UBC Faculty Research and Publications

A case control study of differences in non-work injury and accidents among sawmill workers in rural compared… Ostry, Aleck; Maggi, Stefania; Hershler, Ruth; Chen, Lisa; Louie, Amber; Hertzman, Clyde Nov 25, 2009

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

Item Metadata


52383-12889_2009_Article_1835.pdf [ 186.47kB ]
JSON: 52383-1.0223181.json
JSON-LD: 52383-1.0223181-ld.json
RDF/XML (Pretty): 52383-1.0223181-rdf.xml
RDF/JSON: 52383-1.0223181-rdf.json
Turtle: 52383-1.0223181-turtle.txt
N-Triples: 52383-1.0223181-rdf-ntriples.txt
Original Record: 52383-1.0223181-source.json
Full Text

Full Text

ralssBioMed CentBMC Public HealthOpen AcceResearch articleA case control study of differences in non-work injury and accidents among sawmill workers in rural compared to urban British Columbia, CanadaAleck Ostry*1, Stefania Maggi2, Ruth Hershler3, Lisa Chen3, Amber Louie3 and Clyde Hertzman3Address: 1Department of Geography, PO BOX 3060 STN CSC, University of Victoria, Victoria, B.C., V8W 3R4, Canada, 2Department of Psychology and Institute of Interdisciplinary Studies, Dunton Tower Room 2210, Carleton University, 1125 Colonel By Drive, Ottawa, ON K1S 5B6, Canada and 3Human Early Learning Program, University of British Columbia, 4th Floor, Library Processing Centre, 2206 East Mall, Vancouver, B.C., V6T 1Z3, CanadaEmail: Aleck Ostry* - ostry@uvic.ca; Stefania Maggi - Stefania_Maggi@carleton.ca; Ruth Hershler - ruth.hershler@ubc.ca; Lisa Chen - lisa.chen@ubc.ca; Amber Louie - amber.louie@ubc.ca; Clyde Hertzman - hertzman@interchange.ubc.ca* Corresponding author    AbstractBackground: Using a cohort of British Columbian male sawmill workers, we conducted a nestedcase-control study of the impact of rural compared to urban residence as well as rural/urbanmigration patterns in relation to hospitalization for non-work injury. We postulate that for manytypes of non-work injuries, rates will be higher in rural communities than in urban ones and thatrates will also be higher for workers who migrate from urban to rural communities.Methods: Using conditional logistic regression, univariate models were first run with each of fivenon-work injury outcomes. These outcomes were hospitalizations due to assault, accidentalpoisoning, medical mis-adventure, motor vehicle trauma, and other non-work injuries. Inmultivariate models marital status, ethnicity, duration of employment, and occupation were forcedinto the model and associations with urban, compared to rural, residence and various urban/migration patterns were tested.Results: Urban or rural residence and migration status from urban to other communities, andacross rural communities, were not associated with hospitalization for medical misadventure,assault, or accidental poisoning. The likelihood of a rural resident being hospitalized for motorvehicle trauma is higher than for an urban resident. The likelihood that a rural resident ishospitalized for "other" non-work injury is higher than for an urban resident.Conclusion: In a relatively homogenous group of workers, and using a rigorous study design, wehave demonstrated that the odds of other non-work injury are much higher for workers residentin and migrating to rural regions of Canada than they are for workers resident in or migrating tourban places.Published: 25 November 2009BMC Public Health 2009, 9:432 doi:10.1186/1471-2458-9-432Received: 12 March 2009Accepted: 25 November 2009This article is available from: http://www.biomedcentral.com/1471-2458/9/432© 2009 Ostry 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 7(page number not for citation purposes)BMC Public Health 2009, 9:432 http://www.biomedcentral.com/1471-2458/9/432BackgroundAccording to an investigation into rural health conductedby the Canadian Population Health Initiative, the healthstatus of rural Canadians is systematically worse than it isfor urban Canadians, for most, but not all, outcomes[1,2]. This report demonstrated that urban-rural differ-ences in health status, in morbidity and mortality frommost illnesses, and health behaviours remain even aftercontrolling for socio-economic status, suggesting thatsomething about rural life in itself accounts for these dif-ferences.The report also clearly illustrated the major differences insocio-economic status between urban and rural Canadi-ans. For example, in remote areas of Canada approxi-mately 50 percent of the population has little formaleducation compared to a figure of approximately 25 per-cent in urban regions. This underscores the fact that anyanalyses comparing health outcomes between residents inrural and urban areas must take into account the large dif-ferences in labour market, income, and educational sta-tus. When investigating health outcomes across the rural/urban continuum, it is important to be able to control forconfounding by various measures of socio-economic sta-tus and/or conduct studies among sub-populations thatare similar across regions.Non-work injury, except perhaps in the case of motorvehicle trauma, remains under-investigated in studies ofrural health. There is evidence that in the case of motorvehicle trauma, that in North America, both injuries andfatalities occur in higher proportion and with greaterseverity in rural compared to urban areas [3,4]. For non-work injury outcomes such as assault, medical misadven-ture, and accidental poisoning, there is little understand-ing of differences across rural and urban places.In recent years a growing body of evidence has been gen-erated investigating occupational and community influ-ences on a wide range of health outcomes using a largecohort of British Columbian sawmill workers and theirchildren [5-16]. The present study explores differencesbetween urban and rural sawmill worker members of thiscohort in hospitalization for non-work injury.Because the nature and organization of work in sawmillslocated in rural and urban regions of the province is fairlysimilar and because educational requirements and wagespaid have been similar throughout the province, thecohort members are a homogeneous group. As well, thereis information, for cohort members, on potential socio-economic confounders. Using this cohort in investiga-tions of rural/urban differences in health, provides avariables are available to further control for socio-eco-nomic confounding.The aim of the study was to explore the relationshipbetween rural and urban residency and migrationbetween rural and urban places and the risk of hospitali-zation for non-work related injury among BC sawmillworkersLiterature ReviewThis literature review is divided into four sections. In thefirst section we review the Canadian literature on rural/urban differences in motor vehicle and other vehicularaccidents. Because there is no available Canadian researchon rural/urban differences in accidental poisonings andnon-work injuries, in the second section we review the,although limited, international literature on this topic. Inthe third section we review the limited Canadian literatureon rural/urban differences in assault. Finally, in the fourthsection, we review the international literature on rural/urban differences in medical misadventure.Motor Vehicle TraumaAcross North America injuries and fatalities due to motorvehicle trauma occur in higher proportion and greaterseverity in rural compared to urban areas [3,4]. Accordingto Transport Canada in 2002, 63.2% of all fatal crashesoccurred on rural roadways [17]. In Alberta in 2002,nearly 75% of fatal crashes occurred in rural areas [17-19].These studies also illustrated that speeding and not wear-ing seat belts were more prevalent in rural than in urbanareas.In Alberta, rural residents and men were more likely tosustain spinal injuries, mainly due to vehicle accidents[20]. Using data from the Canadian Institute for HealthInformation (CIHI), Macpherson et al. [19] investigatedall Canadian children hospitalized due to bicycling-related injuries (1994-1998, n = 9367). The averageannual incidence rate for bicycle-related head injuries inchildren was 18.5 per 100, 000 for children living in ruralcompared with 10.9 in urban areas, 15.5 in mixed urbanand 17.4 in mixed rural areas. Logistic regression, control-ling for age, sex, socio-economic status (SES), collisionwith a motor vehicle, and the presence of provincial hel-met legislation, suggested that this variation may beexplained by differences in bicycling exposure, helmetuse, hospital admission criteria, or road environmentsacross geographic areas.A population-based study of motor vehicle trauma amongchildren and youth in Alberta examined police report datafor the period from 1997 to 2002 [21]. Across all age andPage 2 of 7(page number not for citation purposes)unique opportunity to compare health outcomes within arelatively homogeneous population of workers for whichsex strata, both hospitalization and fatality rates were sig-nificantly higher in rural compared with urban regions.BMC Public Health 2009, 9:432 http://www.biomedcentral.com/1471-2458/9/432After adjusting for age, sex, and calendar year, the relativerisk for motor vehicle trauma hospitalization (rural versusurban) was 3.0 (95% CI: 2.8, 3.2), and for fatality, 5.4(95% CI: 4.2, 6.9).Accidental poisoning and Other non-work injuriesThe research on rural/urban differences in accidental poi-soning and non-work injuries in Canada is non-existent.However, a limited number of researchers in other coun-tries have examined this topic. Boland and colleagues [22]conducted a study of urban/rural differences in mortalityand hospital admission rates for non-work injuries in theRepublic of Ireland. Central Statistics Office mortalitydata from 1980-2000 were used to calculate standardizedmortality ratios (SMRs) in residents of urban and ruralareas, and standardized hospital admission ratios (SARs)in urban and rural residents were calculated using hospi-tal admission data (Hospital In-Patient Enquiry) from1993-2000. The overall rate of non-work injury mortalitywas significantly higher among rural residents (SMR103.0, 95% CI 101-105), and also for deaths related todrowning, accidents and injury from machinery, and fire-arms. Among rural residents SARs were significantlyhigher for injuries from falls, being struck by or against anobject, fire or burns, and accident and injury frommachinery.A cross-sectional study of poisoning of children aged 0-4years based on crude rates of hospitalizations in Australiaduring the financial year 1996-97 found significantlyhigher rates among children living in rural and remoteareas compared with those living in metropolitan areas[23]. Rate differentials increased with geographicalremoteness.A UK study calculated SMRs using data from the longitu-dinal study of the Office of Population Censuses and Sur-veys, a quasi-random 1% sample of the population ofEngland and Wales [24]. In general, the results demon-strated a striking similarity between metropolitan andnon-metropolitan areas, for deaths from accident, vio-lence, and poison.AssaultA 1991 Canadian study on domestic homicide involvingfirearms [25] showed that almost half (49%) of domestichomicides occurred in rural areas (i.e., places with a pop-ulation less than 10,000), even though rural residencesaccount for only 23% of the population. However, con-sistent with previous research, urban dwellers reporthigher rates of personal victimization--including sexualassault, robbery, assault, break and enter, motor vehicle/parts theft and vandalism--than those from rural areas.rate over 40% higher than that of rural dwellers (199 ver-sus 138 per 1,000) [26].Estimates of the rates of violence against women in ruralCanada are few [27]. The Statistic Canada General SocialSurvey found no variation in reported rates of spousal vio-lence between urban and rural men and women [28].However, in rural areas, 2% of women and 1% of menreported spousal violence in the past 12 months by theircurrent partners, compared to 1% of women and 2% ofmen in urban areas. Notably though, availability of serv-ices that address domestic violence, including health serv-ices, is lower in rural areas. This may reduce reportingrates for rural citizens [27]. In a cross-sectional survey in arural health region in Alberta, among 526 women, 5% ofwomen reported experiencing physical assault in the last12 months and 23% reported experiencing sexual assaultin their lifetime, indicating that rates of spousal abuse inrural regions are moderately high [29].Medical MisadventureThere is a lack of research on how patient safety and qual-ity of care differ between rural and urban settings [30]. Areview of the limited available research, mainly from theUnited States, suggests that patient safety events and med-ical errors may be less likely to occur in rural than in urbanhospitals [31]. For example, Romano et al. [32] conducteda study of patient safety in the U.S. [33]. They found thatthe incidence of most Patient Safety Incidents was highestat urban teaching hospitals. The Harvard Medical PracticeStudy, conducted in acute care hospitals in New York Statein 1984 also showed significantly lower medical injuryrates in rural compared to metropolitan hospitals [34].MethodsUsing all the International Classification of Diseases E-codes available from the BC Linked Health Database(BCLHDB) [35] we were able to study rural/urban differ-ences in hospitalization for the following five outcomes:1) assault; 2) accidental poisonings; 3) medical misadven-tures; 4) motor vehicle trauma; and 5) other non-workinjuries. These E-codes pertain only to non-work injury.As well, because we have information on migration pat-terns and not just current place of residence, we were ableto assess the influence of different migration patternsbetween urban and rural, as well as across rural commu-nities.This study is based on a cohort of male sawmill workerswhich was assembled in the 1980s to assess the effects ofchemical exposure on mortality and cancer among BCsawmill workers; it has been adapted for use in thispresent investigation. In the original study, fourteen largePage 3 of 7(page number not for citation purposes)Urban residents reported a total personal victimization sawmills (150 to 450 workers each) were identified, somein urban and others in rural areas. The personnel recordsBMC Public Health 2009, 9:432 http://www.biomedcentral.com/1471-2458/9/432of workers employed for at least one year between 1950and 1998 were used to identify the study participants. Thisprovided us with detailed personal and job history infor-mation for 28,794 workers employed at these mills from1950 through to 1998 [7].Definition of Rural and Urban Utilized in the Present StudyFor our definition of rural and urban we determined if thepopulation of each place where the workers were diag-nosed was greater or less than 100,000. If it was less than100,000 the place of diagnosis was classified as rural, andif it was over 100,000 it was urban.Obtaining Information on Non-Work Injury OutcomesHealth information for each cohort member was obtainedby probabilistic linkage to the BC Linked Health Database(BCLHDB) which has files on physician services utiliza-tion and hospital discharges from 1985 to the present.The records are housed at the University of British Colum-bia's Population Health Observatory. The BCLHDB ismanaged according to the provisions of British Colum-bia's Freedom on Information and Protection of PrivacyAct. Ethical approval to conduct this study was obtainedfrom the University of British Columbia (UBC) and theBritish Columbia Ministry of HealthThere are approximately 120 "E" codes characterizing arange of non-work accidents and injuries in the hospitaldischarge data. We utilized approximately 100 of thesecodes (See Additional File 1) in order to develop five gen-eral categories of non-work injury and accident for thepurposes of this investigation. The five categories are asfollows: 1) assault; 2) accidental poisonings; 3) medicalmisadventures; 4) motor vehicle trauma; 5) other non-work injuries and accidents. Approximately 20 codes didnot fit in these 5 basic categories and were thus excludedfrom analysis.Selection of Cases and ControlsThere were several reasons for using a nested case controldesign. First, we were able to determine non-work injuryoutcomes across several major diagnostic categories. Sec-ond, the study statistically controlled for residual con-founding by socio-economic factors, thus increasingcomparability between workers living in urban and ruralsettings. Third, the study was longitudinal in design andutilized common International Classification of Disease(ICD 9) codes for non-work injury outcomes, based upona common data source for urban and rural study subjects.Finally, because of its historical prospective character wewere able to address confounding by migration, identify-ing workers who migrated between rural and urban envi-ronments before and after diagnosis.Complete hospital diagnoses for these five categories ofnon-work accidents and injury were available in the BCL-HDB [35] from January 1st 1994 until December 31st,2001. Cases were eligible for selection from this 8-yearperiod. Cases included all subjects with a first ICD9 diag-nostic code for these five categories. We identified 151hospital discharges for assault, 75 cases of accidental poi-sonings, 1,073 cases of medical misadventure, 470 casesfor motor vehicle trauma, and 2,046 cases of other non-work injury.For each case we identified the place they were livingwhen diagnosed with a non-work related injury, usingpostal codes available in the BCLHDB. In this way wewere able to determine whether a case that originated atan urban mill had remained in their same urban location(urban stay), or had moved away from this mill (urbanmigrate). Similarly, we determined whether a case thatoriginated at a rural mill remained at the same location(rural stay), moved to an urban location (rural urban), ormoved to another rural location (rural rural). This classi-fication scheme therefore identified two types of cases thatwere non-migrators (those who stayed in the same urbanlocation and those who remained in the same rural loca-tion), as well as three types of cases involving migration(those urban dwellers who migrated away from their orig-inal urban location, rural dwellers who migrated to anurban place, and rural dwellers who migrated away fromtheir original rural place to another rural place). Note thatwe did not determine whether the urban dwellers whomigrated away from their original urban location movedto another urban place or to a rural one, only that theymigrated away from an urban location.Using STTOCC (survival-time to case-control) on STATA8.0, three controls were selected for each case matched onage. Controls were chosen randomly with replacementfrom the set at risk, that is, all the members of the cohortwho worked in a study sawmill for at least one year. Thus,a control could be anyone at risk who also satisfied thematching criteria, and who had not had a non-workrelated injury up to the time of diagnosis of the case.ResultsNo significant associations were observed in univariateanalyses for assault, accidental poisoning, or medical mis-adventure. Univariate analyses indicate that those whostayed in urban regions had a lower Odds Ratio for motorvehicle trauma (.60; CI .42-.86), and those who migratedfrom rural to a different rural area had elevated OddsRatios for motor vehicle trauma (1.49; CI 1.19-1.86). Inmultivariate models, after controlling for socio-demo-graphic variables, duration of employment, and occupa-Page 4 of 7(page number not for citation purposes)tion, workers who migrated from one rural community toanother had approximately twice the odds of motor vehi-BMC Public Health 2009, 9:432 http://www.biomedcentral.com/1471-2458/9/432cle trauma than workers who remained in an urban com-munity. Statistically significant and elevated rates werealso observed for rural compared to urban residents (OR= 1.56; CI 1.01-2.40), and for workers who moved from arural to urban community (OR = 1.76; CI 1.11-2.79). Aswell, in this model, relative to managers, skilled workerswere approximately 2.5 times as likely, and unskilledworkers twice as likely, to sustain a motor vehicle trauma(Tables 1, 2).Univariate analyses for other non-work injury indicatereduced odds for urban stayers (OR = 0.65; CI .53-.80)and for those workers who migrated from rural to urbancommunities (OR = 0.82; CI .67-.99). In contrast elevatedodds were observed for rural stayers (OR = 1.43; CI 1.18-1.74) and those who migrated from rural to another ruralcommunity (OR = 1.33; CI 1.16-1.53). Multivariate mod-els, after controlling for socio-demographic variables,duration of employment, and occupation, showed evenhigher odds for rural stayers (OR = 1.94; CI 1.49-2.53)and for migrants from rural to other rural communities(OR = 1.63; CI 1.29-2.07). Additionally, these multivari-ate models showed elevated odds (OR = 1.34; CI 1.04-1.72) for workers who migrated away from urban com-munities (Tables 1 &2).DiscussionThere are three main results from this study. First, urbanor rural residence and migration status from urban toother communities, and across rural communities, wasnot associated with hospitalization for assault, accidentalpoisoning, or medical misadventure. The results forassault accord with the limited literature on rural/urbandifferences in Canada.Second, in accord with existing Canadian research, thelikelihood of a rural resident being hospitalized for motorvehicle trauma is higher than for an urban resident. Ourresearch indicates that, relative to urban dwellers, rates areeven higher for cohort members who move from onerural community to another rural community. As well,workers who migrate from urban communities have agreater likelihood of being involved in a motor vehicletrauma. We did not track the types of communities thatthese urban migrators moved to, it may be that mostmoved to rural communities, however in this study theirdestination remains unknown.Third, in accord with the limited Canadian research, thelikelihood that a rural resident is hospitalized for othernon-work injury is higher than for an urban resident.Although the odds are somewhat lower for workers whomigrate from one rural community to another, they arestill higher than for workers who remain in urban com-munities.Even among a group of workers employed in the sametype of industry, and even after strict controls for con-founding related to demographic and occupational fac-tors, the odds for motor vehicle trauma and other non-work injury are significantly higher for rural workers com-pared to their urban counterparts. This finding points tostructural features of rural non-work and recreational lifeand activity as risks for greater hospitalization for thesetwo outcomes. Road safety in rural areas is a major healthissue, and improvements in this regard may help toredress some of the imbalance in motor vehicle traumaoutcomes for rural residents. As well, safety in outdoorrural recreational pursuits and in non-paid work pursuitssuch as do-it-yourself home renovations may pay off interms of reducing hospitalization for these outcomes.There are several limitations to this study. Outcomes wereICD 9 codes based on hospitalized cases. They do not cap-ture less severe non-work injury cases. As well, this studyis based on males only. As it is an unusual population ofworkers this study is not representative of the general pop-ulation and so the findings cannot be generalized. Thedefinition of rural used in this investigation is very broad;rural place was defined simply as any population centerwith less than 100,000 people. So, in effect we are meas-uring the difference between residents of Census Metro-politan Areas (CMAs) versus "elsewhere." This thresholdfor rurality is much higher than is used in most otherurban/rural investigations, and so limits the comparabil-ity of this study to others.Table 1: Univariate analyses: Odds ratios for motor vehicle trauma and other non-work injury among sawmill workers for the period 1994 through 2001Motor Vehicle Trauma(N = 470)Other non-work injury(N = 2,046)Urban stay .60 (.42, .86). .65 (.53, .80)Urban migrate .79 (.60, 1.04) .96 (.82, 1.13)Rural stay 1.01 (.76, 1.35) 1.43 (1.18, 1.74)Rural to urban 1.15 (.86, 1.54) .82 (.67, .99)Rural to rural 1.49 (1.19, 1.86) 1.33 (1.16, 1.53)Page 5 of 7(page number not for citation purposes)* Numbers in parentheses are 95% Confidence Intervals.BMC Public Health 2009, 9:432 http://www.biomedcentral.com/1471-2458/9/432Finally, our classification of workers migratory trajectorieswas crude. In particular, we did not divide urban migra-tors into those who migrate to other urban places andthose who migrate from urban to rural places. However,despite these limitations there are a great many strengthsto this study.The study was rigorous in design utilizing objective meas-ures of non-work injury, fine control for socio-economicdifferences among participants, and it was longitudinal.Furthermore, this study is based on a population that wasselected based on its employment status, so it largelyexcluded unhealthy people. Finally, as most researcherson the rural/urban health divide have noted, it is impor-tant in studies of this type to measure not only non-workinjury outcomes among rural and urban residents but alsoto assess outcomes among migrants; this study doesexactly that.ConclusionIn a relatively homogenous group of workers, and using arigorous study design, we have demonstrated that theodds of other non-work injury are much higher for work-ers resident in and migrating to rural regions of Canadathan they are for workers resident in or migrating to urbanplaces.Competing interestsThe authors declare that they have no competing interests.Authors' contributionsAO was PI on the study to obtain funding for this research,directed the analysis, and was the lead writer. SM took awith the literature review. CH conducted the research,helped direct the analysis, and read drafts of the paper.All authors read and approved the final manuscript.Additional materialReferences1. Canadian Institute for Health Information: How healthy are rural Cana-dians? An assessment of their health status and health determinantsOttawa: Canadian Institute for Health Information; 2006. 2. Statistics Canada: Canadian community health survey Ottawa: StatisticsCanada, Special Surveys Division; 2005. 3. Miles-Doan R, Kelly S: Inequities in health care and survivalafter injury among pedestrians: explaining the urban/ruraldifferential.  J Rural Health 1995, 11:177-184.4. Thompson EJ, Russell ML: Risk factors for non-use of seatbelts inrural and urban Alberta.  Can J Public Health 1994, 85:304-306.5. Teschke K, Marion SA, Ostry A, Hertzman C, Hershler R, Dimich-Ward H, Kelly S: Retrospective Chlorophenol ExposureAssessment in the Sawmill Industry: Reliability of Estimatesof Worker Exposure Over Five Decades.  Am J Ind Med 1996,30:616-622.6. Dimich-Ward H, Hertzman C, Teschke K, Hershler R, Marion SA,Ostry A, Kelly S: Reproductive Effects of Paternal Exposure toChlorophenate Wood Preservatives in the Sawmill Industry.Scand J Work Env Hea 1996, 22:267-273.7. Hertzman C, Teschke K, Ostry A, Hershler R, Dimich-Ward H, KellyS, Spinelli J, Gallagher R, McBride M, Marion SA: Cancer IncidenceAdditional file 1"E-Codes" available in BCLHDB Hospital Discharge Records and Collapsed into Five Major Categories. This is a table outlining how the E codes for injury were aggregated into five categories.Click here for file[http://www.biomedcentral.com/content/supplementary/1471-2458-9-432-S1.DOCX]Table 2: Multivariate analyses: Odds ratios for motor vehicle trauma and other non-work injury among sawmill workers for the period 1994 through 2001Motor Vehicle TraumaN = 470Other non-Work InjuryN = 2,046Duration of job atsawmill (years) .99 (.97, 1.01) .99 (.98, 1.01)Marital status 1.02 (0.97, 1.07) 1.02 (.99, 1.05)Tradesman 1.99 (.98, 4.04) 1.04(.72, 1.45)Skilled 2.49 (1.20, 5.17) .95 (.66, 1.35)Unskilled 2.05 (.1.01, 4.17) .96 (.68, 1.34)Caucasian 1 1Chinese 1.07 (.34, 3.35) .23 (.08, .70)Sikh 1.30 (.86, 1.96) .84 (.61, 1.16)Urban stay 1 1Urban migrate 1.35 (.87, .2.09) 1.34 (1.04, 1.72)Rural stay 1.56 (1.01, 2.40) 1.94 (1.49, 2.53)Rural to urban 1.76 (1.11, 2.79) 1.12 (.85, 1.48)Rural to rural 2.09 (1.40, 3.13) 1.63 (1.29, 2.07)* Numbers in parentheses are 95% Confidence Intervals.Page 6 of 7(page number not for citation purposes)lead on the analysis and reviewed drafts. RH conductedthe analysis and LC managed the database. AL assistedand Mortality in Sawmill Workers Exposed to Chlorophe-nols.  Am J Public Health 1997, 87(1):71-79.Publish 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 2009, 9:432 http://www.biomedcentral.com/1471-2458/9/4328. Teschke K, Ostry A, Hertzman C, Demers PA, Barroetavena M, Dav-ies H, Dimich-Ward H, Heacock H, Marion SA: Opportunities fora Broader Understanding of Work and Health: Multiple Usesof an Occupational Cohort Database.  Can J Public Health 1998,89(2):132-136.9. Ostry A, Marion SA, Green L, Demers PA, Hershler R, Kelly S,Teschke K, Hertzman C: Downsizing and Industrial Restructur-ing in Relation to Changes in Psychosocial Conditions ofWork in British Columbia Sawmills.  Scand J Work Env Hea 2000,26(3):273-277.10. Ostry A, Marion SA, Green L, Demers PA, Hershler R, Kelly S,Teschke K, Hertzman C: The Relationship between Unemploy-ment, Technological Change and Psychosocial Work Condi-tions in British Columbia Sawmills.  Crit Public Health 2000,10(2):179-192.11. Ostry A, Marion SA, Green L, Demers PA, Hershler R, Kelly S,Teschke K, Mustard C, Hertzman C: Comparison of Expert-raterMethods for Assessing Psychosocial Job Strain.  Scand J WorkEnv Hea 2001, 27(1):1-6.12. Ostry A, Marion SA, Green L, Demers PA, Hershler R, Kelly S,Teschke K, Hertzman C: Measuring Psychosocial Job Strainwith the Job Content Instrument Using Experienced JobEvaluators.  Am J Ind Med 2001, 39(4):397-401.13. Ostry A, Maggi S, Tansey J, Dunn J, Hershler R, Chen L, Louie A,Hertzman C: The Impact of Father's Physical and PsychosocialWork Experience on Attempted and Completed Suicideamong their Children.  BMC Public Health 2006, 6:77.14. Ostry A, Maggi S, Tansey J, Dunn J, Hershler R, Chen L, Hertzman C:The Impact of Psychosocial and Physical Work Experienceon Mental Health: A nested case control study.  Can J CommunMent Health 2006, 25(2/Fall):59-70.15. Alamgir H, Tompa E, Ostry A, Koehoorn M, Demers P: Costs andcompensation of work-related injuries in British Columbiasawmills.  Occup Environ Med 2007, 64(3):196-201.16. Ostry A, Maggi S, Tansey J, Dunn J, Hershler R, Chen L, Hertzman C:The Impact of Physical and Psychosocial Work Conditionson Suicide Among Sawmill Workers.  Scand J Public Health 2007,35:265-271.17. Transport Canada: Canadian motor vehicle traffic collision statistics.Ottawa 2002.18. Alberta Transportation: Alberta traffic collision statistics. Edmonton2003.19. Macpherson AK, To TM, Parkin PC, Moldofsky B, Wright JG, Chip-man ML, Macarthur C: Urban/rural variation in children's bicy-cle-related injuries.  Accident Anal Prev 2004, 36(4):649-54.20. Dryden DM, Saunders LD, Rowe BH, May LA, Yiannakoulias N, Sven-son LW, Schopflocher DP, Voaklander DC: The epidemiology oftraumatic spinal cord injury in Alberta, Canada.  Can J NeurolSci 2003, 30(2):113-21.21. Kmet L, Macarthur C: Urban-rural differences in motor vehiclecrash fatality and hospitalization rates among children andyouth.  Accident Anal Prev 2006, 38(1):122-7.22. Boland M, Staines A, Fitzpatrick P, Scallan E: Urban-rural variationin mortality and hospital admission rates for unintentionalinjury in Ireland.  Inj Prev 2005, 11:38-42.23. O'Connor PJ: Differentials in poisoning rates of young Austral-ian children according to residential location and geographi-cal remoteness.  Inj Prev 2005, 11:204-206.24. Jessop EG: Deprivation and mortality in non-metropolitanareas of England and Wales.  J Epidemiol Commun H 1996,50:524-526.25. Dansys Consultants Inc: Domestic Homicides Involving the Use of Fire-arms. WD 1992-20e Ottawa: Research & Development Directorate,Department of Justice Canada; 1992. 26. Canadian Institute for Health Information: How healthy are rural Cana-dians? An assessment of their health status and health determinantsOttawa: Canadian Institute for Health Information; 2006. 27. Sev'er A, Dawson M, Johnson H: Guest Editors' Introduction:Lethal and nonlethal violence against women by intimatepartners: Trends and prospects in the United States, theUnited Kingdom, and Canada.  Violence Against Women 2004,10:563-576.28. AuCoin K, Statistics Canada, Canadian Centre for Justice Statistics:2005. Family violence in Canada: A statistical profile Cat no 85-224-XIE.29. Thurston WE, Patten S, Lagendyk LE: Prevalence of violenceagainst women reported in a rural health region.  Can J RuralMed 2006, 11:259-267.30. Wholey D, Moscovice I, Hietpas T, Holtzman J: The environmentalcontext of patient safety and medical errors.  J Rural Health2004, 20:304-313.31. Coburn AF, Wakefield M, Casey M, Moscovice I, Payne S, Loux S:Assuring rural hospital patient safety: what should be the pri-orities?  J Rural Health 2004, 20:314-326.32. Romano PS, Geppert JJ, Davies S, Miller MR, Elixhauser A, McDonaldKM: A national profile of patient safety in U.S. hospitals.Health Aff (Millwood) 2003, 22:154-166.33. Agency for Healthcare Research and Quality: Nationwide InpatientSample (NIS). Healthcare Cost and Utilization Project (HCUP). Rock-ville; 2000. 34. Brennan TA, Leape LL, Laird NM, Hebert L, Localio AR, LawthersAG, Newhouse JP, Weiler PC, Hiatt HH: Incidence of adverseevents and negligence in hospitalized patients. Results of theHarvard Medical Practice Study I.  N Engl J Med 1991,324:370-376.35. Chamberlayne B, Green B, Barer ML, Hertzman C, Lawrence WJ,Sheps SB: Creating a population-based linked health database:a new resource for health services research.  Can J Public Health1998, 89:270-273.Pre-publication historyThe pre-publication history for this paper can be accessedhere:http://www.biomedcentral.com/1471-2458/9/432/prepubyours — you keep the copyrightSubmit your manuscript here:http://www.biomedcentral.com/info/publishing_adv.aspBioMedcentralPage 7 of 7(page number not for citation purposes)Ottawa 2005.


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