two measures to individuals. (from 27% to 33%, p,0.05%, Mc Nemar test Further researches are crucial to explore for paired samples). The two most often cited and validate the methodological aspects of Key points items of bullying are ‘‘a constant under- bullying measurement as well as the influ- valuing of my efforts’’ and ‘‘withholding ences of bullying on different health out- N Bullying in the workplace seems to be an necessary information’’. comes by way of well conducted prospective increasing phenomenon in Belgium Repeated absenteeism remains stable in epidemiological studies. It is also too early to the two firms, but its prevalence is higher in evaluate the effects of the new regulation on N In this survey bullying is a significant the unstable firm, for the two measures the prevention of bullying. Explanations and independent predictor of sick leave (around 11% in the stable, 16% in the prevention could be sought in a more global, N More research is needed to validate unstable one). organisational, and economic environment to instruments that measure bullying Cross sectional data (at the first measure prevent this from happening, and not at the N Research is needed to evaluate the effects and at the second one) show very important individual level solely, even if victims must be of the new legislation relations between bullying and different helped and supported. Unstable, fragile N More should be done at the workplace health outcomes (subjective health status, workplaces—nowadays such a common rea- level to prevent it depression, anxiety, somatisation, neuroti- lity for so many workers—deserve special cism, or chronic fatigue). Bullied workers attention.

www.jech.com PostScript 259

I M Godin 12 Kivima¨ki M, Elovainio M, Vahtera J. Workplace The book will not cheer those who wish to bullying and sickness absence in hospital staff. see government policies evidence based. School of Public Health, Universite Libre de Bruxelles, J Epidemiol Community Health: first published as on 13 February 2004. Downloaded from Occupational and Environmental Medicine CP 596 808 Route de Lennik, Brussels 1070, Belgium; Robust evidence for the effectiveness of many 2000;57:656–60. [email protected] of the policies is missing (an absence some- what glossed over in the Acheson report), Funding: this study is supported by the Federal Office while the sheer number of initiatives means for Scientific, Technical and Cultural Affairs, Belgium. BOOK REVIEW that the effectiveness of any one of them will be impossible to evaluate. It will not be References Tackling health inequalities since the possible to assess whether the policies as an Acheson Inquiry aggregate have reduced health inequalities 1 Paoli P, Merllie´D.Third European survey on for some years to come. Inevitably, therefore, working conditions 2000. Dublin: European this book’s early intelligence is both very E Exworthy, M Stuart, D Blane, M Marmot. Foundation for the Improvement of Living and welcome and rather tantalising. Working Conditions, 2001. (Pp 68; £14.95). The Policy Press, Bristol, 2 Ferrie J, Marmot M, Griffiths J, Ziglio E, eds. 2003. ISBN 1-86134-504-6 S Abbott Labour market changes and job insecurity: a challenge for social welfare and A decade ago, the then British government health promotion. Denmark: WHO Regional shunned the very mention of the phrase CORRECTION Publications, 1999. ‘‘health inequalities’’. Its successor, by con- 3 European Parliament. Health and safety at work trast, began by commissioning the Acheson Council Directive 89/391/EEC on the Inquiry (1998) and has since introduced a doi: 10.1136/jech.2003.009506corr1 introduction of measures to encourage large number of policy initiatives across a improvements in the safety and health of workers number of government departments to at work: OJ L 183, 29.6.1989. 89/391/CEE. 20- reduce health inequalities. This useful book An authors’ error occurred in this paper by Dr 9-2001. 4 Perima¨ki-Dietrich R. Violence at the workplace. reviews in detail how three areas of policy Sundquist and others (2004;58:71–7). TUTB Newsletter 2002;19–20:20–2. highlighted by Acheson have been imple- Page 71, left column, second paragraph, 5 Einarsen S. Harassment and bullying at work; a mented so far (tax and benefit reform, line 8, a reference should be inserted Merlo J, review of the Scandinavian approach. performance management, and transport). Lynch J W, Yang M, et al. Effect of neighbor- Aggression and Violent Behavior It is a short book that nevertheless touches hood social participation on individual use of 2000;5:379–401. on some large themes. To take an example: it hormone replacement therapy and antihy- 6 Hirigoyen M-F. Le harce`lement moral. La violence is easier for governments to target identifi- pertensive medication: a multilevel analysis. perverse au quotidien. Paris: La De´couverte et able deprived neighbourhoods than to target Am J Epidemiol 2003;157:774–83. Page 71, Syros, 1998. deprived people, many of whom live in right column, first paragraph, line 8, 7 Siegrist J, Starke D, Chandola T, et al. affluent or middle income areas. The book exchange the existing reference (20) with Measurement of effort-reward imbalance at work: European comparisons. Soc Sci Med (in press). acknowledges that such an approach to reference Merlo J, Asplund K, Lynch J, et al. 8 Quine L. Workplace bullying in NHS community health inequalities is likely to fail in the long Population effects on individual systolic trust: staff questionnaire survey. BMJ run (and that governments are more con- blood pressure—a multilevel analysis of 1999;318:228–32. cerned with the short run). WHO MONICA project. 43rd Annual confer- 9 Karasek R. Job demands, job decision latitude, Similarly, the authors show how the ence on cardiovascular disease epidemiology and mental strain: implications for job redesign. government’s approach seems to be based and prevention in association with the Administrative Science Quarterly on economics rather than social justice: poor council on nutrition, physical activity and 1979;24:285–308. health is a poor use of human resources, and metabolism, 2002. Page 75, right column, last 10 Siegrist J. Adverse health effects of high effort - policies focus on employment and the future paragraph, lines 5 and 12, a reference should low reward conditions at work. Journal of workforce (children) rather than, say, the be inserted Merlo J, Lynch JW, Yang M, et al. Occupational Health Psychology 1996;1:27–43. quality of life of older people. The authors Effect of neighborhood social participation on 11 Voss M, Floderus B, Diderichsen F. Physical, http://jech.bmj.com/ psychosocial, and organisational factors relative might have been more explicit about the individual use of hormone replacement ther- to sickness absence: a study based on Sweden government’s failure to promote policies such apy and antihypertensive medication: a mul- Post. Occupational and Environmental Medicine as progressive taxation and improving public tilevel analysis. Am J Epidemiol 2003;157:774– 2001;58:178–84. transport. 83. on September 26, 2021 by guest. Protected copyright.

www.jech.com