Article Text


Poster session 3

Statistics from


K. Frowen1, J. Cromie2, R. Nixon1.1Occupational Dermatology Research & Education Centre, Melbourne, Victoria, Australia; 2La Trobe University, Melbourne, Victoria, Australia

Introduction: Statistics for occupational contact dermatitis in Australia are gathered from workers’ compensation (WC) data, and previous research has indicated that this data underestimates occurrence by as much as 400%. This study investigated factors that might influence the decisions of workers as to whether or not to claim WC.

Methods: A self administered questionnaire was posted to 168 individuals diagnosed with significantly work related occupational contact dermatitis at a specialised occupational dermatology clinic, therefore fulfilling valid claim criteria under the WC scheme operating in the state of Victoria.

Results: We analysed 70 completed responses. Ages ranged from 18–65 years; there were 40 women (57%) and 30 men (43%). Only 40% of respondents had claimed WC, with females being significantly (p<0.05) less likely to claim. Respondents who had dermatitis present for less than 6 months were also less likely to claim, along with those aged under 45 years. The occupational groups of the respondents included 37% healthcare workers, 10% hairdressers, 7% food handlers, and of the industry groups, 29% worked in hospitals, 24% in manufacturing, 10% in hairdressing salons, and 7% each in vehicle maintenance, food service, and trades. Only 4/20 healthcare workers and 1/ 7 hairdressers claimed, presumably highlighting the predominance of females in these occupations. The other occupational groups were a little more evenly split. Of the respondents, 31% no longer worked for the same employer; however, 90% of respondents were still employed. Those who did not claim WC lost less time from work than those who claimed, but more of the non-claimants still had skin problems quite often or constantly than did the claimants. For non-claimants, 28.5% had all or some of their medical and/or lost time costs paid by their employer, and only 18% of claimants had all of their costs paid by their employer or WC insurer. Although the sample size was small, interesting data were obtained from the qualitative responses.

Conclusions: It would appear that gender, occupation, and duration of skin condition impact upon a worker’s decision to claim WC.


H. Y. Chang1, C. Y. Tsai1, Y. Q. Lin1, T. S. Shih2.1Department of Environmental and Occupational Health, Medical College, National Cheng Kung University, Tainan, Taiwan; 2Institute of Occupational Safety and Health, Council of Labor Affairs, Taipei, Taiwan

Introduction: Many chemical contaminants frequently used in occupational environments pose skin absorption concerns.

Objective: To determine the contribution of dermal exposure to the level of N,N-dimethylformamide (DMF) in the body across a work week between two groups with similar respiratory DMF exposure levels.

Methods: In this study, we recruited 25 workers in a synthetic leather factory (high dermal DMF exposure) and 20 in a copper laminate circuit board factory (low dermal DMF exposure) as DMF exposed groups, along with 20 age and sex matched control subjects. Environmental monitoring of DMF exposure including breathing zone monitoring during the full work shift, dermal exposure of hands and forearms and levels of urinary N-methylformamide (NMF) at pre-shift were implemented for each exposed participant for five consecutive days (from W1 to W5) and for the control subjects on the first and the fifth days.

Results: Environmental and biological monitoring showed no detectable values in controls and similar accumulations in the two groups of monitored workers. The geometric mean (geometric SD) airborne DMF concentration, was insignificantly lower for one group (3.98 (1.91) ppm) than for the other (4.49 (1.84) ppm). Dermal DMF exposure and urinary NMF values, however, were significantly higher for that same group.

Conclusions: This study has demonstrated that dermal exposure to DMF with 5 day consecutive occupational exposure could result in the accumulation of a significant DMF body burden. The long term exposure response under both repeated and intermittent conditions with substantial skin exposure should be investigated.


N. Chau, F. Guillemin, J. F. Ravaud, J. P. Meyer, J. Sanchez, A. Bhattacherjee, E. Bourgkard, S. Guillaume, J. P. Michaely, C. Otero Sierra, B. Legras, A. Dazord, M. Choquet, L. Méjean, N. Tubiana-Rufi, Y. Schléret, J. M. Mur.INSERM—INRS, Département Epidémiologie en Entreprises, Vandoeuvre-lès-Nancy, France

Introduction: Demanding work conditions with multiple hazards can increase occupational injuries. This study assessed the relationship of demanding work conditions with occupational injuries among employed people.

Methods: The sample included 2826 working subjects randomly selected from the Lorraine (northeastern France). A postal questionnaire was filled in by each subject. It included sociodemographic characteristics, smoking habit, alcohol use, living conditions, occupational injuries during the 2 year period before the survey, and highly demanding occupational hazards (use of hammer, screwdriver, hand tools with vibration, and machine tools; shocks; workplace with vibration; incorrect work posture; noise, heat, cold, artificial light, and heat radiation; standing about and walking; tasks at high level, work in adverse climates; handling objects; working on a production line; pace; and physical and mental workload). The data were analysed using adjusted odds ratios.

Results: The annual incidence rate of occupational injuries was 4.0%. The hazards with significant odds ratios adjusted on age and sex were: hand tools with vibration (OR 3.03; 95% confidence interval (CI) 1.99 to 4.61), machine tools (2.40; 1.30 to 4.44), shocks (2.77; 1.54 to 4.99), workplace vibration (2.20; 1.43 to 3.61), incorrect work posture (2.25; 1.65 to 3.07), noise (1.79; 1.36 to 2.35), heat (1.97; 1.48 to 2.63), cold (2.38; 1.78 to 3.19), tasks at high level (4.52; 2.85 to 7.16), work in adverse climates (2.72; 1.84 to 4.02), handling objects (3.30; 2.46 to 4.42), pace (1,79; 1.33 to 2.41), and physical workload (3.58; 2.60 to 4.94). A dose–response effect was found between the number of hazards (NH) and occupational injuries: compared with the NH 0 group, the groups NH 1–2, NH 3–4, and NH ⩾5 had OR (95% CI) adjusted on age, sex, job, smoking habit, and alcohol use of 2.58 (1.66 to 4.02), 6.15 (3.51 to 10.79), and 7.63 (3.58 to 16.25) respectively.

Conclusions: Demanding work conditions may increase the risk of injuries. Cumulative exposure to several hazards strongly increases the incidence of injuries. Prevention should reduce occupational hazards.


V. Santana, R. Oliveira.Program of Environmental and Workers’ Health, Institute of Collective Health, Federal University of Bahia, Brazil

Introduction: We describe non-fatal occupational injuries among construction workers and the factors associated with them, identified through a community based survey carried out in the city of Salvador, the capital of Bahia State, Brazil, are described.

Methods: All 1947 male individuals identified in a random cluster area single stage sample who reported having a paid job and were between 10 and 65 years of age comprised the study population. Data were obtained from household individual interviews. Narrative data about work injuries were also collected and analysed.

Results: There were 287 construction workers, who were older, more frequently black, poorer, and had fewer school years than those in other trades (n = 1660). Most of the construction workers reported having informal job contracts (65.8%). The annual incidence rate of non-fatal work injuries was estimated as 8.5/100 full time equivalent, and was higher among construction workers (incidence rate ratio 1.72; 95% confidence interval 1.49 to 7.42) compared with workers of other trades. Poor self perceived health and <2 years of experience were associated with these injuries. Based on narrative data, lack of information and access to protective equipment were reported. It was also observed that workers tend to blame themselves for the accident occurrence.

Conclusions: Formalisation of job contracts, education, training, and reinforced worksite inspections are strongly recommended in the construction industry in Brazil.


B. Waldvogel1, C. Salim2.1Seade Fundation, São Paulo, São Paulo, Brazil; 2Fundacentro/Ministry of Labor and Employment, Belo Horizonte, Minas Gerais, Brazil

Introduction: This paper seeks to analyse, compare, and debate the results of research carried out by Fundacentro and the SEADE Foundation on health and safety conditions in the transport sector in the Belo Horizonte and São Paulo metropolitan regions. The research sought to identify and measure the occurrence of work accidents among professionals in the sector, with emphasis on drivers and fare collectors, through careful data gathering over two partly superimposed periods; between 1997 and 1999 in São Paulo and 1998 and 2000 in Belo Horizonte.

Methods: The research was carried out in the agencies of the National Social Security Institute (INSS). It was based on a thorough documentary research of the information contained in the official communication of work accidents forms and in any documents that the accident generated, such as police reports. The INSS developed three categories: typical work accidents, transport to and from work accidents, and work related illnesses.

Results: Using selected demographic, socioeconomic, and epidemiological variables, it was possible to measure differences according to marital status, consequences (temporary incapacity, death or permanent injury), time off work, classification of the accident, age group, type of vehicle used, detailed occupation, time of the accident, hours worked, causal agent and, finally, parts of the body affected.

Conclusions: Compared with other similar exercises, the research is marked by the coverage of 3 year periods, the wealth of information raised, the reconstruction of the coding lists of selected variables and, finally, the new relationships established between variables; factors, without doubt, that contribute to the reduction of the occurrence of random factors in the development of the indicators used, widen the scope of the narrow information available in the computerised files of the INSS, and furnish new and important insights for understanding the causes of work accidents in Brazil.


F. M. Fischer1, D. C. Oliveira2, R. Nagai1, L. R. Teixeira1, M. Lombardi-Júnior1, M. R. D. O. Latorre3, S. P. Cooper4.1Department of Enviromnental Health, School of Public Health, University of São Paulo, Brazil; 2Nursing Department, State University of Rio de Janeiro, Brazil; 3Department of Epidemiology, School of Public Health, University of São Paulo, Brazil; 4Texas A&M School of Rural Public Health, Bryan, TX, USA

Introduction: In Brazil in 1997, 4314 adolescents (up to 18 years old) experienced work injuries. Usually they worked in informal jobs, as “office boy”, cleaners, shop clerks. or general helpers.

Objective: To evaluate the variables associated with work injuries among adolescents.

Methods: The study was carried out in a public school of São Paulo, Brazil. All students attending high school classes in the evening period (17:30–22:30) of 14–18 years old were invited to participate in this study. Of these, 354 students (184 current workers, 85 unemployed, and 85 non-workers) answered a questionnaire about living and health conditions. Only those current and former employed answered a questionnaire about working conditions. Association among variables were initially tested using χ2 with Yates’ correction, followed by multiple logistic regression to examine risk factors for occupational injuries. Stepwise forward selection was used for regression modelling. A 5% level of significance was used in all analysis.

Results: Work injuries were reported by 21% of the teenage workers. The univariate analysis showed association between work injuries and high psychological and physical job demands, bad working conditions, and having a job as cleaner, cashier, or general helper. The multivariate analysis also showed work injuries associated with poor environmental conditions (high psychological job demands, thermal and accoustic discomfort, toxic substances).

Conclusions: It is necessary to include an educational programme on occupational health in the high school curriculum.


K. J. Nielsen, O. Carstensen, K. Rasmussen, D. Glasscock.Department of Occupational Medicine, Herning Hospital, Denmark

Introduction: Registration and in depth analysis of occupational accidents and “near misses” is becoming a common preventive strategy for major industrial plants all over the world. We report the main conclusions and recommendations from a recently concluded Danish study of the implementation of an in depth occupational accident registration system in three industrial plants.

Methods: The study used both quantitative (questionnaires administered to all employees involved in production) and qualitative (various types of interviews and observations) data to assess both the employees’ and the management’s perception of and attitude towards safety, registration, and occupational accidents, both before and after the 2.5 year study period. After the first assesment, a new registration system was introduced in the plants and the implementation process and success of the system were studied.

Results: All three plants experienced a decline in the number of lost time accidents (LTAs) during the study period. In two of the plants, the implementation of the new registration system succeeded, whereas it failed in the third plant. In the two plants where the system was successfully implemented, the decline in LTAs was accompanied by an increase in the number of reported minor incidents and near misses.

Conclusions: It is possible to pursue a preventive strategy based on the registration and indepth analysis of occupational accidents and near misses, but it is necessary to focus on both management’s and employees’ perceptions of and attitudes towards safety, registration, and occupational accidents, and the management has to continuously dedicate the necessary time and resources to get (and keep) the system up and running.


S. Porru, D. Placidi, A. Carta, L. Alessio.Institute of Occupational Health, University of Brescia, Italy

Introduction: The first analysis of occupational accidents in an Italian iron foundry showed high accident rates, disabilities, fatal injury, and poor prevention, creating need for an urgent and thorough intervention.

Methods: Starting from 2000, the occupational physician (OP) promoted the institution of an accident prevention team including safety management, OP, and workers’ delegates. The team had periodical meetings, developed a detailed accident report form, and implemented first aid units and workforce education. Interventions occurred in general safety, maintenance, protective equipment, and work procedures. The issues of risk assessment, return to work, and job fitness were specifically addressed in health surveillance. Statistical analysis of accident occurred in 1996–2003 was performed. Record linkage enabled the analysis of several individual health variables. A cross sectional questionnaire evaluated job satisfaction, psychophysical conditions, chronic fatigue, personality, shift adaptation, and sleep.

Results: There were 480 accidents and 1700 person years (white men in “blue collar” jobs) analysed. The average incidence rate was 26.6%/year (range 14.4 to 40.3). Accidents occurred mainly in finishing, foundry, and maintenance; more than 60% of the accident were contusions, foreign ocular bodies, and burns. Pre-intervention and post-intervention analysis showed a significant decrease in frequency (p = 0.03), number injured more than once a year (p = 0.003), and duration (p = 0.05). A non-statistically significant decrease was observed for accidents with >20 work days lost. No relevant effects were noted for alcohol or shiftwork. A borderline statistically significant association was noted for sleep inadequacy, neuroticism, and injury duration. Health surveillance allowed proper human resources evaluation and effective management of return to work. The benefits of such actions outweighed their costs.

Conclusions: Accidents can be effectively prevented only in the context of a comprehensive multidisciplinary approach where the OP is an ideal interface between workforce and management, their contribution being mainly in priority setting, human factor assessment and evaluation, and counselling. Our experience in an iron foundry supported this approach.


C. W. Martin, K. Noertjojo, A. McNestry, C. T. Dunn.Clinical Services Department, Workers’ Compensation Board of British Columbia, Richmond, BC, Canada

Introduction: As a preliminary step in the effort by the Workers’ Compensation Board of British Columbia (WCBBC) to further understand and improve on how workers with mild traumatic brain injury (MTBI) may best be assessed, diagnosed, treated, compensated, and ultimately reintegrated into the workforce and community, a literature review and internal data analysis were undertaken. The internal data analysis is reported in this presentation.

Methods: Data was extracted from the WCBBC administrative data warehouse. The extraction was based on criteria using (ICD 9 codes 08500, 08501, 08502, 08503, 08504, 08505, and 08509) and/or nature of injury (concussion) and/or body part (brain). Analysis was limited to data from the period 1987–2001.

Results: In this period, 8260 claims were filed under MTBI. Of these, 98.6% were classified as concussion, and 46 (0.6%) were registered as having multiple injuries. The average incidence of MTBI was about 0.3% (range 0.20 to 0.43%) annually. Over half (52.6%) of claimants submitted their claims within the first week of the injury. By the third week of injury, 95.5% of MTBI related claims had been submitted. The average age of claimants was relatively stable at 30 years (range 34 to 37). Almost half of the claimants were aged 21–35 years old. Men were 3–4 times more likely to have an MTBI claim compared with women; however, there was a significant increase of female MTBI claimants across time, who tended to be older than their male counterparts. Of the MTBI claims, 96.2% were awarded short term disability (STD) benefits. About half of these were on STD benefits for ⩽7 days and 14.7% for >70 days. Overall the median STD of MTBI claims was 8 days. Women had significantly lower median STD compared with men. Overall, the older the claimants, the longer the median STD. Even though there was a trend that median claims costs were rising over time, the opposite tendency was observed for the annual total claim costs.

Conclusions: The incidence of MTBI at the WCBBC remains stable across time. The majority of MTBI claimants were awarded STD. Multiple linear regression model revealed that sex, age, and length of time between injury and claim submission were significant predictors of the duration of STD.


E. Stikova1, B. Pierkoska2, A. Kuka3.1School of Medicine, University St. Cyril and Methodius, Skopje, Macedonia; 2Pedagogical Faculty, University St. Cyril and Methodius, Skopje, Macedonia; 3Republic Institute for Health Protection, Skopje, Macedonia

Introduction: In the past years of transition of the Republic of Macedonia to an independent state, the total number of employees was reduced by 20%, and the unemployment rate rose to 31.9. We present here problems in occupational health and safety at work.

Methods: The data from official reports for work related injuries and deaths registered in past 10 years was used for statistical analysis.

Results: The mortality rate due to work related accidents was reduced from 1.66 per 100 000 in 1991 to 0.2 per 100 000 in 2001, the number of people injured at work decreased from 39093/year to 1383/year, and the incidence rate (IR) decreased from 1150.3 per 100 000 to 464.4 per 100 000. The most marked decrease occurred in 2001, when the IR was two times lower than the previous period. The situation is similar in all fields of economic activities. The highest IR is in mining and industry (average IR 1337.1 per 100 000; range 684.4 to 1791.5), with construction second (average IR 1296.0 per 100 000; range 675.4 to 1497.7), and agriculture third, with an average of 150 work related injuries (IR 952. 7 per 100 000; range 293.6 to 1465.2). The IR in these three employment areas is 2–3 times higher than among other workers. The largest decrease in accidents was seen among agricultural workers. The reason for the decrease in these sectors is possibly because of the economic situation in the country and decreased activities in the aforementioned sectors.

Conclusion: Compared with EU countries, the situation in Macedonia seems to be much better; however this is possibly a false reflection of the situation, because despite these decreases in incident ratesthere are problems with registration of injuries, as a national register is not yet established.


K. V. Duong, N. N. Nguyen, T. B. Ta, N. Q. Pham.National Institute of Occupational and Environmental Health,Vietnam

Introduction: The footwear industry in Vietnam is expanding rapidly and is recognised as one of the key industries leading Vietnam’s economic development. Because of the typical features of the industry, employees come into contact with substances and elements that are harmful to their health. Assembly lines are not synchronic and are out of date, and ergonomic factors at workplaces are risk factors for musculoskeletal disorders (MSDs) for workers, particularly women, who make up about 85% of the labour force.

Methods: A cross sectional study was carried out on 240 female workers in two footwear factories in Hanoi. Ergonomic analysis techniques and questionnaires on MSDs were employed to analyse the ergonomics of working positions and its relation to MSDs.

Results: The results showed that although workers were engaged in machinery assembly lines, most processes had to be operated manually. Many working positions were arranged incorrectly such as height of chairs (66–67 cm) and drying hangers (131.5–179 cm) being too high and unsuitable for Vietnamese dimensions, causing awkward working postures (bending, twisting, reaching up). The operation frequencies in some processes were very high (35–45 times/minute) and monotonous. Furthermore, the workers have to work with quite heavy shoe forms (1200–2000 g) with 2500–3500 shoes per workshift, so the effort required, particularly by hands and forearms was high. After the working day, 85.4% workers complained of MSDs. The highest prevalence of complaints were in the lower back (55%), shoulders (48.8%), wrists (37.1%), and hands (23.9%). The relationship between the MSD complaints and work characteristics was clearly shown when comparing the prevalences of MSDs among five groups of workers who had to perform different tasks with different work pieces and in different working positions (p<0.01).

Conclusions: The study provides some evidences of the relationship between ergonomic factors and MSDs in female workers in the footwear industry. Recommendations were suggested in order to reduce the hazards of these factors and to prevent MSDs in workers.


T. H. Nguyen, T. T. Nguyen.National Institute of Occupational & Environmental Health, Vietnam

Introduction: In Vietnam, many workers are exposed to whole body vibration. Low back pain (LBP) is the major health problem for these workers.

Objective: To investigate low back pain in workers exposed to whole body vibration.

Methods: We examined 126 lorry drivers and 102 workers in control group through interviews, clinical examination, and x ray scans.

Results: The results showed that among lorry drivers the percentage of LBP was very high (70.6% LBP in history, 59.5% LBP at present) and significantly higher than that among the control group (odds ratio (OR) 5.26 in history, 4.08 at present; p<0.001). Lorry drivers suffered mainly from LBP for <7 days (69.7%), and it occurred mainly after the work shift (88.7%). The degree and frequency of LBP in lorry drivers was also higher than that among the control group (p<0.05). The percentage of positive Schober test for lorry drivers was significantly higher than that among the control group (41.6% and 21.9%; p<0.05). In particular, the percentage of vertebral degeneration on x ray film of lorry drivers was significantly higher than that among the control group (OR 6.35; p<0.02).

Conclusion: We recommend that appropriate methods should be applied in order to reduce the rate of LBP in workers exposed whole body vibration.


B. D. Nguyen.National Institute of Occupational and Environmental Health, Vietnam

Introduction: Dentistry is one of the demanding professions, which requires a lot of concentration and precision. The major health problem among dentists is musculoskeletal discomfort and occupational stress when performing their tasks.

Objective: To investigate the musculoskeletal problems and occupational stress among the dentists and dental assistants in a Dental Clinic.

Methods: All 19 dentists and dental assistants were observed, interviewed by questionnaire and their working postures were assessed by the rapid upper limb assessment method.

Results: The results of this study showed that the working posture of dentists was very awkward (static and prolonged sitting, raised shoulders, bent and twisted neck, bent trunk) and needed to change immediately. For dental assistants, the posture was more comfortable (sitting or standing or walking) and tolerable. Because of the awkward posture of the subjects when performing their tasks, 89% of dentists and 80% of dental assistants reported neck problems. More than 50% of the subjects reported pain in both shoulders. Back pain was reported by a third of the dentists. It is interesting to note that more musculoskeletal problems were found in dental assistants than in dentists. Moreover, women suffered from musculoskeletal discomfort more than men, particularly in the shoulders (χ2 1 d.f. p<0.05). Limitation in work and leisure activity was caused by low back (20%), neck (47%), and shoulder pain (54%). Absenteeism from work was also caused by musculoskeletal problems (back 20%, shoulder 46%, and neck 47%). Furthermore, 30%, 46%, and 33% of the subjects had sought medical help for low back, shoulder, and neck pain, respectively. It was noted during the interview and from the results of the questionnaire that all subjects have a high level of stress at work. The most reported cause of stress was time pressure, which results in mental fatigue. Another major cause of stress was coping with difficult patients. A statistically significant difference (p<0.01) was obtained between dentists and dental assistants.

Conclusion: Both dentists and dental assistants in a dental clinic experienced musculoskeletal discomfort when performing their tasks and also experienced psychosocial problems. It is therefore recommended that ergonomic changes should be implemented.


J. N. Saraji, A. Mousavi, S. J. Shahtaheri, M. Pourmahabadian.Department of Occupational Health, School of Public Health, Tehran University of Medical Sciences, Tehran 14155–6446, Iran

Introduction: In order to evaluate and describe the occupational risk factors of upper extremity musculoskeletal disorders (UEMSDs), a calculation method for assessment of exposure to repetitive movements of the upper limbs is necessary.

Methods: The method used was conceptually based on the procedure recommended by NIOSH for calculating the lifting indices in manual handling activities. Exposure indices were used to obtain integrated and concise assessment of the various risk factors. In this study, we used four data gathering methods: an observational method (for jobs and tasks analysis featuring repetitive movements), interview (to obtain information about upper limb musculoskeletal disorders), the Nordic Musculoskeletal Questionnaire, and the Occupational Repetitive Actions (OCRA) method.

Results: We studied 385 people working in 33 various jobs in a textile factory in Ghaemshahr city, Iran. The results showed that 0.86% of work tasks had low, 87.93% had moderate, and 11.21% had high risk level. Mean exposure indices in four jobs (primary spinning, spinning, primary weaving, and weaving) had significant differences (p = 0.003). The weaving job had the highest risk of occurrence of upper extremity musculoskeletal disorders. A significant difference was seen between the mean values of right and left hand exposure indices (p<0.001). There was a significant relationship between UEMSDs and age, job history, height, weight, education level, and type of job (p<0.05).

Conclusion: It was concluded that, the OCRA method described in this study can be a useful method for evaluation of UEMSDs in repetitive activities in the textile industry.


N. K. Jørgensen, B. Bach.Department of Occupational Medicine, Nykøbing F Central Hospital, Nykøbing F, Denmark

Introduction: Low back pain may be a risk of occupational disease among train drivers. We surveyed metro train drivers for occurrence of low back pain.

Method: This was a cross sectional case–control design comprising 468 engine drivers and 757 mechanics from the Danish State Railways. Telephone interviews with a standard questionnaire included some questions used in the Danish National Cohort Survey of Occupational Diseases. The answers were analysed using the SPSS statistics program.

Results: There was a high response rate (95%). There was a very high occurrence of low back pain among the train drivers as measured by eight questions (18–44%), but also high frequencies in the control group (21–45%). There were no statistical differences, and no differences in treatment or examinations. Compared with an external control group (transport group in a Danish national cohort) this study showed very high frequencies of low back pain among train drivers as measured by eight questions (33–37% v 18–23%). Drivers with high exposure to swaying or vibrations had a higher risk for low back pain than those with low exposure (12–50% v 7–29%).

Conclusions: Danish train drivers in metro trains have a high risk of low back pain compared with other workers in the transport sector, but not higher than an internal control group.


N. K. Jørgensen, B. Bach.Department of Occupational Medicine, Nykøbing F Central Hospital, Nykøbing F, Denmark

Introduction: In a small survey of low back pain among train drivers we found high frequencies of low back pain among a smaller group of train drivers. This survey investigated train drivers on long distance trains.

Method: This was a case–control investigation of 1184 train drivers employed by the Danish State Railway compared with 766 mechanics working at the same company, and was based on telephone interviews with a standard questionnaire. Some of the questions are taken from the Danish National Cohort of Occupational Diseases Among Danish Workers. The answers were analysed with the SPSS statistics program.

Results: There was a high response rate (94%). There was a very high occurrence of low back pain among the train drivers as measured by eight different questions (43–50%), but also high frequencies in the control group (41–45%). There were no statistical differences, and no differences in treatment, examinations, or previous workload. Compared with an external control group (transport group in a Danish national cohort) there were very high frequencies of low back pain among train drivers (43–73% v 18–48%) as measured by eight different questions. Drivers with high exposure to swaying or vibrations had a higher risk of low back pains than those with low exposure (47–52% v 32–43%).

Conclusions: Danish train drivers have a higher risk of low back pain compared with other workers in the transport sector, but not higher than an internal control group.


Y. Rasoulzadeh1, R. Gholamnia1.1Departement of Occupational Health, Faculty of Health, Uremia Medical Science University, Uremia, Iran

Introduction: Work related musculoskeletal disorders (WRMD) are now recognised as a major occupational health problem and are linked to jobs that require continuous or repeated extreme or awkward postures. Owing to the nature of taxi driving as a job, such as prolonged working hours and sitting for long periods, we expected that the incidence rate of disorders among drivers would be high; however there are few studies on the incidence of WRMDs in taxi drivers.

Objective: To determine the type of musculoskeletal symptoms, their incidence rate, and their relationship with factors such as work time, age, and experience.

Methods: The study population was taxi drivers in Urmia city; 200 people (mean (SD) age 37.05(8.72) years; experience 7 (6.1) years) were selected at random. The data were collected using a musculoskeletal disorders questionnaire and statistical analysis was then performed.

Results: During the past year, 73.5% of drivers had experienced various types of musculoskeletal disorder. The highest incidence rates were related to low back and knee (joint stiffness) pain: 62.5% and 27.5% respectively. During this period, 32.9% of drivers saw a physician for these disorders and 1354 days were lost through resting at home and time in hospital. There was a significant relationship between experience and work time with musculoskeletal disorders (p = 0.012 and p = 0.001, respectively) while the relationship with age was not significant (p = 0.345).

Conclusion: The results showed that musculoskeletal disorders prevalence between taxi drivers is considerable and work related. Thus this study suggests that WRMD among taxi drivers can be decreased by using ergonomic principles during car design, especially the driver’s seat, low back support, and clearance spaces, and educating professional drivers in ergonomic principles.


P. R. Pinto1, G. C. Moraes1, B. V. Minghini2.1Workers Health Reference Center of São Paulo State (CEREST/SP), São Paulo, Brazil

Introduction: Cumulative trauma disorders (CTD) or work related musculoskeletal disorders (WMSD) include a group of musculoskeletal afflictions that often present difficulty in treatment owing to the chronic aspects of most cases and to the diverse aetiological origins of the problem. Once different therapeutic forms have been carried out by health services, it is necessary to evaluate their effectiveness and the progression of the clinic cases.

Objective: To present and test an upper extremity evaluation model.

Methods: The study was carried out in a group of 25 CTD/WMSD patients under treatment at Workers Health Reference Center of São Paulo State. The confidence level between observers for some topics of the physical examination, and the correlation between its results and the level of patients’ functional capacity were tested.

Results: The results indicated a good level of confidence between the observers for the special tests (intraclass correlation coefficient (ICC) 0.6 to 1), except for the modified Adson test (ICC 0.14 to 0.38). Similar results were found for the motion measurement ranges (K = 0.62 to 1). Pearson’s coefficient indicated a moderate correlation between the results of the physical examinations and the functional capacity level (p<0.05).

Conclusions: The proposed evaluation model seems to encompass important criteria for attendance of CTD/WMSD patients, and is a practical tool (ease of use, relative speed, and low costs) to be used in routine health services.


M. Ghaffari1, A. Alavi2, A. Alipour2, A. Farshad3, E. Vingard4.1Department of Public Health, Karolinska Institutet, Sweden; 2Department of Occupational Medicine, Irankhodro Company, Iran; 3Department of Occupational Health, Iran University of Medical Science, Iran; 4Department Clinical Neuroscience, Section for Personal Injury Prevention, Karolinska Institute, Sweden

Introduction: Most low back pain epidemiological data relates to developed and industrialised countries with high income; there is very little information about low back pain in the general population in developing and low income countries, and studies in the working population are even fewer.

Objective: To determine the prevalence and risk factors for low back pain among industrial workers in Iran.

Methods: The present study is from the first phase of the Irankhodro cohort (IKCO) study and involved a cross sectional, epidemiological analysis of low back pain among Iranian industrial workers with data gathered by means of a self reported questionnaire. For this study, one of the biggest industrial groups in Iran and also in the Middle East was chosen; IKCO is the biggest automotive manufacturing industry with more than 18 000 employes, all of whom were invited to complete the Standardised Nordic Questionnaire for Analysis of Musculoskeletal Symptoms. Working environment (physical exposures), living habits, and psychosocial factors were investigated.

Results: An entire group of 18 031employees at IKCO was chosen for this study. A total of 14384 completed surveys were received, giving a high overall response rate of 77.7%. Unadjusted odds ratio and multiple logistic regression with 95% confidence interval were applied for analysis. Of the participants, 20.8% had had low back pain during the past 12 months, projecting to about 2859 cases. The prevalence in women was 6.7% higher than in men. Of these cases, 30.4% had been prevented from working, and 50.7% reported low back pain during the past 7 days.

Conclusion: The result shows that LBP is prevalent among workers at IKCO, and there are significant associations between LBP and female sex, regular exercise, lifting heavy items, having sitting position, enough variation at work, and cooperation with fellow workerd. The 1 year prevalence of LBP in our study was higher than a similar survey in the USA.


F. Gobba1–4, S. Martinelli3, G. Artioli2, M. Vinceti2, M. Bergomi2, L. Roccatto4.1Cattedra di Medicina del Lavoro, Italy; 2Corso di Laurea in Infermieristica, sede di Reggio Emilia, Italy; 3Scuola di Specializzazione in Medicina del Lavor, Italyo; 4Dottorato di Ricerca in Sanità Pubblica, Università di Modena e Reggio Emilia, Facoltà di Medicina e Chirurgia, Modena, Italy

Introduction: Low back pain (LBP) is an important occupational risk in nursing personnel, mainly related to the heavy physical demands of patient handling. Intervention programmes for the prevention of LBP have been developed, but their effectiveness has been seldom evaluated. We studied the prevalence of self reported LBP in a group of nursing personnel, its relationship with various occupational and non-occupational factors, and the possible impact of a prevention programme.

Methods: We studied 180 hospital nursing professionals (89% women; mean (SD) age 39.5 (7.3) years) working in different units (medicine, surgery, emergency). The prevalence of self reported LBP episodes during the last 12 months, various occupational and non-occupational risk factors, participation in a prevention programme, and the actual application of learned preventive measures were studied by questionnaire.

Results: The prevalence of LBP in the whole group was 64%, without any difference among different working unit (medicine, surgery, emergency). The number of LBP episodes during the last 12 months was 1 for 26% of the workers, 2–4 for 31%, and >4 for 7%. About half of the subjects believed that their LBP was related to work, mainly to patient handling. This belief is somewhat supported by the observation of a tendency toward an increase in LBP prevalence related to the number of years of work in hospitals, but not with age, at multivariate analysis. Furthermore, the prevalence of LBP was lower (27%) among subjects who received adequate training for LBP prevention and report routine application of correct work techniques learned, compared with 67% in subjects not applying any prevention technique. Regular physical activity seems to reduce LBP prevalence.

Conclusions: The results of this study confirm the relevance of LBP in nursing personnel, and support the hypothesis that routine application of adequate preventive techniques learned during prevention programmes can significantly reduce risk.


P. Frost1, J. H. Andersen2, J. P. Haahr2.1Department Of Occupational Medicine, Aarhus University Hospital, Denmark; 2Department Of Occupational Medicine, Herning Hospital, Denmark

Introduction: Musculoskeletal complaints have always been common in working populations, but job loss and disability related to such complaints has increased in Western countries during the past decades. This study focuses on the potential for prevention of disability due to common musculoskeletal pain problems through modification of physical workloads and negative beliefs about the nature of the pain problems.

Methods: In 2002 we established a cohort of about 4100 subjects working in industry, service, or health care. All subjects completed a questionnaire on physical and psychosocial workloads, fear avoidance beliefs, somatisisation, and other factors with possible influence on disability problems due to pain complaints. By block randomisation at company level, subjects were allocated to one of two intervention groups or to the reference group. The interventions consist of a booklet containing evidence based information about common pain problems (group 1) or the same information in combination with systematic workplace screening for unacceptable levels of physical workload (group 2). Two years’ follow up on pain related sickness absence and alterations in fair avoidance beliefs were the main outcome measures to be evaluated. Companies reported all absence spells quarterly and then the specific cause of each absence spell was obtained by a postal questionnaire.

Results: By randomisation we achieved three comparable groups according to age, gender, work characteristics, fear avoidance beliefs, and other characteristics. The study is still ongoing; however, based on 1 year of follow up, about 30% experience an absence spell each quarter and complete information on the cause of absence has been obtained from about 80%.

Conclusions: It was possible to establish a randomised intervention study to evaluate the possible impact of evidence based information and workload modification on pain related sickness absence.


C. W. Martin, K. Noertjojo, A. McNestry, C. T. Dunn.Clinical Services Department, Workers’ Compensation Board of British Columbia, Richmond, BC, Canada

Introduction: As a part of a 2003 public awareness campaign on low back pain conducted by the Workers’ Compensation Board of British Columbia (WCBBC), the Clinical Services Department analysed the administrative data available at the WCBBC.

Methods: Data was extracted from the WCBBC administrative data warehouse. The extraction was based on broad criteria using a combination of various codes from ICD 9, nature of injury, and body parts in order to identify low back pain related claims. Analysis was limited to data from the period 1987–2001.

Results: In the period 1987–2001, 261 762 new claims were filed for low back related pain (LBP). These claims represented 6–11% of the claims filed in those years. The average incidence of LBP was 24% (range 22 to 25%). Approximately 75% of LBP claimants were men. The trend remained the same across the period measured. The mean (SD) and median age of LBP claimants was 36.9 (10.9) years and 36.0 years, respectively; 64% of the claimants were ⩽40 years old. There was a significant trend for women to be older than men at the time the claim were filed. Of the claimants, 97.5%, 2.3%, and 0.3% were awarded short term disability benefits (STD), long term disability benefits (LTD), and healthcare cost only (HCO), respectively. There was a significant trend across time that more claimants were awarded STD instead of LTD or HCO. Overall, the median number of days of STD was 12 days. Longer periods of STD were observed with increasing age of claimants. The total number of STD days among LBP claimants represented 19–43% of the total STD days awarded by the WCBBC in the period 1987–2001. The median LBP claim cost varied between Can$938 and Can$1483. Overall, total claim costs related to LBP paid by the WCBBC ranged from Can$63 million to Can$165 million, annually. An increase in the median claim cost and a decrease in the total number of claims paid were observed across time.

Conclusions: LBP represents a significant physical and financial burden on the workers and employers of British Columbia. Data on treatment modalities will be presented.


T. Morken1, S. Torp2, B. E. Moen1, T. Riise1.1Section for Occupational Medicine, University of Bergen, Norway; 2Vestfold University College, Norway

Introduction: Musculoskeletal disorders (MSD) frequently cause disability and reduced work capacity, thus these problems are receiving increasing focus in industrialised countries. Several risk factors for MSD in the working environment are known, and the risk factors vary between occupations. To achieve improvement of musculoskeletal health in occupations, we need to know more about the occupations at risk.

Objective: To investigate occupation specific differences in MSD, sickness absence, and psychosocial factors in Norway.

Methods: A cross sectional questionnaire study of a general population of 20 129 individuals aged 40–47 years in Hordaland County, Norway, was carried out. The classification of occupations was based on the International Standard Classification of Occupations (ISCO 88). We analysed the occupation specific prevalence of MSD, sickness absence due to MSD, social support, and psychological strain. Differences between groups were measured by χ2 and t tests.

Results: In this study, 42% of the workers reported having had MSD for more than 3 months in the past year. MSD were more frequent among women (47%) than men (37%) (p<0.001). There were significant occupation specific differences in prevalence of MSD, being highest for drivers (50%) for men and for clerical officers (60%) for women. In men the lowest prevalence was found for professionals and in women for technicians and associate professionals. Sickness absence due to MSD in the past year was significantly different reported among 25% of the men and 29% of the women. The gender difference was significant (p<0.001). In both male drivers and female clerical officers, 36% reported sickness absence due to MSD in the past year. The male drivers reported lower social support (p = 0.003) and higher strain (p<0.001) than the rest of the male study population. The female clerical officers reported higher strain (p<0.001), but the social support did not differ from the rest of the female study population.

Conclusion: In order to prevent MSD and sickness absence due to MSD in the working environment, measures should focus particularly on risk factors among male drivers and female clerical workers.


I. Kim1, S. B. Koh2, J. S. Kim3, D. M. Kang4, M. Son5, Y. Kim1, J. Song1.1Department of Occupational and Environmental Medicine, Hanyang University, Seoul, Korea; 2Department of Preventive Medicine, Yonsei University in Wonju, Wonju, Kangwon-Do, Korea; 3Department of Preventive Medicine, Seoul National University; Seoul, Korea; 4Department of Preventive and Occupational Medicine, Pusan National University, Pusan, Korea; 5Department of Preventive Medicine, Kangwon National University, Chunchon, Korea

Objective: To investigate the various factors related to work related musculoskeletal disorders (WMSD) of shipbuilding workers, and to elucidate the relationship between musculoskeletal symptoms and psychosocial factors, such as change in working conditions, job stress, and physical workload.

Method: We recruited 1059 shipbuilding workers for this study. A structured questionnaire was used to assess general characteristics, job stress, psychosocial wellbeing index, physical workload, change in working conditions, and information concerning musculoskeletal symptoms. We estimated the relationhips of job stress, physical workload, and intensity of labour to musculoskeletal symptoms using univariate and multiple logistic regression analyses.

Results: Prevalence rate of musculoskeletal disorders (one or more) was 89.5% by National Institute for Safety and Health criteria, with back (58.6%) and shoulder (56.3%) being highest. After adjustment for sociodemographic factors, posture factors (odds ratio 1.06; 95% CI 1.00 to 1.12), non-posture factors (1.17; 1.05 to 1.31), Borg scale (1.15; 1.00 to 1.32), relative work intensity increase (1.92; 1.08 to 3.41), labour flexibility increase (2.04; 1.04 to 4.01), high job demand (2.68; 1.48 to 4.88), and high risk stress group (13.50; 3.15 to 57.97) were all found to be significantly associated with musculoskeletal disorders.

Conclusions: As with the results of previous studies, work related musculoskeletal disorders were related to various risk factors such as stress, physical workload, and change in working condition. In particular, high job demand, increased intensification of work, and increased flexibility such as subcontracting, outsourcing and import of contingent work were strongly associated with increasing prevalence of WMSDs.


C. J. Burns1, K. F. Harrison2, B. L. Jammer1, D. Zuccarini3, B. Lafrance3.1The Dow Chemical Company, Midland, MI, USA; 2Dow Chemical Canada Inc. Fort Saskatchewan, AB, Canada; 3Statistics Canada, Ottawa, ON, Canada

Introduction: A cancer incidence and mortality study of the Dow Chemical Canada Inc. sites at Sarnia and Fort Saskatchewan was conducted.

Methods: We followed 5277 men and 1301 women from 1950 to 1999. We observed 390 cancer incidence cases and 597 deaths due to all causes.

Results: Employees experienced lower death rates and cancer incidence rates than the general population for several major causes of death, including heart disease, respiratory cancer, and many other cancers. There were no cases of angiosarcoma of the liver. Among all male workers, we observed a lower mortality rate of prostate cancer (standardised mortality rate 0.79; 95% confidence interval 0.43 to 1.32), but a higher incidence rate of prostate cancer (1.22; 1.00 to 1.48). Among the Sarnia employees, the incidence of pleural neoplasms was increased (5 observed vs. 1.48 expected) (3.37; 1.09 to 7.86). These cancers were included in the 12 deaths with malignant mesothelioma at Sarnia.

Conclusions: As a surveillance tool, the principal objective of this analysis was to identify potential hazards of employment rather than potential hazards of specific chemical exposures. We observed a strong healthy worker effect with highly significant deficits for all causes, all malignant neoplasms, and many other causes of death. The higher incidence rates of prostate cancer may reflect increased screening among current employees and recent retirees. Past asbestos exposure prior to 1970 is probably a contributor to the deaths due to malignant mesothelioma, but is not reflected in lung cancer mortality. We found little indication of any major unidentified hazard either leading to mortality or cancer for current employees.


K. Mohajer1, A. Alavi2, S. H. Vosoughi3, B. Ghadimpour4.2Department of Occupational Medicine, Iran Khodro Company, Tehran, Iran; 3Department of Occupational Health & Safety, Iran Khodro Company, Tehran, Iran

Introduction: Musculoskeletal disorders (MSD) are very common disorders in industrial workplaces. According to reported studies, a risk factor for these disorders is workers’ incorrect posture. Therefore, this project was carried out to investigate the prevalence of MSD and the role of posture as a risk factor in the Iran Khodro Company. This research was undertaken because MSD are known to be common in this company, as confirmed by occupational medicine department records.

Methods: This project was conducted in Bodyshop 1 of the Iran Khodro Company; 53 workers at 12 workstations were studied, using the Ovako Working Posture Analysis System, an assessment method used to investigate problems of lifting and back injuries.

Results: The results of the working posture analysis showed that 37.73% of workers had scores of 1 and 2 (lower risk), while the number of workers with scores of 3 and 4 were 49.06% (n = 26) and 13.2% (n = 7) respectively. Occupational medicine department physical examination showed that back pain was the most prevalent MSD among the studied workers, especially in the workers with scores of 3 or 4, compared with the other workers.

Conclusions: We concluded that 26 workers need to have corrective measures to their working posture as soon as possible. It was also concluded that immediate corrective measures are required for the seven workers with an OWAS score of 4.


S. Porru1, D. Placidi1, M. Campagna1, A. Scotto1, di Carlo1, G. F. Tassi2, L. Alessio1.1Institute of Occupational Health, University of Brescia, Brescia, Italy; 2Pneumology Division, Spedali Civili Hospital, Brescia, Italy

Introduction: An active surveillance of cases of malignant mesotheliomas (MM) admitted to a major university hospital was carried out in a highly industrialised area in northern Italy, where MM incidence is about two cases per 100 000 population.

Methods: Patients affected by histologically confirmed MM were referred to an occupational physician (OP) for a face to face interview during hospital admission. A detailed lifetime history was collected, and a standardised questionnaire was completed and sent to the local and national MM registries. In selected cases, additional information on asbestos exposure was obtained from plant personnel, occupational health services, local health authorities, industrial hygienists, and the literature. Joint imaging readings, search for asbestos bodies in sputum or bronchoalveolar lavage, and histopathological re-evaluations were also performed. A full OP report was written, including statements on aetiological diagnosis, medicolegal obligations were addressed, and counselling was directed to patients and to some health professionals. Occupational histories were blindly coded (ISCO and ISIC classification) and an asbestos exposure matrix was used.

Results: Between 1996 and March 2004, 170 pleural MM (71% in men) and two peritoneal MM (both in women) were evaluated: 47.1% (90% males) were aetiologically attributed to occupational exposure to asbestos (32% metal industry, 20% construction, 15% transport, 15% asbestos manufacture), and 2% to environmental exposure to asbestos. About 50% of the patients with occupational MM had been compensated. The average time since first exposure to asbestos was 40.1 years (range 21–66). Full results of analysis by occupation, industry, and asbestos exposure coding will be discussed, with reference to Registry conclusions.

Conclusion: Our experience highlights the multifaceted contribution of the OP in the process of aetiological diagnosis in MM, with its medicolegal, individual, and social consequences; during exposure assessment, for their contribution to MM Registries and in design and conduct of epidemiological studies; and in cooperation with health professionals for planning and implementation of preventive actions, especially in workplaces.


M. A. V. Rêgo1, C. S. C. Sousa2.1Department of Preventive Medicine, School of Medicine, Federal University of Bahia, Brazil; 2Octávio Mangabeira Hospital, State Health Department, Bahia, Brazil

Introduction: The risk factors associated with non-Hodgkin’s lymphomas (NHL) do not explain the increase in the incidence rates of this disease during the past few decades. Among the possible causes, we discuss the role of some chemicals such as organic solvents; in particular, the specific action of benzene. Acute myeloid leukaemia is the main cancer related to exposure to benzene. Some studies aiming to investigate the relationship between leukaemia and benzene showed an increased risk of lymphoid neoplasm among exposed workers.

Objective: To discuss the role of benzene as a risk factor for NHL.

Methods: MedLine was searched for articles published between 1966 and 2003, using key words relating lymphoma and occupational or environmental exposure to benzene or organic solvents. Selected studies involved adult humans, were published in English, French, Spanish, or Portuguese, had an explicit evaluation of exposure to benzene, a definition of NHL as a specific group, and an epidemiological measure of association. The selected studies were classified according to the way they were performed, whether based in the workplace or the community, and were analysed considering the strength of the described associations and the control for confounder variables.

Results: We found 26 studies that fulfilled the criteria; all of the 13 studies based on occupational settings were cohort mortality studies, and among the 13 community based studies, 11 were case–control studies. Fifteen studies described positive results (six statistically significant) and 11 did not show an association. It is important to observe that a strong association was shown in the multicentric cohort study conducted in China. No study showed a dose–response realtionship.

Conclusions: Considering the limitations of each study, especially regarding the low number of individuals, and the leukaemogenic potential of benzene, these results suggest that exposure to benzene may contribute to the development of NHL. Further specific investigations about this issue are necessary.


P. Gustavsson1,2, M. Pollán3.1Department of Occupational and Environmental Health, Stockholm Centre for Public Health, Sweden; 2Division of Occupational Medicine, Department of Public Health Sciences, Karolinska Institute, Stockholm, Sweden; 3Cancer Epidemiology Unit, National Centre for Epidemiology, Carlos III Institute of Health, Madrid, Spain

Introduction: Alarming reports of increased mortality among Swedish anaesthetists led us to investigate the cancer incidence among Swedish physicians in general.

Methods: A cohort of all Swedish physicians (6932 men and 1599 women) was identified from the national census of 1970. The cohort was followed from 1971 to 1989. Cancer cases were identified from the National Cancer Register and deaths from the National Register of Deaths. Standardised incidence ratio (SIR) for cancer were computed using all gainfully employed people in Sweden to calculate the expected numbers.

Results: Total numbers of cancers were close to the expected (SIR 101; 95% confidence interval (CI) 93 to 110 for men, and 109; 91 to 128 for women). A rather pronounced excess of non-melanoma skin cancer (NMSC) was seen among the men, (260; 190 to 347), based on 46 cases. Women also showed an excess of NMSC, although numbers were lower: 4 cases v 1.7 expected. There was also an excess of female breast cancer (59 cases v 40.0 expected) (SIR 147; 95% CI112 to 190). Men showed an increased risk of bladder cancer (141; 106 to 183).

Conclusions: A similar investigation of physicians in Norway, Denmark, and Finland corroborated an increased risk of NMSC, and it is unlikely that this is a chance finding. Physicians in radiology and nuclear medicine are exposed to ionising radiation, which is a risk factor for NMSC, but exposure levels in modern work environments would not be expected to induce new cases. Possibly, an increased observance of skin problems among health care workers could contribute, although is not clear why this would be so much more expressed for skin cancer than for other forms of cancer. A higher risk of malignant melanoma has previously been noted in higher socioeconomic groups, probably due to sun exposure habits; possibly, the same mechanism could operate for NMSC, although this is not a well documented association.


A. Gilg Soit Ilg1, E. Imbernon1, M. Goldberg1, G. Launoy, theP. N. S. M. teams.1Institut de Veille Sanitaire, Département Santé Travail, Saint-Maurice, France; 2Registre des Tumeurs du Calvados, Caen, France

Introduction: Most previous estimates of mesothelioma incidence in France were based on theoretical models due to the lack of observed data. The French National Mesothelioma Surveillance Program (PNSM) was initiated in 1998 in order to provide an annual estimate of the national mesothelioma incidence. It collects, certifies (histological and clinical proof), and investigates mesothelioma cases in 21 administrative areas, corresponding to French cancer registries.

Methods: Data on mesothelioma incidence collected by the PNSM were compared with the malignant pleural cancer mortality rates, for men and women separately and by age and area. Mesothelioma deaths (ICD9–163) were known from the official national database of deaths by cause in France. The ratios of incidence/mortality (I/M) were computed. Geographical disparities regarding the collection of mesothelioma incident cases were analysed. We considered different hypotheses regarding the national I/M ratio in order to provide estimates of the national mesothelioma incidence.

Results: On all hypotheses considered, the estimates of the national incidence of mesothelioma were reasonably close: in 1998 and 1999 there were 620 and 560 mesothelioma cases respectively in men, and 160 and 215 cases in women, while the estimated incidence rate of mesothelioma was 2.16 and 2.02 per 100 000 in men and 0.55 and 0.71 in women.

Conclusions: Those results are in accordance with previous theoretical estimates provided by simulations, and are corroborated by a recent study using data from the French collaborative network of cancer registries (which covers 9.5% of the French population); they estimated that in the year 2000, 671 (95% confidence interval 535 to 807) new cases of mesothelioma occurred in men and 200 (123 to 277) in women.


C. Bigert, R. Jakobsson, P. Gustavsson.Department of Occupational and Environmental Health, Stockholm Centre for Public Health, Stockholm, Sweden

Introduction: Previous epidemiological studies have indicated an excess risk of lung cancer among professional drivers. It is not clear whether the excess risk is due to smoking habits, exposure to diesel exhaust, or other factors.

Objective: To investigate the risk of lung cancer among bus, taxi, and lorry drivers in relation to tobacco smoking habits and occupational exposure to diesel exhaust.

Methods: All new cases of lung cancer in males aged 40–75 years in Stockholm County, Sweden, were identified from 1985 to 1990. Both living and deceased cases were included. Referents were selected randomly within strata (age, inclusion year) of the study base. Questionnaires administered to subjects or their next of kin gave information on lifetime occupational history and tobacco smoking history. Occupational exposure to diesel exhaust was assessed individually by an industrial hygienist. Response rates of 87% and 85% resulted in 1038 cases and 2359 referents, respectively. In all, there were 145 cases and 230 referents who had worked as professional drivers for at least a year.

Results: There was an increased risk of lung cancer among professional drivers (relative risk 1.38; 95% CI 1.08 to 1.76). The risk excess was most pronounced among short distance lorry drivers. Bus drivers and taxi drivers but not long distance lorry drivers had a tendency to an increased risk. Adjustment for smoking habits reduced the risk (1.12; 0.86 to 1.45), and additional adjustment for diesel exhaust reduced it further (1.07; 0.75 to 1.53).

Conclusions: The increased risk of lung cancer in professional drivers can to a large extent be explained by their smoking habits and partly by exposure to diesel exhaust.


T. P. Brown, L. Rushton.MRC Institute for Environment and Health, Leicester, UK

Introduction: Respirable crystalline silica dust (RCSD) has been defined by the International Agency for Research on Cancer as a lung carcinogen, although there is considerable controversy over the role of silicosis in the development of lung cancer. This paper describes the cancer incidence among a cohort of British industrial silica sand workers.

Methods: Details of 2013 workers, employed for more than a year at one of seven quarries after 1970 were included in the analysis. RCSD samples collected after 1978 were used to assemble a job exposure matrix (JEM). These data were used to estimate exposure between 1971 and 1978 by back extrapolation of linear regression equations. The JEM was used to estimate cumulative RCSD exposure for each cohort member for whom job history was complete (n = 1636). Standardised incidence rates (SIR) were calculated using the national rates as comparison.

Results: Between 1970 and 1998, 160 cancers were registered, giving an SIR for all cancers of 79 (95% confidence interval 67 to 92). The incidence of gastrointestinal (59; 37 to 89) and genitourinary (66; 43 to 97) cancers was significantly reduced, although the incidence of kidney cancer at one quarry was significantly raised (four cases). Lung cancer incidence was slightly raised (110; 80 to 147) but showed no relationship with quarry or job group. SIRs for lung cancer were elevated for all categories of cumulative RCSD exposure, although there was a negative trend with increasing exposure.

Conclusions: These findings suggest a lack of association between cancer and RCSD exposure in this cohort, and in particular lung cancer, in contrast to studies of granite workers. Further investigation is required to investigate the occurrence of silicosis within the cohort, especially in relation to lung cancer.


Z. Mikoczy, L. Hagmar, U. Bergendorf, K. Jakobsson.Department of Occupational and Environmental Medicine, Lund University Hospital, Lund, Sweden

Introduction: The rubber industry has long been associated with cancer risks, especially bladder cancer, lung cancer, and leukaemias. Exposure levels have declined during the past decades, but it is not clear whether this has affected cancer risks in the “modern” rubber industry. In this follow up study we analysed mortality and cancer incidence in a cohort of Swedish rubber industry employees, first employed between 1965 and 2001.

Methods: A cohort of 14 955 workers (of which 5237 were women) from 11 rubber plants, who had been employed for at least 6 months in the industry, was established. The median year of birth was 1953, and the median follow up period was 18 years. Using linkage with Statistics Sweden and the Swedish Cancer Registry, causes of death and incident cancer cases were recorded. Cause specific mortality and cancer incidence 1965–2001 was calculated, based on calendar year, sex, and 5 year age group specific rates. Altogether 282 347 person years under risk were observed.

Results: The overall mortality was lower than expected (972 deaths observed; standardised mortality ratio (SMR) 0.85; 95% confidence interval (CI) 0.80 to 0.91). This was true in deaths from cardiovascular diseases (SMR 0.79) and respiratory diseases (SMR 0.77). A total of 644 incident cancer cases were observed, resulting in a standardised incidence ratio (SIR) of 0.96 (95% CI 0.88 to 1.03). Slight numerical, but non-significant, increases were observed for leukaemias (19 observed; SIR 1.16; 95% CI 0.70 to 1.81), and stomach cancer (25 observed; SIR 1.15; 95% CI 0.74 to 1.70). For lung cancer and bladder cancer we observed SIRs of 0.97 and 0.78, respectively. When reducing the cohort to workers with a minimum employment of five years, the slight numerical increase was still present for leukemias, but vanished for stomach cancer.

Conclusion: Overall, there were no indications of an increased cancer risk in the cohort of rubber workers. We were able to include the majority of Swedish rubber workers in our study, but the study was still underpowered with respect to site specific cancers. Thus, pooled studies are needed to investigate cancer risks in the “modern” rubber industry. A recently developed exposure matrix (EXASRUB), based on exposure assessment from the rubber industry in five European countries, may provide a useful tool for such studies.


E. Stikova1, A. Kuka2, B. Piperkoska3.1School of Medicine, University St. Cyril and Methodius, Skopje, Macedonia; 2Republic Institute for Health Protection, Skopje, Macedonia; 3Pedagogical Faculty, University St. Cyril and Methodius, Skopje, Macedonia

Introduction: The epidemiological transition in the Republic of Macedonia show an increasing trend of mortality rate (from 690/100 000 in 1990 to 880/100 000 in 2002). The second leading causes of death are malignant neoplasms, with average mortality rate of 130/100 000. Thi paper presents the most important epidemiological aspects of cancer incidence in the period 1995–2001.

Methods: The data were collected from three main sources: the official cancer database, histological reports, and death certificates. All deaths were coded according to ICD X and ICD 0. The incidence rate (IR) was classified by sex, age, place of residence, and occupation.

Results: The annual average number of new cancer cases was 5506. Incidence rate was 274.3/100 000, with a steady rising tendency (range 254.6 to 305.9/100 000), and was significantly higher in men. The male: female ratio varied with age group of the population. In the age group 20–44 years, the IR was two times higher in women (mainly from cancers of the breast and female genital organs). In age groups ⩾45 years, the IR in men, due to lung cancer. Regarding the age structure, the IR increased in every next (older) group interval, from 14.1 to 1310.8 per 100 000, with a ratio of 1:2.8:11.2:40.2:93.6. The urban: rural IR ratio was 0.46 with declining tendency in rural area (t1/t7 = 1.3). The analysis of data by occupation showed that among the workers in the agriculture sector the incidence rate was 7–10 times higher than in industry, and decrease from 482.0/100 000 in 1995 to 240/100 000 in 2001. The incidence rate was statistically higher in men. The annual consumption of pesticides decreased from 3.7 to 2.1 kg/hectare.

Conclusion: IR is much lower in Macedonia than countries of Europe and the EU, and similar than in other central and southeastern European countries. The study has identified cancer as a growing problem in Macedonia.


C. W. Martin, K. Noertjojo, C. T. Dunn, A. McNestry.Clinical Services Department, Workers’ Compensation Board of British Columbia, Richmond, BC, Canada

Introduction: As a preliminary step in the effort by the Workers’ Compensation Board of British Columbia (WCBBC) to understand and improve on how workers diagnosed with mesothelioma may best be treated and compensated, a literature review and internal data analysis were undertaken. The internal data analysis is reported in this presentation.

Methods: Data were extracted from the WCBBC administrative data warehouse. The extraction was based on criteria using (nature of injury code 00888, 30000, 31000, 31100, 31900, 31901, 33900, or 39900) and (source of injury code 05203 or 55100). Analysis was limited to data from the period 1987–2002.

Results: In the period 1987–2002, 325 claims were filed as mesothelioma. On average, 20 new mesothelioma cases (range 8 to 29) were filed with the WCBBC annually. Overall, the incidence of mesothelioma at the WCBBC was 10.77 per 100 000 new claims per year (range 4.09 to 22.49). Overall, the mean (SD) and median age at diagnosis of mesothelioma claimants were 64.6 (9.5) years and 64.0 years, respectively, with 56% of the claimants being between the ages of 55 and 70 years old. Of the registered mesothelioma cases 316 (97.2%) were men; 32% of the mesothelioma claimants worked in general construction and 27% worked in metallic and non-metallic mineral product environments. When the data was downloaded in March 2004, 88.6% of these claims were classified as fatal. The median number of days of short term disability awarded to these claimants was 226 days. Overall, the median mesothelioma claim costs Can$165 000. Comparison of the incidence and incidence rate data on mesothelioma for the province of British Columbia revealed a large discrepancy between mesothelioma cases registered at the WCBBC and the British Columbia Cancer Agency (BCCA). Overall, the incidence rate of mesothelioma as registered at the BCCA was twice as high as that registered at the WCBBC, which suggest many occupational mesotheliomas were not being reported to the WCBBC.

Conclusions: Mesothelioma represents a small number of claims at the WCBBC. There was a discrepancy between the number of mesothelioma cases as recorded by the WCBBC and BCCA that warrants further study.


Y. Ogawa1, K. Ohba2, R. Yoshida3, I. Mori1, H. Saitoh1, K. Kitamura1, A. Nakata1, Y. Matsumoto2.1National Institute of Industrial Health, Kawasaki, Japan; 2Department of Medical Hygiene, Kitasato University, Sagamihara, Japan; 3IARC, Lyon, France

Introduction: We studied the relationship between blood dioxin concentration and dioxin related gene expression among incinerator workers.

Methods: Blood samples were collected from fasting workers in the morning and allocated to two tubes. One was used for RNA extraction and another was frozen for later analysis of dioxins. CYP1A1, CYP1B1, AhR, hOGG1, β-actin, and GAPDH were analysed by real time PCR. Blood dioxin levels were analysed according to the standard Japanese manual.

Results: The study population consisted of 35 workers, including one woman. Mean age was 46.3 years (range 22 to 63); mean blood dioxin level was 11.8 pg TEQ/g fat (range 4.0 to 21.1). Dioxin congeners, of which all 35 samples had levels above the detection limit, were 1,2,3,7,8-PeCDD, 1,2,3,4,7,8-HxCDD, 1,2,3,4,6,7,8-HpCDD, OCDD, and 2,3,4,7,8-PeCDF. We divided the amounts of gene expressions of CYPs by each dioxin concentration and constructed histograms. Double instead of single peak distribution appeared in 1,2,3,7,8-PeCDD and 2,3,4,7,8-PeCDF sets. Two groups, (high and low induction groups) were extracted by cluster analysis. A strong correlation existed between the amount of CYP1B1 expressed and blood dioxin levels in the high induction group but there was no correlation in the low induction group.

Conclusion: Two groups were classified among incinerator workers according to CYP induction. The high induction group had high CYP expression, low blood dioxin level, and correlation between CYP1B1 and blood dioxins. Our results suggest that the two groups had different characteristics possibly depending on some type of gene polymorphism.


G. Assennato, L. Bisceglia, G. de Nichilo, M. E. Grassi.Occupational Medicine Division, Department of Internal and Public Medicine, University of Bari, Italy

Objective: To investigate the influence of genetic polymorphisms of the cytochrome p450 1A1 (CYP1A1), and glutathione S transferases μ (GSTM1) and θ (GSTT1) on urinary levels of 1 hydroxypyrene (1-HOP) in workers exposed to polycyclic aromatic hydrocarbons (PAHs).

Methods: We enrolled 355 coke oven workers currently exposed to PAHs from the coke oven plant of one of the largest European steel plants. The coke plant consists of three batteries (A, B, and C). Maintenance workers and industrial cleaners were included. A standardised questionnaire was administered to collect information on variables of interest. Biological samples were collected at the end of the work shift. After log transformation of the 1-HOP levels, univariate and multivariate analyses were performed.

Results: The median value of urinary 1-HOP was 1.05 μmol/mol creatine. The difference between the groups was statistically significant, with the highest value observed in the maintenance group (median 1.71 μmol/mol creatine; range 0.06 to 14.69), followed by the workers of battery C (1.15; 0.13 to 9.27), battery A (0.96; 0.01 to 31.04), battery B (0.90; 0.09 to 9.95), and cleaners (0.69; 0.08 to 3.36). It is noteworthy that 27% of the workers exceeded the proposed benchmark guideline value of 2.3 μmol/mol creatine. Maintenance workers showed the most elevated value of 1-HOP (4.42; 3.25 to 5.58), followed by machine operators in battery C (4.31; 1.29 to 5.27), and by topside workers in battery A (3.28; 0.14 to 7.06). No statistical difference was found in relation to smoking habits. Workers with the GSTT1 null genotype had higher 1-HOP levels than those with wild type (p = 0.06). In multiple logistic regression, the concentration of 1-HOP was associated with GSTT1 polymorphism, adjusting for smoking habits, department, and job title.

Conclusion: A slight influence of the genetic polymorphisms GSTT1 on the urinary concentration of 1-HOP was observed.


D. G. Ellingsen1, J. Efskind2, A. Hartmann3, Y. Thomassen1.1National Institute of Occupational Health, Oslo, Norway; 2Health Department, Borregaard Ind. Ltd., Sarpsborg, Norway; 3Section for Nephrology, Rikshospitalet, Oslo, Norway

Introduction: The urinary excretion of N-acetyl-β-D-glucosaminidase (U-NAG) is used as a biomarker in epidemiological studies assessing injury to the kidney proximal tubulus cells (PTC). The kidney is rich in selenium, and the PTC are supposed to be the main source of serum glutathione peroxidase (S-GSH-px). We have previously shown that selenium status modified the excretion of U-NAG in low mercury vapour exposed chloralkali workers. This study assesses the impact of selenium status on the excretion of U-NAG.

Methods: U-NAG, S-GSH-px, and serum selenium (S-Se) were measured in a cross sectional study of 49 chloralkali workers previously exposed to mercury vapour and 49 age matched referents.

Results: The mean age was similar in the groups (exposed group 46.4 years, range 28.9 to 66.9; referent group 46.4 years, range 28.2 to 68.8). The geometric mean U-NAG concentrations were 0.19 and 0.19 U/mmol creatine respectively (p = 0.99). Linear regression analysis demonstrated statistically significant associations between U-NAG and age. S-Se and S-GSH-px were also included in the statistical models. Based on the calculated regression equations, the material was stratified according to age into three groups, and the indices of selenium status were stratified into “high” and “low”. No significant group differences in U-NAG were found between subjects with “high” or “low” selenium status in the subjects of low or medium age. However, the oldest subjects (range 51.6 to 68.8 years) with “high” S-Se or S-GSH-px had significantly lower U-NAG than those with “low” S-Se or S-GSH-px.

Conclusions: The study shows that “high” S-Se or S-GSH-px is associated more with lower U-NAG excretion than “low” S-Se or S-GSH-px, particularly in the oldest subjects. The potential effect modification from selenium is hardly ever assessed, but should be considered when epidemiological nephrotoxic studies using U-NAG as a biomarker are performed. Knowledge of selenium status is important in order to compare studies, especially across countries, as selenium status may vary considerably.

View Abstract

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.