2013 Volume 55 Issue 2 Pages 98-107
Objective: The aim of this study was to examine the relationship between age and the lost-time workers' compensation claims in British Columbia, Canada, over three time periods (1997–98, 2001–02 and 2005–06). We examined if the relationship between age and risk of lost-time claims is consistent over time and for different nature of injury categories. Methods: Secondary analyses of lost-time workers' compensation claims combined with estimates of person-years of exposure generated from the Canadian Labour Force Survey were performed. Analyses examined the relationship between age and claim risk using sex-stratified regression models, adjusting for time period, occupational characteristics and whether the claimant was employed in the goods or service industry. Multiplicative interaction terms were used to examine if the relationship between age and lost-time claim risk changed over time. Seven separate regression models were generated to explore the variation in the effect of age across nature of injury groups. Results: We observed important differences in the relationship between age and risk of injury depending on the nature of injury examined. A negative relationship was observed between age and lost-time claims for open wounds, while a positive relationship was observed for traumatic injuries to bones, nerves and the spinal cord. We found no evidence that the relationship between age and risk of lost-time claims changed over time periods. Conclusions: The association between age and risk of lost-time claims depends on the nature of injury under investigation. We found no evidence that the relationship between age and overall lost-time claim risk has changed over time in British Columbia.
(J Occup Health 2013; 55: 98–107)
Work-related injuries and illnesses are an important public health concern. The Association of Workers' Compensation Boards of Canada estimates that health care and wage replacement benefit expenditures for work-related injuries and illnesses were in excess of 5 billion dollars in 2010, with an additional 2.3 billion spent on injuries and illnesses from previous years1). Investigations of age group differences in the risk of work-related injury and illness and their consequences are common among epidemiological research in this area. This research generally suggests that the rate of injuries at work declines with age2–4), but that when injuries do occur, older age is associated with extended time away from work, increased health care costs and greater risk of reinjury5, 6).
The past three decades have seen dramatic demographic changes in the Canadian labor force. The percent of the labor force aged 45 years and over has increased from 23% in 1985 to 36% in 20077). Among 55 to 64 year olds, the participation rate (the percent of the population who are currently working or looking for work) increased from 31.5% in 1995 to 43.1% in 2007, with increases also occurring among workers over the age of 65 years7). These changes coincide with findings from the 1991 Survey on Ageing and Independence, and the 2002 and 2007 General Social Surveys, which demonstrate that the average age of planned retirement has been increasing among Canadians in their late 40s and early 50s8).
It is important to understand how these changes in the age composition of the labor market might influence the population burden of work-related injury and illness. Data from the 2007 General Social Survey demonstrated that workers with the lowest levels of self-rated health were the ones most likely to want to stay in the labor market until 65 years of age or older and least likely to expect an adequate income in retirement8). If less healthy workers remain in the labor market due to choice or financial necessity, this could in turn increase the rate of injury among older age groups, relative to younger workers in more recent time periods.
A recent study in Ontario demonstrated that while rates of injuries among older workers generally declined more rapidly than rates of injuries among younger workers between 1991 and 1998, the opposite trend was observed between 1999 and 2007, with rates of injuries among younger workers declining more sharply than older workers9). The authors of this paper suggested that this may be due to a greater percentage of primary prevention resources in Ontario being directed toward specific programs of injury prevention among younger (often male) workers since 2000. However, this relatively slower decline in the injury rate among older age groups may also be due to changes the composition of the older workforce as outlined above. It is important to understand if similar changes in the relative injury rates across age groups have occurred in other jurisdictions.
The relationship between age and risk of injury may also differ by the nature of injury2, 10). While younger age has been associated with a higher risk of all-cause work injury, a recent systematic review concluded that older age is associated with an elevated risk of musculoskeletal disorders11). Physiological changes related to reductions in mineral density and the deterioration of the microarchitecture of the bone may result in a higher risk of fractures among workers over the age of 5012). In addition, decreases in flexibility and range of motion associated with the replacement of elastin in the muscles and ligaments with collagen may result in higher rates of sprains and strains and repetitive motion injuries as workers age13). A previous study in Ontario found higher rates of sprains and strains among workers 25 and older compared with those less than 25, whereas workers less than 25 had higher rates of cuts and open wounds10). It is important to recognize that age differences in the nature of work-related injury may also arise from differences in occupational hazards experienced by older and younger workers2).
The objective of this paper is to examine the relationship between age and the incidence rate of lost-time compensation claims in the Canadian province of British Columbia (BC) over three time periods (1997–98, 2001–02 and 2005–06). Specifically, we want to examine if the relationship between age and work injury is consistent across time periods and across injury types. Similar to previous studies, we expect to find an inverse relationship between age and overall work injury rates. However, we hypothesize that the relationship between age and work injury will have changed over time, with the inverse relationship between age and work injury no longer being present in later time periods. We also hypothesize that older workers will be more likely to have work-related fractures and sprain and strain injuries compared with younger workers, but that younger workers will be more likely to have open wound injuries compared with older workers. Given the increasing number of older workers and differing trends between age and work injury previously reported, this paper offers new insights related to the primary prevention of work-related injury and illness in the context of a changing Canadian labor market. Given the ageing of the labor markets in many developed economies around the world, these objectives are also relevant for many countries including those in the Asia region, in particular Japan, which has both an ageing population and higher labor market participation rates among older male workers than those currently in British Columbia.
This study utilized short- and long-term disability claims (i.e., claims involving time away from work) reported to WorkSafe British Columbia (WorkSafeBC) in three separate time periods: 1997–98, 2001–02 and 2005–06. WorkSafeBC is the monopoly provider of workers' compensation in BC and covers approximately 93% to 94% of the labor force1). Those excluded from coverage with WorkSafeBC are self-employed workers without employees and minor segments of the employed labor force (e.g., athletes, domestic workers employed for less than eight hours per week). Information provided with each claim includes the date and nature of the injury/illness, the gender and age of the claimant, the industry they were employed in (using a classification system developed by WorkSafeBC, which is based on the Canadian Standard Industry Classification 198014)), and the type of occupation they were working in when injured, coded to the National Occupational Classification15).
Claim rates across each time period of interest were calculated by combining claim counts with labor force estimates generated from the Canadian Labour Force Survey (LFS). The LFS follows a complex, rotating panel sample design to efficiently estimate monthly changes in the Canadian labor force16). For the purpose of our analyses, we restricted the LFS estimates to only respondents from BC who were not self-employed (i.e., employees). To account for differences in the hours commonly worked by men and women and across age groups, all denominator counts were estimated as full-time equivalents (FTEs) using the usual number of hours worked as reported by each respondent in the week of the labor force survey. One full-time equivalent is equal to 2,000 hours of work per year.
Labor force estimates were generated across each age by year by gender by occupation by industry group (see below for additional information on these groupings).
The age of the claimant was defined as the age at time of injury using the claim record. Age was grouped into the following categories: 15 to 24 years, 25 to 34 years, 35 to 44 years, 45 to 54 years and 55+ years.
Occupational characteristics contained in the National Occupational Classification Career Handbook were assigned to each claim based on the claimant's occupational title17, 18). The Career Handbook characteristics were assigned to each occupational code by trained occupational analysts using a modified Delphi procedure18). We focused on the following five occupational characteristics:
The WorkSafeBC classification unit observed on each claim record was used to classify claimants as working in either a goods-producing or a services-producing industry. These classification units can be mapped to Standard Industrial Classification 1980 (SIC80) groups at an aggregate level. The goods-producing industry includes primary resources (agriculture, fishing, and forestry), utilities, manufacturing, construction and transportation/warehousing. The services-producing industry includes trade (retail and wholesale), accommodation/food/leisure services, business services, professional/scientific/technical services, education, health care and other services.
For analyses examining the relationship between age and specific types of injuries, we used the major 2-digit codes of the Canadian Standards Association Z-795 coding system for nature of injury19) to group claims into seven different groups. These were trauma to bones, nerves, spinal cord; intracranial injuries; open wounds; trauma to muscles, tendons, ligaments, joints; other Traumatic Injuries; MSK and connective tissue diseases/disorders; and other non-traumatic injuries (which included conditions such as hearing loss, inguinal hernias and carpel tunnel syndrome). Additional details of the types of injuries contained within each of these grouping are provided in Table A1 in the appendix.
Initial descriptive analyses estimated claim rates across each of our independent variables (age, time period, occupational characteristics and industry). As work tasks and the type of work injuries reported to compensation agencies differ for men and women20, 21), all analyses were stratified by gender. We conducted three separate sex-stratified regression analyses. Given we had count data within each age by time period by occupation by industry group, we had planned to use Poisson regression. However, the results of the Poisson model indicated that the data were overdispersed so we employed negative binomial regression models. The first regression model assumed no interactions between any of the variables. The second model investigated the possibility of an interaction between age group and time period by including a product term in the regression model. A final series of regression models examined if the relationship between age and work injury was consistent across different nature of injury groups. As our labor force survey denominator data was stratified separately for each occupational characteristic (i.e., we had separate denominator tables for occupational strength requirements, body position, exposure to noise, working with equipment, and working in an unregulated indoor environment), we could only adjust our regression models for one occupational characteristic at a time. For analyses examining the interaction between age and time period and between age and different nature of injury groups, we compared the results from models that were separately adjusted to each of our occupational characteristics. As the conclusions from each of these models were the same, we have presented models adjusted for occupational strength requirements in this paper, as this occupational factor was most strongly related to rates of work injury in our preliminary regression models.
Table 1 presents descriptive information of the number of claims, number of full-time equivalents (FTEs) and claim rates per 1000 FTEs, plus unadjusted risk estimates calculated from a negative binomial regression model. The relationship between age and claim rates was generally flat for both men and women. For both men and women, there was a reduction in claim rates across our study years. Among both men and women, higher claim rates were present among goods-producing industries; occupations with higher physical demands; occupations that require other body positions such as bending, stooping, kneeling and crouching; occupations with exposures to noise; occupations with hazardous equipment; and occupations with an unregulated indoor environment.
| Men | Women | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Claims | FTEs | Rate/1,000 FTEs | RR‡ | 95% CI | Claims | FTEs | Rate/1,000 FTEs | RR‡ | 95% CI | |
| Age group | ||||||||||
| 15 to 24 | 42,261 | 635,749 | 66.5 | 0.88 | 0.50–1.56 | 14,719 | 531,667 | 27.7 | 0.83 | 0.50–1.38 |
| 25 to 34 | 70,704 | 1,149,995 | 61.5 | 1.00 | 0.56–1.77 | 23,997 | 936,838 | 25.6 | 0.85 | 0.52–1.41 |
| 35 to 44 | 77,963 | 1,271,909 | 61.3 | 1.00 | 34,119 | 1,010,854 | 33.8 | 1.00 | ||
| 45 to 54 | 54,776 | 1,034,882 | 52.9 | 0.88 | 0.49–1.56 | 30,790 | 917,990 | 33.5 | 0.97 | 0.58–1.60 |
| 55 plus | 27,359 | 506,242 | 54.0 | 0.91 | 0.51–1.62 | 13,037 | 363,543 | 35.9 | 0.77 | 0.47–1.28 |
| Time period | ||||||||||
| 1997–98 | 99,950 | 1,461,245 | 68.4 | 1.00 | 39,078 | 1,148,334 | 34 | 1.00 | ||
| 2001–02 | 87,888 | 1,504,094 | 58.4 | 1.06 | 0.68–1.66 | 38,910 | 1,240,294 | 31.4 | 0.98 | 0.66–1.46 |
| 2005–06 | 85,225 | 1,633,438 | 52.2 | 0.96 | 0.62–1.50 | 38,674 | 1,372,265 | 28.2 | 0.84 | 0.57–1.25 |
| Industry | ||||||||||
| Goods | 167,858 | 1,566,180 | 107.2 | 2.16 | 1.54–3.05 ** | 17,499 | 353,640 | 49.5 | 1.67 | 1.22–2.28 ** |
| Services | 105,205 | 3,032,596 | 34.7 | 1.00 | 99,163 | 3,407,253 | 29.1 | 1.00 | ||
| Strength requirements | ||||||||||
| Limited | 13,565 | 1,832,294 | 7.4 | 1.00 | 17,681 | 2,056,725 | 8.6 | 1.00 | ||
| Light | 27,205 | 988,129 | 27.5 | 3.18 | 2.54–2.97 ** | 28,054 | 874,935 | 32.1 | 4.39 | 3.57–5.41 ** |
| Medium | 95,490 | 922,146 | 103.6 | 11.54 | 9.23–14.42 ** | 57,789 | 646,014 | 89.5 | 11.70 | 9.51–14.4 ** |
| Heavy | 136,803 | 856,207 | 159.8 | 17.25 | 13.81–21.56 ** | 13,138 | 183,218 | 71.7 | 11.12 | 9.03–13.68 ** |
| Body position | ||||||||||
| Sit | 27,548 | 1,172,426 | 23.5 | 1.00 | 11,526 | 1,548,578 | 7.4 | 1.00 | ||
| Stand/walk | 17,095 | 451,071 | 37.9 | 0.97 | 0.71–1.34 | 20,021 | 574,617 | 34.8 | 3.72 | 3.09–4.48 ** |
| Sit/stand/walk | 23,989 | 1,079,915 | 22.2 | 0.58 | 0.42–0.79 ** | 16,443 | 623,965 | 26.4 | 3.39 | 2.81–4.08 ** |
| Other | 204,431 | 1,895,365 | 107.9 | 2.59 | 1.89–3.55 ** | 68,672 | 1,013,733 | 67.7 | 8.17 | 6.80–9.82 ** |
| Noise exposure | ||||||||||
| No | 85,638 | 3,160,470 | 27.1 | 1.00 | 97,384 | 3,399,714 | 28.6 | 1.00 | ||
| Yes | 187,425 | 1,438,307 | 130.3 | 3.76 | 3.06–4.61 ** | 19,278 | 361,179 | 53.4 | 2.35 | 1.88–2.95 ** |
| Equipment exposure | ||||||||||
| No | 72,285 | 2,694,475 | 26.8 | 1.00 | 80,183 | 2,851,265 | 28.1 | 1.00 | ||
| Yes | 200,778 | 1,904,302 | 105.4 | 2.84 | 2.21–3.63 ** | 36,479 | 909,628 | 40.1 | 1.96 | 1.62–2.37 ** |
| Unregulated indoor environment | ||||||||||
| No | 164,601 | 3,876,451 | 42.5 | 1.00 | 105,823 | 3,574,656 | 29.6 | 1.00 | ||
| Yes | 108,462 | 722,326 | 150.1 | 2.83 | 2.22–3.61 ** | 10,839 | 186,237 | 58.2 | 2.18 | 1.88–2.54 ** |
Adjustment for other variables had a negligible impact on the unadjusted estimates (see Table 2). We did not observe any interaction between age and lost-time claim risk over the years in our study, suggesting that the relationship between age and claim risk has remained relatively stable for both men and women over the three time periods we examined. The results of this analysis are presented in Table A2 in the appendix.
| Males | Females | |||
|---|---|---|---|---|
| Variable | RR | 95% CI | RR | 95% CI |
| Age group | ||||
| 15 to 24 | 0.87 | 0.74–1.03 | 0.83 | 0.68–1.02 |
| 25 to 34 | 0.93 | 0.79–1.10 | 0.84 | 0.69–1.03 |
| 35 to 44 | 1.00 | 1.00 | ||
| 45 to 54 | 0.95 | 0.80–1.11 | 0.99 | 0.81–1.21 |
| 55 plus | 1.01 | 0.86–1.20 | 0.91 | 0.74–1.12 |
| Time period | ||||
| 1997–98 | 1.00 | 1.00 | ||
| 2001–02 | 0.89 | 0.78–1.01 | 0.92 | 0.78–1.08 |
| 2005–06 | 0.76 | 0.67–0.87 ** | 0.79 | 0.67–0.92 ** |
| Industry | ||||
| Goods | 1.37 | 1.30–1.44 ** | 1.16 | 1.09–1.24 ** |
| Services | 1.00 | 1.00 | ||
| Strength requirements | ||||
| Limited | 1.00 | 1.00 | ||
| Light | 3.44 | 2.96–3.99 ** | 4.18 | 3.47–5.03 ** |
| Medium | 12.16 | 10.49–14.10 ** | 11.27 | 9.37–13.55 ** |
| Heavy | 17.00 | 14.65–19.72 ** | 10.01 | 8.26–12.12 ** |
| Body position | ||||
| Sit | 1.00 | 1.00 | ||
| Stand/walk | 1.40 | 1.08–1.80 ** | 3.95 | 3.43–4.54 ** |
| Sit/stand/walk | 0.80 | 0.62–1.02 | 3.45 | 3.00–3.96 ** |
| Other | 3.33 | 2.61–4.24 ** | 8.32 | 7.25–9.55 ** |
| Noise exposure | ||||
| No | 1.00 | 1.00 | ||
| Yes | 3.76 | 3.44–4.12 ** | 2.04 | 1.68–2.47 ** |
| Equipment exposure | ||||
| No | 1.00 | 1.00 | ||
| Yes | 3.07 | 2.74–3.44 ** | 1.79 | 1.53–2.09 ** |
| Unregulated indoor environment | ||||
| No | 1.00 | 1.00 | ||
| Yes | 3.13 | 2.82–3.47 ** | 2.18 | 1.88–2.54 ** |
Table 3 presents the relationship between age and different nature of injury groups after adjustment for industry, year and occupational strength characteristics. Among both men and women, younger age was associated with a greater risk of open wound injuries; older age was associated with a greater risk of traumatic bone, nerve and spinal cord injuries; and middle age (35 to 44 years) was associated with the highest risk of trauma to muscles, tendons, ligaments, joints, and musculoskeletal and connective tissue diseases and disorders. The relationship between age and lost-time claims for intracranial injuries varied by sex, with younger men and older women at an increased risk of these types of injuries compared with 35 to 44 year olds. Older men, but not older women, had a greater risk for other nontraumatic injuries, with this effect being produced mainly by an elevated risk of hearing loss claims among older men.
| Trauma to bones, nerves, spinal cord | Intracranial injuries | Open wounds | Trauma to muscles, tendons, ligaments, joints | Other trauma | MSK and connective tissue diseases/disorders | Other non-trauma | ||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Sex | Age | n | RR | 95% CI | n | RR | 95% CI | n | RR | 95% CI | n | RR | 95% CI | n | RR | 95% CI | n | RR | 95% CI | n | RR | 95% CI |
| Women | 15 to 24 | 564 | 0.75 | 0.61–0.92 ** | 255 | 1.07 | 0.91–1.26 | 2,149 | 1.35 | 1.17–1.55 ** | 6,934 | 0.66 | 0.55–0.79 ** | 3,807 | 1.06 | 0.91–1.23 | 602 | 0.54 | 0.45–0.65 ** | 408 | 0.6 | 0.46–0.79 ** |
| 25 to 34 | 849 | 0.73 | 0.60–0.89 ** | 300 | 0.9 | 0.78–1.05 | 2,011 | 0.97 | 0.84–1.10 | 14,053 | 0.78 | 0.65–0.94 ** | 4,758 | 0.89 | 0.77–1.03 | 1,177 | 0.69 | 0.58–0.82 ** | 848 | 0.73 | 0.56–0.95 * | |
| 35 to 44 | 1,287 | 1.00 | 358 | 1.00 | 2,301 | 1.00 | 21,050 | 1.00 | 5,992 | 1.00 | 1,874 | 1.00 | 1,256 | 1.00 | ||||||||
| 45 to 54 | 1,575 | 1.32 | 1.09–1.59 ** | 373 | 1.12 | 0.96–1.29 | 1,817 | 0.87 | 0.76–1.00 * | 19,157 | 0.96 | 0.80–1.15 | 5,198 | 0.93 | 0.80–1.07 | 1,537 | 0.96 | 0.81–1.14 | 1,133 | 0.92 | 0.71–1.20 | |
| 55 plus | 1,231 | 2.39 | 1.97–2.90 ** | 173 | 1.22 | 1.02–1.46 * | 700 | 0.79 | 0.68–0.92 ** | 7,639 | 0.88 | 0.73–1.06 | 2,331 | 0.98 | 0.84–1.14 | 510 | 0.73 | 0.61–0.89 ** | 453 | 0.85 | 0.65–1.12 | |
| Men | 15 to 24 | 2,958 | 0.68 | 0.58–0.80 ** | 615 | 1.22 | 1.10–1.35 ** | 9,036 | 1.67 | 1.43–1.95 ** | 17,756 | 0.7 | 0.61–0.81 ** | 10,202 | 1.13 | 0.99–1.28 | 883 | 0.47 | 0.39–0.56 ** | 811 | 0.6 | 0.51–0.70 ** |
| 25 to 34 | 5,351 | 0.82 | 0.71–0.96 ** | 813 | 1.11 | 1.00–1.22 * | 10,355 | 1.2 | 1.03–1.40 * | 35,203 | 0.9 | 0.79–1.03 | 15,199 | 1.02 | 0.89–1.15 | 2,175 | 0.76 | 0.64–0.90 ** | 1,608 | 0.75 | 0.65–0.87 ** | |
| 35 to 44 | 6,768 | 1.00 | 829 | 1.00 | 9,115 | 1.00 | 40,662 | 1.00 | 15,392 | 1.00 | 2,942 | 1.00 | 2,255 | 1.00 | ||||||||
| 45 to 54 | 5,624 | 1.13 | 0.98–1.32 | 587 | 0.88 | 0.79–0.98 * | 5,770 | 0.81 | 0.70–0.95 ** | 28,690 | 0.94 | 0.82–1.07 | 9,993 | 0.88 | 0.78–1.00 | 1,998 | 0.93 | 0.79–1.10 | 2,114 | 1.24 | 1.08–1.42 ** | |
| 55 plus | 3,245 | 1.41 | 1.21–1.64 ** | 343 | 1.05 | 0.92–1.19 | 2,760 | 0.82 | 0.70–0.97 * | 12,537 | 0.88 | 0.77–1.01 | 4,426 | 0.85 | 0.74–0.96 * | 807 | 0.8 | 0.67–0.96 * | 3,241 | 3.8 | 3.32–4.36 ** | |
The objective of this paper was to describe the relationship between age and lost-time injury claim rates in BC over three time periods, specifically examining if the age and work-injury relationship was consistent across time periods and across nature of injury categories. While we found no clear relationship between age and risk of all types of lost-time work-related injury or illness among either men or women, we did find important age-related differences in injury risk for different nature of injury categories. In line with our hypotheses, we found a negative relationship observed between age and open wound injuries, but a positive relationship observed for traumatic injuries to bones, nerves and the spinal cord. However, contrary to our hypotheses, we did not find that older workers were more likely to report sprain and strain injuries (trauma to muscles, tendons, ligaments and joints) compared to younger workers. Instead we found that the relationship between age and these types of injuries was an inverse U shape, with the greatest risk being present for men and women 35 to 44 years of age. We found no evidence that the relationship between age and work injury risk has changed over the time periods we examined.
Our results, however, should be interpreted in light of the following limitations. Our study relied on accepted workers' compensation lost-time claims in BC over three time periods. Previous reports from Canada have demonstrated that workers compensation claims do not represent all injuries and illnesses that may be eligible for compensation, with workers with injuries perceived as less severe, employed in precarious employment relationships or with injuries involving a more complex etiology being less likely to file compensation claims22–25). Specifically focusing on age differences, a study conducted in BC found that older workers were less likely than their younger counterparts to claim compensation for injuries that resulted in hospitalization24). Another Canadian-based study focusing on work-related absences of one week or greater found older workers were less likely to report income from workers' compensation sources in the year of their absence or the year following their absence23). As a result, administrative records from workers' compensation insurance plans may underestimate the injury burden among older workers compared with younger workers. While we have used an accepted Canadian coding standard to group nature of injury groups, this coding standard may not always accurately reflect the underlying clinical diagnosis for a given injury. In particular, while trauma to muscles, tendons, ligaments, joints and musculoskeletal and connective tissue diseases/disorders are separate sections of the nature of injury coding system, these two sections do contain some similar types of injuries (e.g., tendonitis); with the distinction between assigning an injury to one section over the other revolving around whether the onset of symptoms was acute or gradual in nature. Given this information is not always readily available to workers' compensation coding staff, there may be misclassification between these two nature of injury groups.
Our study also has a number of strengths, which include the use of a sample of claims from a Canadian province with near complete coverage of labor market participants by the one workers' compensation agency, and the inclusion of three separate time periods. We were also able to adjust our models for important occupational and industrial characteristics, and account for age-related differences in hours worked through the use of full-time equivalents.
Unlike a previous study in the Canadian province of Ontario, which reported that injury rates among younger workers declined more sharply than those among older workers, we did not find that the relative risk of injury across age groups changed over a similar time period9). The lack of decline in injury rates among younger workers in BC may be due to the rapid growth in the BC economy between 2003 and 2007, in particular in the construction industry26). This lead to much larger growth in the 15- to 24-year-old labour force in BC compared with Ontario over this time period, with employment growth often associated with relative increases in injury rates27, 28).
Our finding of no clear relationship between age and overall claim rates for either men or women differs from previous studies using compensation claims in Ontario, which have reported an inverse relationship between age and injury risk, in particular among male labor market participants9, 10, 29). This difference could be due to either a lower rate of injury among younger workers in BC compared with Ontario or a higher rate of injury among workers 35 and older in BC. Previous cross-jurisdictional comparisons of self-reported work-related injury rates have noted that rates of injury are higher in BC than Ontario30), including specifically among workers aged 15 to 24 years of age31). A comparison of crude lost-time claim rates in BC and Ontario between 2000 and 2007 suggest that the lack of a relative increased risk among younger workers in BC (compared with Ontario) is the result of comparatively higher rates of injury among workers 35 years of age and older in BC compared to Ontario, relative to the differences in injury rates of workers 15 to 24 years of age26). In addition, unlike a recent review11), we did not find that older age was associated with greater risk of musculoskeletal disorders; instead, we found that the risk of injury related to both trauma to muscles, and non-traumatic musculoskeletal diseases and disorders, was greatest for the 35 to 44 year age group in both men and women. The similar age gradients for each of these injury groups may reflect misclassification between each of these groups as discussed previously.
Our finding that the relationship between age and injury risk varies by nature of injury has important implications for primary prevention programs. The choice to focus on younger (or older) ages to identify workers at risk of injury depends on whether primary prevention is targeting injuries in general or a specific nature of injury. Similar to other studies4, 29), we found the strongest predictors of overall injury risk were occupational characteristics. We did find an elevated risk of traumatic injuries to the bones, nerves and spinal cord among older men and women. This risk was predominantly due to an elevated risk of fracture injuries among older workers, in particular female labor market participants. This increased risk of fracture could potentially be the result of decreased bone mineral density and other changes in bone microarchitecture that occur with age12). Conversely, younger workers had an elevated risk of open wounds, which is consistent with a previous paper examining differences in injury types across age groups10). We suspect that this elevated risk, rather than being related to something inherent to younger workers, is likely due to specific differences in occupations worked between younger and older workers, which are not accounted for with our broad occupational groupings32). Finally, we found no evidence that the relationship between age and injury risk has changed over time. As a result, the greater participation of older workers and the potential for older workers to remain in particular occupational groups as outlined in our introduction has yet to result in any changes in the relationship between age and injury risk in BC.
This study was supported with funds from WorkSafeBC and the Workers' Compensation Board of Nova Scotia through the Focus on Tomorrow research program (RS2009-OG03). WorkSafeBC approved access to and use of the data facilitated by Population Data BC for this study. During the study, Peter Smith was supported by a New Investigator Award from the Canadian Institutes of Health Research. He is currently supported by a Discovery Early Career Researcher Award from the Australian Research Council. M. Koehoorn was supported in part by a Michael Smith Foundation for Health Research Senior Scholar Award. Approval for the secondary data analyses was obtained through the University of Toronto, Health Sciences I Ethics Review Committee. The Toronto Region Statistics Canada Research Data Centre provided access to the microdata used for this study for generating denominator estimates. Access to the public use files for the Labour Force Survey was made available through the Statistics Canada's Data Liberation Initiative via the University of Toronto.
