Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please "contact us" to request a format other than those available.
In the years 2002 to 2004, acute injuries occurring on the job resulted in an average of 465 deaths annually, and close to 300,000 compensated time-loss claims.1 The consequences of occupational injuries can be appreciable: lost work time and income, medical expenses, compensation costs, possible long-term health problems or disability, and a burden on the family of the injured worker.
As with all injuries, a substantial share of those that occur on the job can be prevented. An improved understanding of the circumstances associated with occupational injuries should contribute to more effective preventive strategies.
To date, much of the research on work-related injury in Canada has focused on specific occupational categories — such as farmers — and usually in specific geographic regions. A search of the PubMed data base2 for Canadian papers on occupational injury published from 1990 to January 2007 yielded 33 descriptive or analytic studies, 14 of which concerned the agriculture sector; only 6 were based on data for all of Canada.
Most statistics on occupational injury in Canada are collected by administrative agencies involved in injury compensation, and are thus limited in coverage and the information they provide. For example, self-employed people and some professionals may not be included, and data on socio-economic characteristics and other health-related risk factors are not collected. As well, only compensated injuries are documented, although fewer than half of workers who sustain an injury file a claim.4,5 Thus, injury statistics from compensation boards would not be expected to correspond with estimates from survey data.
The availability of data from Statistics Canada's Canadian Community Health Survey (CCHS) offers several advantages in the study of occupational injury. The dataset is large, permitting more analytical precision than is usually possible. The data are population-based and provide information on a broad range of social and personal factors, as well as work-related variables and injury occurrence, and they do not rely on workplace-based reporting.10,11 An additional advantage of the CCHS is that within the questionnaire, items on work-related injury are separated from those on work conditions, thus reducing the bias that may arise in studies based on more focused questionnaires.
A typical theoretical model of occupational injury suggests that risk arises from an interplay of tangible conditions in the work environment or tasks directly related to the job (for example, exposure to hazardous materials or equipment), the organization of the work (such as working overtime or shifts), and individual or behavioural characteristics, including socio-demographic and psychological characteristics and chronic disease morbidity.13-16 To the extent that variables reflecting these factors were available, the models described by Schuster and Rhodes13 and Veazie et al.14 serve as the basis for this study.
Using data from the 2003 CCHS (cycle 2.1), this article provides estimates of the number of employed Canadians aged 18 to 75 who sustained at least one non-fatal, activity-limiting injury on the job in 2003 (see Methods and Definitions). The analysis concerns only acute injuries; repetitive strain injuries are not included. The specific objectives were to compare injury occurrence by occupational category, and to examine relationships between on-the-job injuries and selected work-related and personal factors. Results are presented first for unadjusted, weighted estimates, and then for adjusted (multivariate) models.
Substantial share of injuries work-related
Blue-collar workers at higher risk
Cooks, machinists injured more
Hand, lower back injuries most common
Strenuous jobs risky
Long hours linked to injury
Young men at higher risk
Smokers more often injured
Work-related influences persist
Link to job stress persists in women
In 2003, an estimated 630,000 Canadians experienced at least one activity-limiting occupational injury, representing 5% and 2% of employed men and women, respectively (Table 1). Because the estimates pertain only to the most serious injury, and also because of respondents' memory decay (see Limitations), these figures underestimate the actual frequency and proportions of work-related injury.
Occupational injuries comprised a substantial proportion of all injuries. More than a quarter (28%) of employed people aged 18 to 75 who reported an activity-limiting injury in 2003 (one-third of men and one-fifth of women) sustained their most serious injury at work.
Despite methodological differences, these results were strikingly similar to the estimated share of medically treated injuries reported to be work-related in the United States over the 1997-to-1999 period (29%), based on data from the National Health Interview Survey.10
Not surprisingly, injury was more common in "blue-collar" than "white-collar" jobs (see Definitionsand Geographic differences). Close to one worker in ten (9%) in trades, transport and equipment operation sustained an on-the-job injury, more than four times the rate (2%) for those employed in business, finance or administration, or in social science, education, government service or religion (Table 2). People employed in processing or manufacturing, or in primary industries, were also at higher risk of work injury, relative to the total workforce.
Men in trades, transport and equipment operation; processing or manufacturing; and primary industries were significantly more likely to be injured, compared with male workers overall. For women, the likelihood of injury was significantly elevated in the same occupational categories, as well as in sales and service.
In both the white- and blue-collar sectors, men's injury rate significantly exceeded women's. Men in management; business, finance and administration; trades, transport and equipment operation; and processing and manufacturing were about twice as likely as their female counterparts to have been injured.
In white-collar occupations, health care workers were more likely to be injured (3%), compared with all white-collar workers combined (2% - data not shown). Those in business, finance or administration had a significantly lower likelihood of injury.
For blue-collar workers, the higher frequency of injury (relative to white-collar workers), together with the large sample size of the CCHS, permitted comparisons of injury occurrence between occupations within each category.
In the sales and service group, 7% of those employed as chefs or cooks were injured, significantly higher than the proportion (3%) for all workers in that group (Table 3). Also at higher risk were workers in food and beverage services, and those in protective services.
Among workers in trades, transport and equipment operation, a strikingly high share (13%) of machinists, metal formers, shapers and erectors were injured on the job. The risk was also significantly higher for mechanics, compared with all workers in the category.
Agriculture workers (a group that includes contractors, operators and supervisors, but not labourers) accounted for 55% of workers in primary industries (data not shown), so it was not surprising that the proportion who were injured did not differ significantly from the proportion for the category overall.
Within processing, manufacturing and utilities, machine operators were injured significantly more frequently, compared with all people in this category (Table 3).
Two mechanisms — overexertion/strenuous movement and falls — accounted for nearly half (49%) of occupational injuries reported to the CCHS (Table 4). These findings corresponded closely with those from the United States. According to data from the National Health Interview Survey of 1997-1999, 48% of occupational injuries were caused by falls and overexertion or strenuous movement.10
CCHS results indicated that another 32% of work-related injuries were due to accidental contact with a sharp object, tool or machine, or being accidentally struck or crushed by an object.
The mechanisms underlying work injuries differed from those of non-work injuries. Work injuries were more than twice as likely as those occurring elsewhere to have involved accidental contact with a sharp object, tool or machine or being struck or crushed by (an) object(s). Accidental contact with a hot object, liquid or gas was four times as likely to have resulted in a work injury, compared with a non-work injury.
The hand was the body part most frequently injured on the job (Table 5). Over one-quarter (28%) of work injuries were to the hand, followed by the lower back (16%). Compared with all workers who sustained an occupational injury, those in sales and service were more likely, and white-collar workers (combined) less likely, to injure a hand. White-collar workers were more likely to injure the lower back: one-fifth of the occupational injuries sustained by white-collar workers affected this part of the body.
The most frequent type of occupational injury was a sprain or strain, followed by cuts, then fractures (Table 6). Sprains or strains were more common among white-collar workers, compared with workers overall. Fractures accounted for nearly one-fifth (19%) of injuries to primary industry workers, significantly above the share (11%) for all workers. Burns or scalds made up 15% of injuries sustained by sales and service workers — significantly higher than the proportion of all work injuries (6%).
To assess the severity of occupational injuries reported to the CCHS, the proportion of people injured at work who sought treatment was compared with the corresponding proportion for people injured in other circumstances. Two-thirds (66%) of people injured at work sought treatment, significantly higher than the proportions for people injured during sports (52%) or while performing household chores or other unpaid work (60%) (Table 7). Although this suggests that occupational injuries were more severe, the requirement that injuries be medically certified for compensation benefits might have influenced treatment-seeking behaviour.
To further examine the gravity of work injuries, those that are usually more harmful (burn or scald, concussion, fracture, internal injury, multiple injuries) were grouped together in a "serious" category. Fully 15% of the people who sustained such injuries were hospitalized, compared with 1% of those who had other types of injuries (data not shown), a difference that validates the designation of these injuries as serious.
Based on this definition, 20% of work injuries were "serious," significantly below the corresponding proportions for injuries sustained during other activities (Table 7). For example, 28% of people injured while travelling to or from work had a serious injury, as did 23% of those injured while engaging in sports or exercise. Clearly, these findings do not support the suggestion that occupational injuries are relatively more serious than those sustained elsewhere.
However, the proportion of work injuries that were serious varied by occupational category — 27% of injuries sustained by sales and service workers were serious, more than twice the share among white-collar workers (13%) (data not shown). Serious injuries were also significantly more common among workers in trades and transport, primary industries, and processing and manufacturing.
Most work injuries for which treatment was sought were treated in emergency rooms (51%), followed by doctors' offices (Table 8). Fewer than one in twenty non-fatal work injuries required hospitalization.
Although CCHS respondents were not asked about their specific job tasks or the materials and tools they used, they did provide information about the physical effort their daily activities involved. As reported in previous research, the likelihood of occupational injury was markedly elevated for workers whose jobs required strenuous effort, compared with those without such demands.11,15,17-19 The risk of injury for people doing heavy work was twice as high in men, and three times as high for women, compared with those with less physically demanding jobs (Table 9).
In addition to occupational category and the effort involved in a job, several organizational aspects of employment emerged as significant correlates of work injury.
For men, the number of hours worked per week was associated with injury (Table 9). The likelihood of injury was greater for men who worked 35 or more hours, compared with those who typically worked fewer hours. Furthermore, the results suggested a gradient in risk corresponding to hours of work. This was consistent with previous research, showing that jobs routinely requiring overtime increase the risk of occupational injury.20 No significant differences by hours of work emerged for women.
Both men and women with a regular daytime schedule were at significantly lower risk of injury, compared with those who worked other shifts. These results corroborate previous research indicating that shiftwork is associated with a higher risk of occupational injury.21,22 As well, men and women who were self-employed were less likely to sustain an occupational injury, compared with those who worked for others.
The amount that workers are paid is another aspect of the organization of work. Among men with annual earnings of $60,000 or more, 3% were injured on the job, significantly below the proportions for men with income less than $60,000 (Table 9). Women whose annual income was less than $40,000 had a significantly higher risk of work injury, compared with women earning $40,000 or more. Of course, analysis of injury in relation to income is far more informative when occupation is taken into account; this was undertaken in multivariate analysis (see below).
For both sexes, having more than one job was significantly associated with an elevated likelihood of work injury, compared with workers who had only one job.
The results of previous research on the relationship between stress and work injury are inconclusive, perhaps because of the variety of measures that have been used to assess stress as well as injury.23 A significant association has emerged in some studies,24-26 but not in others.27 Analysis of the CCHS data indicated that women's injury risk differed significantly in relation to the perceived stressfulness of their job, but no pattern emerged for men. Just over 4% of women who described their job as "extremely" stressful had sustained a work injury, compared with less than 2% of those who said their job was "not a bit" or "not very" stressful.
Factors such as a disposition to risk-taking have been shown to influence the likelihood of work injury.28 Although the CCHS contained no direct measures of such factors, it did collect data on personal and socio-demographic characteristics that are relevant to the study of occupational injury, including sex, age, race, educational attainment, smoking, alcohol use, body mass index, diagnosed chronic conditions and life stress.
For men, the proportion injured generally decreased with age, which could, in some cases, be related to experience on the job. Compared with male workers aged 18 to 24, those aged 25 to 34 were significantly more likely to be injured at work, and those aged 45 or older were less so (Table 10). To some extent, the low proportion of men aged 18 to 24 who were injured may have reflected a greater likelihood of part-time work among this group, and thus, less exposure to risk. In fact, when hours of work were taken into account, the percentage of men aged 18 to 24 who were injured was not significantly different from that for those aged 25 to 34 (data not shown). Among women, the likelihood of work injury was similar at all ages. The findings for men support those of previous studies showing a higher risk of occupational injury at younger ages than in middle or older adulthood.29-31
Other socio-demographic characteristics — race and level of household education — were also related to the risk of work injury. Non-white men had a lower risk of occupational injury than did white men, a finding consistent with reports from Canada and the United States.28,40 Workers living in households in which at least one member had been educated beyond secondary graduation were also at lower risk of injury, compared with those from households in which no one had gone beyond secondary completion. The findings related to education are consistent with those of a study conducted in Israel, showing that workers with less than 12 years of education had elevated odds of injury, compared with those with more years of schooling.19
Occupational injury was also linked to health-related risk factors. Workers with at least three chronic conditions were more likely to have had a work injury, compared with those reporting two or fewer conditions. Specific conditions that were significantly associated with work injury included migraine, arthritis, stomach ulcer and multiple chemical sensitivities (data not shown).
Consistent with observations reported elsewhere, men and women who smoked daily had a higher likelihood of injury, compared with occasional or non-smokers (Table 10).24,42-44 Although the mechanism linking smoking to injury risk is not fully understood, previous research has identified smoking as a precursor to injury.45
Heavy alcohol consumption was also related to the likelihood of injury. Of men who reported heavy episodic drinking (that is, over the past year they had consumed at least five drinks per occasion, at least once a week), 6.2% were injured at work, significantly higher than the corresponding percentage (4.8%) for those who had consumed one or more drink(s) in the past year, but whose heavy drinking (if any) had occurred less than once per month. For women, 3.5% of weekly heavy drinkers were injured, compared with 2.1% of those whose heavy drinking occurred less than monthly; because of smaller numbers, this difference fell just short of statistical significance (p=0.06). In a number of previous studies, a positive association has been observed between heavy drinking and occupational injury,17,18,28,46-48 although in other research, this relationship failed to emerge.38
Obese women were twice as likely to be injured at work as were those whose weight was in the normal range: 4% compared with 2%. The finding for women is consistent with previous research indicating a positive relationship between obesity and risk of occupational injury.19,49 No significant difference emerged for men.
Before the influence of other factors (including work stress) was taken into account, personal life stress was also associated with work injury. Among men and women who reported that on most days their lives were "extremely" stressful, the likelihood of injury was significantly higher than among those reporting lives that were "not at all," or "not very"stressful. This observation corroborates findings of previous studies on the role of stress, although the variety of measures used makes comparability problematic.23
To examine the relationship between work-related variables and injury, while controlling for other influences, successive multivariate models were fitted for each sex — the first containing work-related variables, and the second containing socio-demographic and other health-related risk factors as well (Table 11).
In each of the second, fully controlled models, the relationships observed in the first models between work-related factors and occupational injury generally persisted. For men, regardless of age, race, household education, health-related risk factors and level of job stress, those employed in trades, transport or equipment operation; primary industries; and processing, manufacturing or utilities had over twice the odds of work injury as did those in white-collar occupations (Table 11). Aside from occupational category, physical exertion on the job was also significant for men; the odds of injury associated with heavy lifting or strenuous activity were 70% higher than the odds associated wtih less physically taxing jobs.
Men whose annual income was below $60,000 had higher odds of injury, compared with those at $60,000 or above. The relationship between injury and long hours also persisted; compared with men who worked less than 35 hours per week, the odds of injury were 40% higher for those who worked 45 to 79 hours per week, and nearly twice as high for those who worked 80 hours or more. Male shift workers also had higher odds of on-the-job injury.
Some of the findings for women were similar to those for men (Table 11). As well as in the three occupational categories in which men's odds of injury were elevated, women in sales and services had high odds of injury. Physical exertion on the job also conferred higher odds of injury for women, as did shiftwork and having more than one job. Being self-employed was protective. However, for women, no significant relationships emerged between injury and level of employment income or number of hours worked.
In contrast to the findings for men, women's odds of injury rose in relation to work stress. Compared with women who felt that their job was "not at all" or "not very" stressful, those perceiving more job stress had increasingly higher odds of injury (Table 11). Women in jobs they reported as "extremely" stressful had nearly three times the odds of work injury as did those in jobs identified as not stressful, even when personal life stress was taken into account. This indicates that the association between work stress and injury risk was unique, over and above any influence from "life" stress in general. Of course, this observation is based on cross-sectional data and may reflect reverse causation: the experience and consequences of having sustained an injury on the job may lead to increased stress.
Independent of work-related factors, some health-related risk factors were significantly related to occupational injury. Those that are modifiable are worth noting. Both men and women who smoked daily had significantly elevated odds of on-the-job injury. For women, obesity was also independently significant; obese women had nearly twice the odds of injury as did those whose weight was in the normal range. The findings of previous research about smoking and obesity in relation to occupational injury are contradictory. Some studies have suggested that individual factors including weight and smoking were not significantly associated with occupational injury when work-related factors were considered,38,50,51 while others are more consistent with results from the CCHS.44,49
Well over half a million Canadian workers were injured on the job in 2003. The majority (72%) of them were men, and nearly three-quarters were employed in blue-collar occupations. These findings serve as a reminder that despite recent improvements,52 large numbers of workers are still being injured, and the risk is unevenly distributed within the workforce.
A better understanding of the circumstances that give rise to injury should contribute to the success of injury prevention efforts. Analysis of the CCHS data revealed a number of factors that were strongly associated with occupational injury. Some of these factors were intrinsic to the job, while others reflected personal or socio-economic conditions.
A number of factors associated with work injury were similar between the sexes, including shiftwork, physically demanding jobs, chronic health conditions, and smoking. Other factors differed between men and women. Women with more than one job, and those in sales and service, were at higher risk of injury relative to female white-collar workers, but these relationships were not significant in men. Income and long working hours were associated with injury in men, but not in women. Obese women were at higher risk of injury, but this relationship did not emerge in men. Finally, perceived work stress was highly related to occupational injury in women, but not at all in men.
Findings from the CCHS help to identify individual characteristics and work-related conditions that are linked to an increased risk of work injury, and thus to suggest areas where injury prevention strategies might be directed. As well as programs to promote smoking cessation, healthy weight and stress reduction, workplace-based safety programs could be emphasized for workers in occupations at higher risk. The findings also underscore the importance to injury risk of factors that are modifiable by employers, including workplace hazards, equipment design, work schedules and workload distribution.