O24.1 IMPAIRMENT OF HEARING AND VISION AND THE RISK OF ACCIDENTAL INJURY: A SYSTEMATIC LITERATURE REVIEW

K. T. Palmer, E. C. Harris, D. Coggon.MRC Environmental Epidemiology Unit, University of Southampton, Southampton SO16 6YD, UK

Introduction: A systematic literature review was performed to estimate the risks of accidental injury arising from impairments of hearing and vision. The findings from occupational studies were compared with those on non-vocational accidents (mainly road traffic accidents (RTAs).

Methods: Systematic searches were conducted of the MedLine (1966–Oct 2002) and EmBase (1980–Oct 2002) databases using MESH headings and key words for hearing and visual impairment, and accidents (workplace, occupational, work related; traffic, motor vehicle, car, driving). Abstracts were read by two independent observers and primary research reports with control data were retrieved and summarised.

Results: 13 studies on hearing (6 occupational, 7 on RTAs) and 20 on visual problems (4 occupational, 16 others) were retrieved, covering each of the main study designs. Half reported >1 positive association (risk ratio (RR) >1.5, p<0.05), but most risk estimates were moderate (<2). Studies of RTAs and vision entailed many measurements of visual performance (visual acuity, field of vision (FOV), binocularity, contrast sensitivity) and covered several specific eye diseases (cataracts, glaucoma, macular degeneration). Only reduced FOV was consistently associated with accident risk (RR estimates ranging from 1.9 to 22.0 for a >40% reduced FOV). There was less evidence that risks were increased by hearing impairment, and none for the only two studies that directly measured hearing loss. Of the occupational studies identified across both impairments, only one attempted any direct measure of functional loss, and none distinguished risks for different categories of accidental event.

Discussion: Different impairments may give rise to different kinds of accident, especially in the workplace where work tasks and circumstances vary substantially. The data on RTAs (which represent one kind of accident) confirm a differential association, but in the occupational setting the relationship between category of accident and category of sensory impairment has not been well assessed. From the viewpoint of accident risk, the evidence base on which to decide fitness for work in those with sensory difficulties is surprisingly limited.

O24.2 INCIDENCE OF NON-FATAL INJURIES IN A COHORT OF COASTAL FISHERMEN

D. Loomis^1,2, S. W. Marshall^1,2, K. L. Kucera¹, M. A. McDonald³, H. Lipscomb³.¹Department of Epidemiology, University of North Carolina, Chapel Hill, NC, USA; ²Injury Prevention Research Center, University of North Carolina, Chapel Hill, NC, USA; ³Division of Occupational & Environmental Medicine, Duke University Medical Center, Durham, NC, USA

Introduction: Fishing is known as a dangerous occupation with a high rate of fatalities. However, research has focused primarily on fatalities in deep sea fleets operating in cold water. We conducted a prospective cohort study to determine the incidence and risk factors of occupational injuries among fishermen working in coastal waters in the southeastern United States.

Methods: A cohort of 219 fishermen was followed from August 1999 to May 2002. Demographic information and descriptions of work activities were obtained at baseline, and telephone interviews were conducted at regular intervals to ascertain incident traumatic injuries and assess exposure to potential risk factors. Person days at risk were estimated, and injury incidence rates and 95% confidence intervals (CIs) were calculated, with Poisson regression used to estimate rate ratios (RRs) for the contributions of specific factors.

Results: Participants contributed 46 093 person days at risk, and 81 fishermen reported a total of 142 incident injuries, a rate of 3.08 per 1000 person days (95% CI 0.26 to 3.59). Most injured fishermen (58%) reported only one injury, but 16% had three or more injuries. The hand and fingers were most often injured (57%), followed by the back and arms (each 9%). Penetrating wounds (40%), then strains/sprains (18%) were the most common injuries. The rates of injury while doing fishing work on the water and on shore were similar (1.8 per 1000 person days). For injuries on the water, higher rates were associated with fishing in the open ocean (RR 1.37; 0.76 to 2.47), and finfishing with hook and line (2.69; 1.16 to 6.23) and pound nets (2.68; 1.20 to 9.04). Low rates (0.43; 0.26 to 0.73) were associated with crab potting.

Conclusions: Non-fatal injuries are common among commercial fishermen, with rates varying by type of gear and location. Further research is needed to develop interventions to reduce injury rates.

O24.3 A CASE CROSSOVER STUDY OF TRIGGERS FOR HAND INJURIES IN COMMERCIAL FISHING

K. L. Kucera¹, D. Loomis^1,2, S. W. Marshall^1,2.¹Department of Epidemiology, University of North Carolina, Chapel Hill, NC, USA; ²Injury Prevention Research Center, University of North Carolina, Chapel Hill, NC, USA

Introduction: Previous studies have estimated that hand and wrist injuries are common in commercial fishing. Risk factors influencing these injuries have been identified from reports of previous activities and injury contacts such as working with catch; handling pots, nets, and lines; and slips or falls. No studies have attempted to look at the influence of transient risk factors, or triggers, of commercial fishing hand injuries. The case crossover design is an epidemiological study design for identifying transient risk factors of health outcomes, which is well suited for this purpose.

Methods: A case crossover study was nested within a previously established prospective cohort of 217 southeastern United States commercial fishermen assembled in April 1999 and followed to October 2001. Odds ratios (OR) and 95% confidence intervals (CI) were calculated to determine if transient risk factors such as use of personal protective equipment, engaging in more than one type of fishing, maintenance activities, and other covariates of interest increased the risk of occupational traumatic hand/wrist/digit injuries while commercial fishing. Hand injury cases were matched by worker (self matched) to five randomly selected control periods.

Results: In the study, 21% (46/217) of fishermen reported a hand injury during follow up for a total of 65 eligible hand injury cases. Fishermen performing maintenance work (any versus none) during the week of the injury were at increased risk of hand injury (OR 2.74, 95% CI 1.59 to 4.73). Glove use while fishing (any versus none) during the injury week appeared to be protective for hand injury (0.71; 0.40 to 1.26). Using multiple types compared with only one type of fishing equipment was associated with increased risk of hand injury (1.24; 0.85 to 1.80).

Conclusions: The case crossover design is a useful way of determining triggers of commercial fishing related hand injuries. Maintenance work was strongly associated with hand injury for these fishermen.

O24.4 REGIONAL DIFFERENCES IN THE INCIDENCE OF WORK INJURY AND NON-WORK INJURY AMONG CANADIAN ADOLESCENTS

C. Breslin^1,2, P. Smith¹, C. Mustard^1,2, J. Etches¹.¹ Institute for Work & Health, Toronto, Canada; ² Department of Public Health Sciences, University of Toronto Faculty of Medicine, Toronto, Canada

Introduction: Unintentional injury is a leading cause of morbidity and mortality among young people in the developed economies. The adolescent and young adult period may be a period of elevated risk for occupational injury.

Objective: To estimate differences in the incidence of work and non-work injury among adolescents and young adults across geographic regions in Canada.

Methods: The sample for this study were respondents to the 2000 Canada Community Health Survey (CCHS), aged 15–24 years who were employed (n  = 14 565) or not employed (n = 18 071) at the time of the survey. Measures of unintentional injury were based on respondent report of an injury requiring medical attention in the previous 12 months and were categorised as work related and non-work related causation. The health interview survey also provided information on hours of work, occupation, age, gender, province of residence, and socioeconomic status measures.

Results: Logistic regression was used to estimate provincial differences in the risk of unintentional work injury and unintentional non-work injury, controlling for relevant covariates. For young men, but not young women, hours of work were positively associated with the likelihood of an unintentional injury. We found significant provincial differences in injury risk. Respondents resident in the western provinces of Alberta and Saskatchewan had a higher risk of work and non-work injury. These geographic risks were not explained by occupation, socioeconomic status, age, or urban/rural distributions.

Conclusions: Risk factors measured at the level of the individual did not account for the substantial regional variation in work and non-work injuries for adolescents and young adults. These findings raise concerns about the consistency of occupational health and safety regulation across provincial jurisdictions in Canada.