Search tips
Search criteria 


Logo of oenvmedOccupational and Environmental MedicineVisit this articleSubmit a manuscriptReceive email alertsContact usBMJ
Occup Environ Med. 2007 December; 64(12): e44.
PMCID: PMC2095399

Surveillance and health care

235 Hospitalisation due to occupational diseases in Manitoba, Canada

A. G. Kraut. University of Manitoba

ObjectivesOccupational diseases have been estimated to cause considerable morbidity and mortality in developed countries. Limited data are available on the frequency of occupational diseases in Canada. The purpose of this study is to determine what proportion of hospitalisations of men to the general medical wards from the emergency room in a large tertiary care hospital is related to occupational exposures.

MethodsA trained nurse took a standardised occupational history on a random selection of men who were hospitalised. This information was linked with demographic variables, admission diagnoses, co‐morbidities and outcome variables on the internal medicine database which is maintained on all admissions. Each participant's diagnoses and occupational history were reviewed by the principal investigator, a certified specialist in internal medicine and occupational medicine to determine if the admission was definitely, probably, or possibly related to occupational exposures. When required, the hospital chart was reviewed and/or the participant contacted by telephone to obtain additional information. The proportion of all the study participants thought to have an admission or co‐morbidity due to occupational exposures was then calculated.

ResultsOf 153 men entered into the study, none had a condition that was definitely related to their work. Eight individuals (5.2%) had a condition that was possibly or probably related to their work. These included five cases of cancer, two of heart disease and one case each of inflammatory bowel disease and COPD. One individual had two conditions. For four individuals this condition directly led to the admission and in the other four it was thought to be a significant co‐morbidity and indirect cause of the admission. In another 10 individuals an occupational cause for a medical condition was considered, but on review of the chart or interview with the patient the case was ultimately not felt to be work related.

ConclusionOccupational exposures probably or possibly lead to conditions causing or contributing to 5% of admissions to the general medical wards from the emergency room. Increased training in occupational medicine will likely lead to an increase in the number of hospitalisation where workplace factors are thought to play a role.

Key wordsoccupational diseases; hospitalisations; medical wards

236 What determines the size of occupational health services in UK universities?

K. M. Venables, S. Allender. University of Oxford

ObjectivesThere is no requirement on UK employers to provide or procure an occupational health service for their staff. Some employers view occupational health provision as an infrastructure cost rather than a benefit to their operations. This may contribute to the wide, largely unexamined, variation in occupational health provision between UK employers. This study aimed to explain the variation in occupational health provision across the UK university sector, making use of data from a recent national survey.

MethodsAnalyses of data from a survey of university occupational health services and data from the Higher Education Statistics Agency. The outcome variable was clinical (doctor+nurse) staffing of the university's occupational health service. The explanatory variables examined were: university size (expressed as number of staff employed), university income from all sources, research activity score derived from the periodic national audit of university research, and presence or absence of academic disciplines categorised by an expert panel as requiring a high level of occupational health provision (eg, medicine, dentistry, and nursing).

ResultsAll 117 UK universities were included and 93 (79%) responded; with exclusions and incomplete data, between 80 and 89 were included in analyses. There was wide variation in clinical occupational health staffing (range 0–8.4 full‐time equivalents). The median number of staff employed by the university was 1605 with a wide range. Number of staff explained 34% of the variation in occupational health staffing. After adjusting for other factors, neither the research activity score nor the presence of high needs disciplines appeared to be factors currently used by employers to determine their investment in occupational health.

ConclusionGovernment or other guidelines for employers should take organisational size into account. Employers may need guidance on how to provide a level of occupational health provision which is proportionate to specific occupational hazards or to other occupational health needs, such as patient safety, or employee rehabilitation.

Key wordsoccupational health provision; higher education sector; evaluation

237 Use of surveillance data to provide unbiased estimates of time‐trends in UK work‐related disease

R. McNamee1, M. Carder1, Y. Chen2, R. Agius1. 1University of Manchester; 2Alberta Cancer Board

ObjectivesEstimation of change in incidence may be possible from surveillance schemes with well‐defined reporter bases even if coverage is incomplete. The University of Manchester ODIN/THOR network collected data on work‐related ill‐health in the UK as diagnosed by specialist physicians. We used these data to estimate UK time trends during 1996–2005 in incidence of specialist diagnosed work‐related disease, particularly skin and respiratory diseases, and to investigate whether there was evidence of reporter behaviour that might cause bias.

MethodsReporters in two schemes (EPIDERM and SWORD) were specialist physicians for skin and respiratory diseases, respectively, while those in OPRA were occupational physicians. All were volunteers who reported new cases of disease which, in their opinion, were work‐related, either reporting monthly or for 1 month in every year. “Report cards” were to be returned even when no new cases were seen. Response – whether a card was returned – and number of cases were recorded each month. Probability of non‐response and, for returned cards, of a “zero” return, were modelled as a function of calendar time and/or membership time using two‐level logistic models. Annual change in disease incidence (specific diseases and all work‐related, all skin and all respiratory) was estimated using two‐level Poisson models controlling for reporter type, season and whether a first report and with, and without, control for membership time.

ResultsNon‐response was low overall (~15%) but increased over time. The probability of a zero return increased with membership time, independently of calendar time, in some but not all reporter groups. Annual change in incidence of all work‐related diseases reported through OPRA was estimated as 2.5% (95% CI 1.5 to 3.4) ignoring membership time compared to 7.9% (95% CI 3.3 to 12.6) after control for membership time. Corresponding figures for EPIDERM sample reporters were −2.0% (95% CI −4.5 to 0.5) and 3.6% (95% CI −1.0 to 8.4), respectively, and for SWORD (1999–2005 only) sample reporters −3.9% (95% CI −6.8% to −0.9) and −2.0% (95% CI −5.7% to 1.8%). Time trends for some specific diseases (eg, occupational asthma) varied between groups of reporters.

ConclusionGiven variation between reporter groups and according to model assumptions, time trends from surveillance data need to be interpreted with caution.

Key wordssurveillance; estimation of incidence; bias

238 The French National Mesothelioma Surveillance Program

A. Gilg SoitIlg1, E. Imbernon2, P. Rolland2, S. Ducamp2, A. De Quillacq3, C. Frenay4, S. Chammings5, G. Launoy6, J. C. Pairon5, P. Astoul4, F. Galateau‐Salle3, P. Brochard7, M. Goldberg2. 1Institut de Veille Sanitaire, Département Santé Travail; 2Institut de Veille Sanitaire, Département Santé Travail, Saint‐Maurice; 3Laboratoire d'Anatomie Pathologique CHU, Groupe MESOPATH, INSERM ERI3, Caen; 4Département des Maladies Respiratoires, Service d'Oncologie Thoracique, Hôpital Sainte‐Marguerite, UPRES 3287, Université de la Méditerranée, Faculté de Médecine de la Timone, Marseille; 5Institut Interuniversitaire de Médecine du Travail de Paris, Ile de France, Paris, INSERM E03‐37, Créteil; 6Réseau des Registres du Cancer FRANCIM, INSERM ERI3, Caen; 7Laboratoire Santé Travail Environnement, Institut de Santé Publique, d'Epidémiologie et de Développement, Bordeaux

ObjectivesThe French National Mesothelioma Surveillance Program (PNSM) was established in 1998 by the National Institute for Health Surveillance (InVS). Its objectives are to estimate trends in mesothelioma incidence and the proportion attributable to occupational asbestos exposure, to help improve its pathology diagnosis, to assess compensation for it as an occupational disease, and to contribute to research.

MethodsThe PNSM records incident pleural tumours in 21 French districts that cover a population of approximately 16 million people. A standardised procedure of pathologic and clinical diagnosis ascertainment is used. Lifetime exposure to asbestos and to other factors (man‐made mineral fibres, ionising radiation, SV40 virus) is reconstructed, and a case‐control study was also conducted. We assessed the proportion of mesothelioma compensated as an occupational disease.

ResultsDuring the 1998–2001 period, we estimated the annual incidence to be about 600 cases among men (incidence rate: 2.2 per 100 000), and between 150 and 200 for women (incidence rate: from 0.5 to 0.75 per 100 000); mean age was 69 for women and 70 for men. Pathology review confirmed the initial pathologist's diagnosis in 67% of cases, ruled it out in 13%, and left it uncertain in the others; for half of the latter, the clinical findings strongly supported a mesothelioma diagnosis. Analysis of the histological variants showed mostly epithelioid mesothelioma (approximately 70%). The mixed form was observed in 15% of the cases and the sarcomatoid form in 11%; the desmoplastic form accounts for less than 2% of the latter. In all, 62% applied for designation of an occupational disease, and 91% of these were receiving workers' compensation; of the 38% of subjects who did not make this request, about half were considered asbestos‐exposed.

ConclusionThe PNSM is a large‐scale epidemiological surveillance system with several original aspects, providing important information to improve knowledge of malignant pleural mesothelioma, such as by monitoring the evolution of its incidence, of high risk occupations and economic sectors, and improve pathology techniques.

Key wordsmesothelioma; epidemiological surveillance; occupational disease compensation

239 Occupation and cause of death: the Cosmop surveillance program

B. Geoffroy‐Perez, S. Julliard, A. Fouquet, M. Goldberg, E. Imbernon. Institut de Veille Sanitaire

ObjectivesThe Cosmop program is a national surveillance program on mortality and causes of death by occupational groups. It aims at providing indicators contributing to guide and evaluate national health policy as regards occupational risks.

MethodsThis program relies on existing permanent sources of pertinent data collected at the French population level. The present analyses were based on the “Échantillon Démographique Permanent” (EDP), a 1% representative sample of the French population, set up at the 1968 census by the Department of Demography of the National Institute of Statistics. For each individual, civil life events, census occupational data and vital status were prospectively collected. The causes of 95% of all deaths observed were obtained by direct linkage with the French National Death Registry. For different causes of death, relative risks were calculated for each economic sector by reference to the others, taking into account the distribution of social categories in the sector.

ResultsOverall and cause specific mortality among the 137 860 men (35 968 deaths) and 105 290 women (13 797 deaths) active during the study period (1968–1999) and selected for the analysis was described. An excess of death was observed in men in the sectors of production and industries, while the agricultural sector, commerce and services showed lower mortality risk. An excess of death from malignant diseases was observed among construction (RR 1.26), metallurgy (RR 1.27), coal production (RR 1.35), meat industry (RR 1.42), other agri‐food industry (RR 1.27) and repair (RR 1.30) workers.

ConclusionDespite several limitations of the EDP sample, particularly the lack of specificity of occupational factors studied, the first analyses provided, for the first time in France, mortality indicators by cause of death and economic activity. The advantage and the perspectives of such a program for the surveillance of working population will be given.

Key wordscauses of death; occupation; surveillance program

240 Dynamic blood pressure changes in female nurses under different work shifts and recovery

S. H. Lo1,2, J. D. Wang2,3, C. S. Liau3. 1Department of Cardiology of Internal Medicine, Taipei City Hospital, Zhongxing Branch; 2Institute of Occupational Medicine and Industrial Hygiene, College of Public Health, National Taiwan University; 3Department of Internal Medicine, National Taiwan University Hospital, Taipei, Taiwan

ObjectivesTo evaluate the impact of dynamic blood pressure changes in nurses of self‐repeated measurement in three different work shifts and recovery patterns among them; also to explore the impact on non‐dipper and dipper status on different shifts and off‐duty days.

MethodsSixteen young female nurses working in rotating shifts were recruited for repeated ambulatory blood pressure monitoring (ABPM) for three different shifts and the next off‐duty days to explore the dynamic changes of blood pressure (BP) from shift work and their recovery patterns. Six nurses on day shift only were also observed for comparison.

ResultsStatistical analysis with mixed model construction revealed significant decreases in systolic BP and diastolic BP (−13.0 (SD 1.1) and −8.9 (0.8) mm Hg, respectively) during sleeping time, and increases during duty (6.6 (0.9) and 4.7 (0.7) mm Hg, respectively), on a work day (5 (1.0) and 4.4 (0.8) mm Hg, respectively) and during sleeping time after night shift (4.7 (1.4) and 4.3 (1.0) mm Hg, respectively) or evening shift (4.1 (1.4) and 4.0 (1.0) mm Hg, respectively). These changes on evening shift returned almost to normal after an off‐duty day, while the recovery from night shift seemed less satisfactory (p<0.005). About 19% of study subjects showed no change in non‐dipper or dipper status in the three different shifts, while about 40–50% of nurses showed changes in such states within the same person between a work day and an off‐duty day after the three different shifts.

ConclusionWe concluded that after control of potential confounding by repeated measurements within the same person, evening/night shift working as a nurse was associated with elevation of SBP and DBP. Such impacts usually recover after a full off‐duty day, but are less complete after night shift. Shift work and the workload of nursing tasks may produce a change in non‐dipper and dipper states.

Key wordsshift work; recovery; blood pressure monitoring, ambulatory

241 Predictors of workplace absenteeism in cancer care workers

A. Carosi, N. Lightfoot, K. Alkema. Regional Cancer Program, Sudbury Regional Hospital

ObjectivesThe aim of this study was to identify predictors of workplace absenteeism in cancer care workers.

MethodsThis study included 244 cancer care workers employed by Cancer Care Ontario at the Northeastern Ontario Regional Cancer Centre, Sudbury, Ontario, between 1 January 1998 and 31 December 2003. Sickness absence data were obtained from employee attendance records and human resource databases. Logistic regression analyses were used to estimate the risk of high sickness absence (more than three absence events per year), high sickness absence duration (more than 5 days of absence per year), and high mean duration sickness absence (more than 1.5 days per absence event).

ResultsThe highest rates of sickness absence were observed during the winter months (January–March) and lowest during the summer months (July–September) (p = 0.001). Younger employees, less than 40 years of age (odds ratio (OR) 2.15, 95% CI 1.20 to 3.83), permanent employment (OR 3.34, 95% CI 1.68 to 6.66), and low job level (OR 2.87, 95% CI 1.38 to 5.98) were associated with a significantly increased risk of high sickness absence. Male gender was associated with a significantly lower risk of sickness absence (OR 0.34, 95% CI 0.16 to 0.72). Permanent employment (OR 3.37, 95% CI 0.13 to 0.62) and low job level (OR 2.95, 95% CI 1.40 to 6.21) were associated with a significantly increased risk of sickness absence duration. Males had a significantly lower risk of sickness absence duration (OR 0.28, 95% CI 0.13 to 0.62). Permanent employment (OR 2.69, 95% CI 1.29 to 5.60) was associated with a significantly increased risk of high mean duration sickness absence. Clinical workers (OR 8.23, 95% CI 1.77 to 38.20) and non‐clinical workers (OR 5.54, 95% CI 1.14 to 26.85) displayed a significantly increased risk of high mean duration sickness absence compared to administration workers.

ConclusionOur findings indicate that socio‐demographic and work‐related factors, and seasonal variation had significant effects on workplace absenteeism in cancer care workers and highlight the need for interventions and sickness absence management policies to address these concerns in cancer care settings.

Key wordsworkplace absenteeism; sickness absence; healthcare workers

242 Occupational exposures of Australian nurses

T. R. Driscoll1, A. Hogan2, G. Kearney3. 1School of Public Health, University of Sydney; 2Australian Safety and Compensation Council; 3Australian Nurses Federation

ObjectivesChemical, biological, psychosocial or other hazards in the course of their work. The project was a quantitative survey of Australian nurses intended to identify the key hazards to which they are exposed as a result of their work and to pilot the use of an Australian adaptation of the NIOSH Employee Core Module used as part of the National Exposure at Work Survey.

MethodsParticipants were a stratified random selection of members of the Australian Nurses' Federation (the federation covers approximately 55% of Australian registered nurses) who were invited via email to participate in the study by completing an on‐line questionnaire. The questionnaire was based on the NIOSH Employee Core Module, adapted for the Australian workforce. The exact number of nurses who received the invitation was not clear, because many of the email addresses were no longer active, but the number who received an invitation was probably about 6000.

ResultsNine hundred and fifty‐five subjects participated (85% female). Exposures from which the respondents most commonly reported they considered themselves at moderate or high risk were workplace stress (50%), lifting/repositioning heavy objects (44%), needlesticks and other sharps (43%), prolonged standing (43%) and blood‐borne pathogens (40%). Fifty‐one per cent of respondents had ever sustained a workplace injury or disease (while nursing) that had required time off work (14% had sustained such an injury in the previous 12 months). The most common reported disorders (respondents could report more than one) were musculoskeletal disease/injury (71% of respondents who had taken time off work), stress (20%), bullying (15%) and infection (12%). The questionnaire, which was extensive, was completed within 30 min by 77% of respondents. Only 5% of respondents reported that it took more than 45 min to complete.

ConclusionWorkplace stress, musculoskeletal stressors and pathogens are the exposures of most concern to Australian nurses. Use of a web‐based questionnaire appears to be a feasible method of collecting self‐report information on workplace exposures.

Key wordsnursing; exposure; surveillance

Articles from Occupational and Environmental Medicine are provided here courtesy of BMJ Group