UBC Faculty Research and Publications

Psychosocial and other working conditions in relation to body mass index in a representative sample of… Ostry, Aleck S; Radi, Samia; Louie, Amber M; LaMontagne, Anthony D Mar 2, 2006

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-12889_2005_Article_350.pdf [ 517.69kB ]
JSON: 52383-1.0223937.json
JSON-LD: 52383-1.0223937-ld.json
RDF/XML (Pretty): 52383-1.0223937-rdf.xml
RDF/JSON: 52383-1.0223937-rdf.json
Turtle: 52383-1.0223937-turtle.txt
N-Triples: 52383-1.0223937-rdf-ntriples.txt
Original Record: 52383-1.0223937-source.json
Full Text

Full Text

ralssBioMed CentBMC Public HealthOpen AcceResearch articlePsychosocial and other working conditions in relation to body mass index in a representative sample of Australian workersAleck S Ostry*1, Samia Radi2,3,4, Amber M Louie1 and Anthony D LaMontagne*4Address: 1Department of Health Care and Epidemiology, University of British Columbia, 5804 Fairview Avenue, Vancouver BC, V6T 1Z3, Canada, 2Department of Epidemiology and Preventive Medicine, Monash University, Melbourne, Australia, 3Unité INSERM 558, Toulouse, France and 4Centre for Health & Society, School of Population Health, University of Melbourne, 207 Bouverie Street, Melbourne VIC 3070, AustraliaEmail: Aleck S Ostry* - ostry@interchange.ubc.ca; Samia Radi - srad@primusonline.com.au; Amber M Louie - ammlouie@interchange.ubc.ca; Anthony D LaMontagne* - alamonta@unimelb.edu.au* Corresponding authors    AbstractBackground: The aim of the study was to examine the relationship between psychosocial andother working conditions and body-mass index (BMI) in a working population. This studycontributes to the approximately dozen investigations of job stress, which have demonstratedmixed positive and negative results in relation to obesity, overweight and BMI.Methods: A cross-sectional population-based survey was conducted among working Australiansin the state of Victoria. Participants were contacted by telephone from a random sample of phonebook listings. Information on body mass index was self-reported as were psychosocial workconditions assessed using the demand/control and effort/reward imbalance models. Other workingconditions measured included working hours, shift work, and physical demand. Separate linearregression analyses were undertaken for males and females, with adjustment for potentialconfounders.Results: A total of 1101 interviews (526 men and 575 women) were completed. Multivariatemodels (adjusted for socio-demographics) demonstrated no associations between job strain, asmeasured using the demand/control model, or ERI using the effort/reward imbalance model (afterfurther adjustment for over commitment) and BMI among men and women. Multivariate modelsdemonstrated a negative association between low reward and BMI among women. Among men,multivariate models demonstrated positive associations between high effort, high psychologicaldemand, long working hours and BMI and a negative association between high physical demand andBMI. After controlling for the effort/reward imbalance or the demand/control model, theassociation between physical demand and working longer hours and BMI remained.Conclusion: Among men and women the were differing patterns of both exposures topsychosocial working conditions and associations with BMI. Among men, working long hours waspositively associated with higher BMI and this association was partly independent of job stress.Among men physical demand was negatively associated with BMI and this association wasPublished: 02 March 2006BMC Public Health2006, 6:53 doi:10.1186/1471-2458-6-53Received: 08 July 2005Accepted: 02 March 2006This article is available from: http://www.biomedcentral.com/1471-2458/6/53© 2006Ostry et al; licensee BioMed Central Ltd.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.Page 1 of 8(page number not for citation purposes)independent of job stress.BMC Public Health 2006, 6:53 http://www.biomedcentral.com/1471-2458/6/53BackgroundThe prevalence of overweight and obesity has been high inmost industrialized nations, since the early 1950s; how-ever, this trend accelerated in the 1990s. Australia has notescaped this phenomenon [1]. The risk of cardiovasculardisease increases with overweight and obesity [2]. Accord-ing to the Australian National Health Survey conducted in2001, 16% of men and 17% of women (aged more than18) were obese, 42% of men and 25% of women wereoverweight, and 1% of men and 5% of women wereunderweight [1].An extensive literature has been published on job stressand cardiovascular disease (CVD), mainly among men[5]. This literature demonstrates that job strain and higheffort/low reward conditions predict CVD [6-8], but therelative contributions of direct and indirect mechanismsremain unclear. Some evidence suggests indirect effects ofpsychosocial and other work conditions on healththrough health behaviours [5].Indirect pathways may include effects of job stress onphysical activity, eating behaviours, and other behavioursthat may be related to BMI [9]. Previous studies havefound associations between working conditions andhealth behaviours, such as diet, physical activity and alco-hol consumption, which impact weight change [10-12].For example, in a cross-sectional analysis of Japaneseworkers (n = 6,759) job strain was associated with lowvegetable and high alcohol consumption [11]. Anothercross-sectional survey of American workers (n = 3,843)showed that job demands were positively associated withhigh fat intake in men, while decision latitude was posi-tively associated with physical activity in both men andwomen [12]. More recently, the Helsinki Health Study (n= 6,243) found that among women, mentally strenuouswork and high job control were associated with a healthydiet [10].The two most widely used instruments to measure occu-pational stress are Karasek's demand/control (DC) andSiegrist's effort/reward imbalance (ERI) models. Thedemand/control model focuses on task-level job charac-teristics. It postulates that psychological strain resultsfrom the interaction of job demands and job control, withthe combination of low control and high demands pro-ducing "job strain". In contrast, the effort/reward imbal-ance model includes personal characteristics of the worker(i.e., over commitment) and also conceptualizes andmeasures work conditions more broadly than thedemand/control model. It focuses on the reciprocity ofexchange at work where high costs/low gain conditions(i.e., high effort and low reward) are considered particu-While the demand/control and effort/reward imbalancemodels have been tested in relation to CVD outcomesthey have been less widely investigated in relation to riskfactors for CVD including body mass index (BMI) [12].Nonetheless, fourteen studies have been conducted usingjob stress to test for associations with body mass index[4,13-25]. The DC model was used in ten of these studies[12-16,20,21,23-25], two of which also used the ERImodel [13,14].Six of the studies with the DC model showed a positiveand statistically significant relationship with BMI [12-15,20,24] but the remaining four showed no association.Both studies with the ERI model showed a positive andstatistically significant relationship with BMI [13,14].And, two of the remaining four studies utilizing othermeasures of job strain showed positive and statisticallysignificant relationships with BMI [4,18].In these positive investigations with the DC and ERI mod-els, high job strain [13], low control [13,24] and high ERI[13] have been linked with increased BMI. Among thefour workplace studies of stress and obesity that did notuse the DC or ERI models, the results varied [4,22,26,27].House et al. [22], using data from the Tecumseh Commu-nity Health Study from 1967 to 1969, demonstrated neg-ative associations between occupational position and"pressures on the job" in relation to obesity for both menand women.In a study of 1,137 Swedish women, Rosmond and Bjorn-top [4] found that education, satisfaction with manage-ment, attempts to alter work situation" (i.e., a proxymeasure of participation and control), and physical exer-cise were all negatively associated with BMI.In a cohort study, Kornitzer and Kittel [26] found no asso-ciation between job stress and obesity. The stress measurethey used in this study of obesity was also tested in rela-tion to coronary heart disease and was also non-predic-tive. Conversely, Georges et al. [27] found a borderlinesignificant positive association (P= 0.06) between jobdemands and increased BMI. This survey also demon-strated that men with high job strain were more likely tohave a pattern of central body fat distribution then menwith low job strain [27].Non-psychosocial working conditions have been investi-gated in relation to body mass index also. Using data fromthe National Population Health Survey in Canada, Shields[28] demonstrated that, after statistical adjustment, menwho worked more than 35 hours a week had an odds ratioof 1.4 for being overweight (BMI> 25). No associationPage 2 of 8(page number not for citation purposes)larly stressful. between long hours of work and overweight was demon-strated for women.BMC Public Health 2006, 6:53 http://www.biomedcentral.com/1471-2458/6/53Using a representative survey of workers in the Australianstate of Victoria, we have assessed associations of jobstress (i.e., demand/control; effort-reward imbalancemodels), shift work, physical demand and hours worked,with BMI among men and women.MethodsThis study was reviewed and approved by the Universityof Melbourne's Human Research Ethics Committee(HREC protocol #030398).Study design and sampleA cross-sectional population-based survey was conductedby telephone from a random sample of White Pages list-ings in the state of Victoria in Australia. In order to reflectgeneral population occupational group proportions, quo-tas were set to match Australian Bureau of Statistics (ABS)census proportions of upper white-collar, lower white-collar, and blue-collar groups (29%, 30%, and 41%,respectively). We also quota sampled for urban/Mel-bourne (72%) versus rural/regional Victoria (28%). Theinclusion criteria were 1) being aged 18 years or older, and2) working at the time of the survey for profit or pay(including self-employed). Interviews were completed inNovember 2003 with a 66 % response rate from in-framehouseholds (i.e., had one or more working residents aged18 or over) to yield a representative sample of 1,101 work-ing Victorians (526 men and 575 women.).MeasuresA brief version [29] of Karasek's DC model [30] was usedto measure psychological demand (sum of 3 items, Cron-bach's alpha = 0.66) and job control (2 equally weightedscales of 6 and 3 items measuring skill discretion and deci-sion authority respectively, Cronbach's alpha = 0.80), andphysical demand at work (sum of 2 items). Each of the 3dimensions was dichotomised at the median. Dichot-omised psychological demand and job control were com-bined to create four categories: low strain (low demandand high control), active jobs (high demand and highcontrol), passive jobs (low demand and low control), andjob strain (high demand and low control). In subjectswith missing data, scores were recalculated using thelower and the higher theoretical score for each missingitem and dimensions dichotomised according to theirmedian. If the classification of participants was the samefor any possible value of the missing item, participantswere considered as having non-missing answers for thedimension of interest (38/88 participants with missingdata). If the classification differed according to thereplaced value, participants were considered as having amissing answer for the dimension.bach's alpha = 0.81), and over commitment (6 items,Cronbach's alpha = 0.82). Effort and reward items weresummed into scales. The effort and reward scales weredichotomized at the median to create two variables meas-uring high and low effort and high and low reward. Aswell, a ratio of effort to reward was computed using a cor-rection factor to give equal weight to both scales. The ratiowas dichotomised using a cut-point of 1. A ratio greaterthan 1 indicated that effort was higher than reward. Forparticipants who did not answer each question in thescale, scores were calculated if at least 80% of the itemswere answered (4/5 items for effort and 9/11 items forreward) [32]. Participants exposed to over commitmentwere defined as those in the sample upper tertile.Shift work was defined as work performed at least partlyduring the night, excluding day shift work and those whoworked exclusively during the day. Weekly working hourswere calculated as the average number of hours workedper week over the previous month, and treated categori-cally as up to Australian standard full-time hours (≤ 35hours/week), long working hours (36–49 hours/week),and excessive working hours (≥ 50 hours/week).Demographic and other covariate data were collected onage (treated categorically as < 30 years, 30–40 years, 41–50 years, and 51+ years), highest level of education com-pleted (post-graduate, undergraduate, vocational, highschool completion, and some primary or secondaryschool completion), and children living at home. Occupa-tions were collapsed into three categories (upper white,lower white, and blue-collar). Marital status was assessedas living as a couple versus living alone. Hostility wasassessed using the sum of a 3-item 5-point Likert scale[33] with higher scores indicating greater hostility. Thiswas dichotomized at the median.OutcomesBody mass index was based on self-reports, calculated asweight in kilograms divided by height in meters squared.Information on BMI was missing for 7 men and 43women.Statistical analysisLinear regression analyses were conducted with the SPSSstatistical package (version 12, SPSS Inc., Chicago, 2003)to assess the relationship of BMI with socio-demographicsand working conditions. Categorical variables with morethan two categories were included in models as dummyvariables. All analyses were conducted separately formales and females. One observation was excluded fromanalyses: a woman with BMI = 63.7. This was a markedoutlier that would be likely to unduly influence regressionPage 3 of 8(page number not for citation purposes)Siegrist's ERI model [8] was used to measure effort (5items, Cronbach's alpha = 0.80), reward (11 items, Cron-analyses because the nearest observation in the distribu-tion was a BMI of 44.6. Descriptive statistics and meanBMC Public Health 2006, 6:53 http://www.biomedcentral.com/1471-2458/6/53Page 4 of 8(page number not for citation purposes)Table 1: Mean Body Mass Index in Relation to Socio-demographics and Working Conditions in both GendersMen (N = 519)** Women (N = 531)BMI (SD) N (%) p value* BMI (SD) N (%) p value*Age <0.001 <0.001< 30 years 24.3 (3.2) 118 (22.7) 22.7 (4.8) 125 (23.5)30–40 years 26.0 (4.5) 157 (30.3) 24.7 (4.7) 150 (28.2)41–50 years 26.8 (3.8) 122 (23.5) 24.9 (4.4) 149 (28.1)? 51 years 27.2 (5.0) 122 (23.5) 25.7 (4.8) 107 (20.2)Educational level 0.037 0.036Post-graduate 26.1 (4.7) 47 (9.1) 24.4 (3.8) 52 (9.8)Undergraduate 25.9 (3.7) 128 (24.9) 24.5 (4.8) 204 (38.6)Vocational 26.6 (3.9) 127 (24.7) 24.0 (4.4) 71 (13.4)High school 24.9 (5.1) 89 (17.3) 23.5 (4.5) 101 (19.1)Primary or secondary 26.7 (4.6) 123 (23.9) 25.5 (5.3) 101 (19.1)Marital status <0.001 0.004Living as a couple 25.1 (4.0) 216 (41.6) 23.8 (4.8) 230 (43.4)Living alone 26.8 (4.4) 303 (58.4) 25.0 (4.7) 300 (56.6)Living with children 0.767 0.536No 26.1 (4.4) 294 (56.6) 24.4 (5.2) 272 (51.3)Yes 26.2 (4.2) 225 (43.4) 24.6 (4.3) 258 (48.7)Hostility 0.975 0.359Low 26.1 (4.1) 272 (52.4) 24.7 (5.0) 243 (45.8)High 26.1 (4.6) 247 (47.6) 24.3 (4.6) 288 (54.2)Occupational category 0.531 0.156Upper white collar 26.1 (3.9) 152 (29.3) 24.9 (4.7) 198 (37.3)Middle white collar 25.7 (4.8) 122 (23.5) 24.4 (4.8) 266 (50.1)Blue collar 26.3 (4.3) 245 (47.2) 23.6 (4.7) 67 (12.6)Hour worked <0.001 0.467<35 hours 24.2 (3.3) 77 (15.3) 24.6 (4.7) 241 (46.9)35 to 49 hours 26.2 (4.4) 258 (51.3) 24.2 (5.1) 198 (38.5)>49 hours 26.9 (4.0) 168 (33.4) 24.9 (4.5) 75 (14.6)Psychological work demand 0.162 0.473Low 25.8 (3.9) 293 (59.1) 24.3 (4.9) 271 (52.0)High 26.3 (4.8) 203 (40.9) 24.6 (4.7) 250 (48.0)Control 0.144 0.119High 26.4 (4.2) 255 (49.2) 24.9 (4.6) 214 (41.0)Low 25.8 (4.4) 263 (50.8) 24.2 (4.8) 308 (59.0)Demand/Control 0.151 0.479Low strain 26.1 (4.3) 138 (27.9) 24.8 (4.5) 99 (19.3)Active jobs 26.6 (4.1) 110 (22.2) 24.8 (4.6) 112 (21.8)Passive jobs 25.5 (3.6) 155 (31.3) 24.0 (4.9) 167 (32.6)Job strain 26.1 (5.5) 92 (18.6) 24.4 (4.8) 135 (26.3)Effort 0.004 0.338Low 25.6 (3.8) 290 (56.8) 24.3 (4.7) 282 (53.9)High 26.7 (4.9) 221 (43.2) 24.7 (4.9) 241 (46.1)Reward 0.938 0.024High 26.0 (3.7) 228 (47.7) 24.9 (4.7) 261 (51.0)Low 26.0 (5.0) 250 (52.3) 24.0 (4.7) 251 (49.0)Effort/Reward ratio 0.898 0.404? 1: Balance 26.0 (4.4) 455 (96.0) 24.5 (4.7) 488 (96.4)> 1: Imbalance 25.9 (4.0) 19 (4.0) 23.6 (4.1) 18 (3.6)Physical work demand 0.009 0.388Low 26.7 (4.3) 202 (39.2) 24.3 (4.6) 258 (48.9)High 25.7 (4.3) 313 (60.8) 24.7 (4.9) 270 (51.1)Shift work 0.262 0.823No 26.2 (4.3) 454 (87.5) 24.5 (4.9) 466 (87.8)Yes 25.5 (4.6) 65 (12.5) 24.6 (4.0) 65 (12.2)*Analysis of variance.** The original sample consisted of 526 men and 575 women. BMI information was missing for 7 men so this table shows data only for (526-7) = 519 men. BMI information was missing for 43 women and one women with BMI >60 was dropped from the analysis so this table shows data only for (574-43-1) = 531 women.BMC Public Health 2006, 6:53 http://www.biomedcentral.com/1471-2458/6/53value for BMI by socio-demographics and working condi-tions were first calculated for women and men. All multi-variate models controlled for age, education, and maritalstatus.Finally, job strain and effort-reward imbalance were mod-eled alone and in combination to assess the independenceof observed relationships between working conditionsand BMI, and to comprehensively assess the full range ofmeasured psychosocial working conditions in relation toBMI with simultaneous adjustment.ResultsDescriptive statistics and mean value for BMI by socio-demographics and psychosocial and other working condi-tions for women and men are shown in Table 1. Womenhad a lower mean BMI than men 24.5 versus 26.1 (p <0.001). For men and women, mean BMI increased signif-icantly with age, education level and living with a partner.For men, both high effort and working longer hours weresignificantly associated with a higher mean BMI. And,high physical demand was significantly associated withlower BMI. For women, low reward was associated withlower mean BMI. Hostility and living with children werenot significantly related to BMI and therefore these varia-bles were not included in the subsequent multivariateanalysis.The results of multivariate linear regression analyses forwomen and men are shown in Table 2. After adjusting forhigh effort were associated with increased BMI, and a neg-ative association was observed between high physicaldemand and BMI. Among women, after controlling forpotential confounders, a negative association was foundbetween low reward and BMI.We conducted a final set of multivariate analyses includ-ing statistically significant work-related predictors frombivariate analyses along with the DC and ERI models, sin-gly and in combination, with adjustment for sociodemo-graphic variables. Model 1 in Table 3 shows that DCmeasures are not significantly associated with BMI whenmodeled along with other work-related predictors. Mod-els 2 and 3 show that ERI is also not associated with BMI,either independently or in combination with DC meas-ures. The negative assocaiton between high physicaldemand and BMI remains stable and unaffected by inclu-sion of either or both job stress measures (Table 3, bottomrow). Associations with longer (35–49 hours/week) andexcessive (>50 hours/week) working hours are attentuatedslightly by inclusion of DC and ERI job stress models sin-gly, but when both the DC and ERI models are added tothe model, working long hours becomes statistically non-significant (Model 3).Model 4 is the most parsimonious final model presentingsignificant work-related predictors of BMI in men for oursample. The unstandardised regression coefficients showthat men with high physical demand have a mean BMIthat is 1.04 BMI units less than those in low physicalTable 2: Conditions and Sociodemographics: Multivariate Linear Regression AnalysisMen (N = 427) Women (N = 455)Beta* p value Beta p valueHigh Psychological work demand (Ref = Low) 0.87 0.04 0.23 0.61Low Control (Ref = High) -0.40 0.34 -0.48 0.31Demand/Control (Ref = Low strain)Active jobs 0.86 0.14 -0.06 0.94Passive jobs -0.34 0.54 -0.68 0.29Job strain 0.50 0.42 -0.31 0.65High Effort (Ref = Low) 1.26 >0.01 0.67 0.14Low Reward (Ref = High) -0.13 0.76 -0.87 0.05Effort/Reward Ratio > 1: Imbalance (Ref = Ratio 1: Balance)** -1.19 0.26 -0.98 0.44Hours worked (Ref = <35 hours)35 to 49 hours 1.35 0.03 0.05 0.92>49 hours 1.86 >0.01 0.48 0.47High physical work demand (Ref=Low) -1.15 0.01 0.30 0.50Each model (row) adjusted for age, educational level and marital status.* Unstandardised regression coefficient.** Additionally adjusted for over commitment.Ref = reference category.Page 5 of 8(page number not for citation purposes)age, education and marital status, for men, working morethan 35 hours a week, high psychological demand, anddemand jobs. In contrast, men with longer working hourshave a mean BMI that 1.37 units higher than those work-BMC Public Health 2006, 6:53 http://www.biomedcentral.com/1471-2458/6/53ing up to full time, and increasing in stepwise fashion,those working excessive hours have a mean BMI that is1.88 units higher than those working up to full time.For women, there were no statistically significant associa-tions between working conditions and BMI after includ-ing DCM and ERI, independently or in combination,along with socio-demographic adjustments (data notshown).DiscussionThere are four main results which arise from this study.First, while the effort/reward imbalance ratio itself is notassociated with BMI, high effort in men and low reward inwomen were associated with BMI. Second, for men, highpsychological demand, as measured in the DC model, waspositively associated with BMI. Third, after controlling formain job stress measures (ERI or DC), for men, longer andexcessive working hours were positively – and physicaldemand was negatively – associated with BMI. Fourth,among women, after similar adjustments neither psycho-social or other working conditions were significantly asso-ciated with BMI.Our population-based sample is representative of occupa-tional groups in the general population. Therefore, ourfindings can be generalised to the general working popu-lation in Victoria. Further, this study was strengthened bythe comparative assessment of a range of psychosocialand other working conditions (two measures of job stress,working hours, physical demand, and shift work).However, three limitations must be taken into account.First, the cross-sectional design of our study cannot sup-port causal inferences between occupational factors andquestionnaire was designed to minimize self-report biasin responses, some items may be subject to this type ofbias. In particular, psychological demand has been shownto have a strong subjective component. And, self-reportsof BMI produce under-estimates as men tend to overesti-mate their height and women tend to under-estimate theirweight. Use of self-reports for BMI, as in this paper, willtend to attenuate any associations observed.Studies of psychosocial working conditions and BMI havedemonstrated an association between high job strain [13],low control [13,24], and high ERI [13], "pressures on thejob" [22], a proxy measure of participation and control[4], and a borderline association between high psycholog-ical demand and BMI [27]. These results indicate thatdemands, control, or some combination of these may beassociated with increased BMI but clear results from thissmall number, of largely cross-sectional studies, do notemerge.The results from our study suggest that "demand" ratherthan "control" factors may be more salient in relation toBMI, at least for men. After fully controlling for confound-ing, two workload-related factors, psychological demand(from the DC model) and working long hours were bothstrongly associated with BMI. As well, in univariate mod-els high effort was associated with BMI. Given that higheffort combines measures of physical and psychologicaldemand into a single variable, and given that high physi-cal demand is negatively associated with BMI and highpsychological demand with increased BMI, the effects ofone may cancel the effects of the other, at least in relationto BMI.There are plausible mechanisms through which longTable 3: Body Mass Index in Relation to Working Conditions and Sociodemographics in Men: Multivariate Linear Regression ResultsModel 1 Model 2 Model 3 Model 4Men (N = 427) Beta* p value Beta p value Beta p value Beta p valueDemand/Control (Ref = Low strain)Active jobs 0.69 0.25 -- 0.61 0.30 --Passive jobs -0.14 0.80 -- -0.11 0.85 --Job strain 0.49 0.43 -- 0.48 0.44 --Effort/Reward Ratio > 1: Imbalance (Ref = Ratio 1: Balance) ** -- -1.18 0.26 -1.32 0.21 --Hours worked (Ref = <35 hours) 1.22 0.05 1.26 0.05 1.15 0.07 1.37 0.0335 to 49 hours 1.57 0.03 1.56 0.02 1.33 0.07 1.88 >0.01>49 hours -1.05 0.02 -1.05 0.02 -1.05 0.02 -1.04 0.02High physical work demand (Ref = Low)Each regression model includes covariates with reported coefficients, after adjustment for age, educational level and marital status.* Unstandardised regression coefficient.** Additionally adjusted for over commitment.Ref = reference category.Page 6 of 8(page number not for citation purposes)BMI. Second, information on independent and depend-ant variables were collected using self-reports. Even if theworking hours could be related to BMI. Working long orexcessive hours may reduce the opportunity for leisureBMC Public Health 2006, 6:53 http://www.biomedcentral.com/1471-2458/6/53time physical activity, might also increase the frequency ofeating higher caloric value take-away or restaurant food,and might also lead to higher alcohol consumption as anunwinding or coping mechanism. It is important to notealso that this assocaiton is of a similar magnitude to theeffect of increasing age on men (data not shown, highestage group [51+ years)] has mean BMI 1.60 units higherthan youngest [<30 years]).In this study, while half the men worked longer workinghours and one third excessive working hours less than halfthe women worked up to full time hours, 38% workedlonger hours, and only 15% worked excessive hours(Table 1). Thus, working more than 35 hours per week ismainly a male phenomenon so that the lack of observedassociation between working long hours and BMI amongwomen may be due to the small numbers of womenworking long hours.ConclusionOur findings suggest that psychosocial work conditionsmay impact BMI, particularly among men, and thatlargely independently of job stress, both low physicaldemand at work and longer working hours among menmay increase BMI.AbbreviationsBMI = Body Mass IndexCVD = Cardio-vascular DiseaseERI = Effort/reward ImbalanceDC = Demand/controlCompeting interestsThe author(s) declare that they have no competing inter-ests.Authors' contributionsAll authors read and approved the final manuscript.ASO:- Developed the conceptual and analytical models.Developed and wrote the literature review and drafted thepaper.SR:- Contributed to statistical analyses.ADL:- Designed, conducted, and directed the VictorianJob Stress Survey, and contributed to statistical analysesand the writing of the paper.AML:- Assisted in conducting the literature review and sta-AcknowledgementsDr. Ostry is funded through a new investigator award from the Canadian Institutes for Health Research and a scholar award from the Michael Smith Foundation for Health Research in British Columbia. He is also an Associate Scientist at the Institute for Work and Health in Toronto and a Faculty Associate at the School of Population Health at the University of Mel-bourne. Project funding was provided by a grant from the Australian National Heart Foundation (#G 01M 0345) to ADL, a Victorian Health Pro-motion Foundation Senior Research Fellowship (#2001-1088) to ADL, Australian National Health and Medical Research Council (NHMRC) Post-Doctoral Fellowship # 3165812 to SR. Support for collaboration between the University of British Columbia and the University of Melbourne was provided by an international collaborations small grant from the Canadian Institute for Health Research (Grant #20R 91434).References1. Australian Institute of Health and Welfare: Australia's Health 2004: Theninth biennial health report of the Australian Institute of Health and Wel-fare Canberra: Australian Institute of Health and Welfare; 2004. 2. Torrance GM, Hooper MD, Reeder BA: Trends in overweight andobesity among adults in Canada (1970–1992): evidence fromnational surveys using measured height and weight.  Int J ObesRelat Metab Disord 2002, 26:797-804.3. Troiano RP, Frongillo EA Jr, Sobal J, Levitsky DA: The relationshipbetween body weight and mortality: a quantitative analysisof combined information from existing studies.  Int J Obes RelatMetab Disord 1996, 20:63-75.4. Rosmond R, Bjorntorp P: Psychosocial and socio-economic fac-tors in women and their relationship to obesity and regionalbody fat distribution.  Int J Obes Relat Metab Disord 1999,23:138-145.5. Belkic KL, Landsbergis PA, Schnall PL, Baker D: Is job strain a majorsource of cardiovascular disease risk?  Scand J Work EnvironHealth 2004, 30:85-128.6. Bosma H, Peter R, Siegrist J, Marmot M: Two alternative jobstress models and the risk of coronary heart disease.  AmericanJournal of Public Health 1998, 88:68-74.7. Peter R, Siegrist J: Psychosocial work environment and the riskof coronary heart disease.  Int Arch Occup Environ Health 2000,73:S41-S45.8. Siegrist J: Adverse health effects of high-effort/low-rewardconditions.  Journal of Occupational Health Psychology 1996, 1:27-41.9. Landsbergis PA, Schnall PL, Deitz DK, Warren K, Pickering TG,Schwartz JE: Job strain and health behaviours: results of a pro-spective study.  American Journal of Health Promotion 1998,12:237-245.10. Lallukka T, Sarlio-Lahteenkorva S, Roos E, Laaksonen M, Rahkonen O,Lahelma E: Working conditions and health behaviours amongemployed women and men: the Helsinki Health Study.  PrevMed 2004, 38:48-56.11. Tsutsumi A, Kayaba K, Yoshimura M, Sawada M, Ishikawa S, Sakai K,Gotoh T, Nago N, Jichi Medical School Cohort Study Group: Asso-ciation between job characteristics and health behaviors inJapanese rural workers.  International Journal of Behavioral Medicine2003:125-142.12. Hellerstedt WL, Jeffery RW: The association of job strain andhealth behaviours in men and women.  International Journal ofEpidemiology 1997:575-583.13. Kouvonen A, Kivimaki M, Cox SJ, Cox T, Vahtera J: Relationshipbetween work stress and body mass index among 45,810female and male employees.  Psychosom Med 2005, 67:577-583.14. Kivimaki M, Leino-Arjas P, Luukkonen R, Riihimaki H, Vahtera J, Kir-jonen J: Work stress and risk of cardiovascular mortality: pro-spective cohort study of industrial employees.  BMJ 2002,325:857.15. Niedhammer I, Goldberg M, Leclerc A, David S, Bugel I, Landre M:Psychosocial work environment and cardiovascular risk fac-tors in an occupational cohort in France.  Journal of edpidemiol-ogy and community health 1998, 52:93-100.16. Brisson C, Lorocque B, Moisan J, Vezina M, Dagenais GR: Psychoso-Page 7 of 8(page number not for citation purposes)tistical analysis.cial factors at work, smoking sedentary behavior, and BodyMass Index: a prevalence study among 6995 whilte collarPublish with BioMed Central   and  every scientist can read your work free of charge"BioMed Central will be the most significant development for disseminating the results of biomedical research in our lifetime."Sir Paul Nurse, Cancer Research UKYour research papers will be:available free of charge to the entire biomedical communitypeer reviewed and published immediately upon acceptancecited in PubMed and archived on PubMed Central BMC Public Health 2006, 6:53 http://www.biomedcentral.com/1471-2458/6/53workers.  Journal of Occupational and Environmental Medicine 2000,42:40-52.17. Hellerstedt WL, Jeffery RW: The association of job strain andhealth behaviours in men and women.  International Journal ofEpidemiology 1997, 26:575-583.18. Georges E, Wear M: Body fat distribution and job stress in Mex-ican-American men of the Hispanic Health and NutritionExamination Study.  American Journal of Human Biology 1992,4:657-667.19. Kornitzer M, Kittel F: How does stress exert its effects – smok-ing, diet and obesity, physical activity?  Postgrad Med J 1986,62:695-696.20. Netterstrom B, Kristensen TS, Damsgaard MT, Olsen O, Sjol A: Jobstrain and cardiovascular risk factors: A cross-sectional studyof employed Danish men and women.  British Journal of IndustrialMedicine 1991, 48:684-689.21. Reed DM, Lacroix AZ, Karasek RA, Miller D, MacLean CA: Occupa-tional strain and the incidence of coronary heart disease.American Journal of Epidemiology 1989, 129:495-502.22. House JS, Strecher V, Metzner HL, Robbins CA: Occupationalstress and health among men and women in the TecumsehCommunity Health Study.  Journal of health and social behavior1986, 27:62-77.23. Jonsson D, Rosengren A, Dotevall A, Lappas G, Wilhelmsen L: Jobcontrol, job demands and social support at work in relationto cardiovascular risk factors in MONICA 1995, Goteborg.Journal of cardiovascular risk 1995, 6:379-385.24. Martikainen PT, Marmot MG: Socioeconomic differences inweight gain and determinants and consequences of coronaryrisk factors.  Am J Clin Nutr 1999, 69:719-726.25. Wamala SP, Wolk A, Orth-Gomer K: Determinants of obesity inrelation to socioeconomic status among middle-aged Swed-ish women.  Prev Med 1997, 26:734-744.26. Kornitzer M, Kittel F: How does stress exert its effects – smok-ing, diet and obesity, physical activity?  Postgrad Med J 1986,62:695-696.27. Georges E, Wear M: Body fat distribution and job stress in Mex-ican-American men of the Hispanic Health and NutritionExamination Study.  American Journal of Human Biology 1992,4:657-667.28. Shields M: Long working hours and health.  Health Rep 1999,11:33-48.29. Mausner-Dorsch H, Eaton WW: Psychosocial work environ-ment and depression: epidemiologic assessment of thedemand-control model.  American Journal of Public Health 2000,90:1765-1770.30. Karasek RA: Job Demands, Job Decision Latitude, and MentalStrain – Implications for Job Redesign.  Administrative ScienceQuarterly 1979, 24:285-308.31. Karasek R, Gordon G, Pietrokovsky C, Frese M, Pieper C, SchwartzJ, Fry L, Schirer D: Job content instrument: Questionnaire and users' guideLowell: University of Massachusetts; 1985. 32. Pikhart H, Bobak M, Pajak A, Malyutina S, Kubinova R, Topor R, Seba-kova H, Nikitin Y, Marmot M: Psychosocial factors at work anddepression in three countries of Central and Eastern Europe.social science & medicine 2004, 58:1475-1482.33. Koskenvuo M, Kaprio J, Rose RJ, Kesaniemi A, Sarna S, Heikkila K,Langinvainio H: Hostility as a risk factor for mortality andischemic heart disease in men.  Psychosom Med 1988,50:330-340.Pre-publication historyThe pre-publication history for this paper can be accessedhere:http://www.biomedcentral.com/1471-2458/6/53/prepubyours — you keep the copyrightSubmit your manuscript here:http://www.biomedcentral.com/info/publishing_adv.aspBioMedcentralPage 8 of 8(page number not for citation purposes)


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items