The Canadian population is aging. Canadian Census data indicate that the percent of the population aged 50 to 74 years has increased dramatically in the past two decades, from 20.6% in 1986 to 21.4% in 1996 and 26.4% in 2006.1 These aging trends have important implications for labor markets in Canada and elsewhere in the developed world. The decline in various aspects of physical and cognitive function and increases in chronic conditions with age is well documented.2–5 A recent report from the Health Council of Canada, focusing on seven long-term chronic conditions (arthritis, cancer, chronic obstructive pulmonary disease, diabetes, heart disease, hypertension, and mood disorders) reported that one third of Canadians have one or more of these conditions, with 11% having two or more conditions.6 Although the prevalence of some chronic conditions has been stable over the last 20 years, others are increasing rapidly. For example, Type 2 diabetes and hypertension prevalence rates have both increased substantially, exceeding previous predictions, in part because of underestimations of increases in obesity.7,8 Rates of arthritis have also increased above what would be predicted on the basis of demographic changes alone.9
Understanding the impact of chronic conditions on labor market experiences, in particular on work injury, is an important area of research because of the aging labor market in many developed countries. Cross-sectional data from the Canadian National Population Health and Canadian Community Health Surveys (CCHS)10,11—all based on representative population-based samples—indicate that the prevalence of arthritis, hypertension, heart disease, diabetes, and back problems have all increased between 1994 and 2007 among the working population, with some (hypertension and diabetes) more than doubling in prevalence.
Despite the importance of this research area, there are relatively few high-quality studies that have examined chronic conditions and work injuries. A review by Palmer and colleagues12 concluded that for prevalent chronic conditions such as diabetes, arthritis, and cardiovascular disease, the evidence base linking these conditions and an increased risk of work injury is “remarkably thin.” This review also identified a lack of high-quality studies in this area with many previous studies that often failed to provide controls for important confounders, or used measures of chronic conditions that did not include physician diagnosis.12 This review concluded from the available studies that there may be elevated risk of injury associated with diabetes (or taking diabetic medication), but there was less evidence for an elevated risk of injury associated with cardiovascular disease or arthritis.12
The objective of this article was to address this research gap by examining the relationship between five medically diagnosed chronic conditions associated with aging (arthritis, hypertension, heart disease, diabetes, and back problems) on risk of work injury, using two population-based samples in Canada. On the basis of the previous research in this area, we hypothesize that risk of work injury will be elevated for respondents with diabetes, with no elevated risk associated with the presence of other chronic conditions.
This study used secondary data from the 2003 and 2005 CCHS. Each CCHS uses a multistaged, stratified sampling frame to target individuals aged 12 years and older, living in private dwellings within Canada. People living on Indian reserves or Crown lands, residents of institutions, full-time members of the Canadian Armed Forces, and residents of certain remote regions are excluded from the sampling frame. The household response rate to the 2003 CCHS was 87%, with the selected person response rate at 93%.13 For the 2005 CCHS, the household response rate was 85%, with the selected person response rate at 94%.14 For the purpose of this article both samples were restricted to respondents aged 15 years and olderand who had been employed (excluding self-employed) at some point in the previous 12 months (n = 65,961 respondents to the 2003 cycle; n = 16,744 respondents in the 2005 cycle). The 2005 sample was smaller than the 2003 sample is because labor market questions were only posed to a random third of the CCHS sample in 2005. Approval for the secondary data analyses was obtained through the University of Toronto, Health Sciences I Ethics Committee.
Main Dependent Variables
Respondents were asked whether they had been injured, not including repetitive strain injuries, in the past 12 months seriously enough to limit their normal activities. Respondents who reported an activity limiting injury were further asked whether the injury occurred at work and whether they received medical attention for their injury within 48 hours. Using this information, respondents were classified as having a work-related injury (WRI) requiring medical attention (yes/no). Respondents with more than one injury were asked to detail their most serious injury.
Work-related Repetitive Strain Injury
Respondents were also asked whether they had been injured in the last 12 months due to repetitive strain that were serious enough to limit their normal activities. The examples of carpal tunnel syndrome, tennis elbow, and tendonitis were given. Respondents who answered yes to this question were further asked whether this repetitive strain injury (RSI) occurred while on the job. Respondents with more than one RSI were asked to detail their most serious injury. Using this information, we classified respondents as having an RSI (yes/no).
For WRIs, but not for RSIs, additional information was also available about the type and cause of the injury.
Main Independent Variables: Medically Diagnosed Chronic Conditions
CCHS respondents are asked for information about chronic conditions defined by the interviewer as “long-term conditions which are expected to or have already lasted 6 months or more and that have been diagnosed by a health professional.”13,14 Respondents were asked directly about 29 different conditions with the option to list additional chronic physical or mental conditions not included in the list provided. For the purpose of our analyses, we focused on five conditions that have a high prevalence among older age groups: hypertension, heart disease, diabetes, arthritis, and back problems.
Respondents who reported suffering from arthritis were asked to specify which kind of arthritis they had (osteoarthritis, rheumatoid arthritis, or other type)). Respondents who reported having diabetes were additionally asked how old they were at first diagnosis, from which the length of time since diabetes diagnosis was estimated (less than 1 year, 1 year to less than 5 years, 5 years to less than 10 years, 10 or more years).
The following measures were included as potential confounders in the relationship between chronic conditions and WRIs based on the quality assessment outlined in the review by Palmer and colleagues12: age (grouped); year of the survey; working hours in the previous 12 months; whether the respondent was an immigrant and whether they identified as part of a visible minority; highest level of education completed; interview language (English, French, or other); current living location (urban or rural); province of residence; and alcohol consumption (grouped).
Occupational characteristics were also included and assigned on the basis of the respondents' occupational title, using the National Occupational Classification Career Handbook.15,16 These occupational characteristics were assigned to each occupational code by trained occupational analysts using a modified Delphi procedure.16 Focus was on the following four occupational characteristics: strength requirements of the job; whether the occupation involved working near or with equipment, instruments, machinery, or power/hand tools that may be a potential source of injury; whether the occupation involved an environment where there is sufficient constant or intermittent noise to cause distraction or possible loss of hearing; and whether the occupation involved working inside where the temperature or humidity may be considerably different from normal room conditions (unregulated indoor climate).
The combined sample of respondents who were older than 15 years and had been employed in the previous 12 months was 82,705 respondents. Of this sample, 212 respondents (0.2%) were missing information about injuries in the previous 12 months and 2723 (3.3%) were missing information about either occupation code or number of hours worked in the previous 12 months, leaving a sample of 79,770. Respondents with missing information were more likely to be female, older, and recruited using the telephone frame of the CCHS, rather than the area frame (49% of the CCHS sample is recruited using a listing of households from the Labour Force Survey—the area frame; 50% are recruited from a phone listing from the Canadian Phone Directory; and 1% are recruited using random digit dialing). Of this sample, an additional 656 respondents (0.8%) were missing information about one of the covariates of interest, leaving a final sample of 79,114 respondents (96% of the original sample).
A preliminary descriptive analysis examined the incidence of WRIs and RSIs across each chronic condition of interest. A series of logistic regression models then examined the association between WRI and RSI and each chronic condition. Initial models included each chronic condition separately along with all potential confounding variables, except occupation. The second model included all chronic conditions simultaneously along with all potential confounding variables, except occupation. The final model involved adding occupation to the second model. The number of hours worked in the past 12 months was included as an offset variable in each of the models.17 Occupational characteristics were not included in the initial or second models because it is not clear whether these characteristics are confounders or mediators in the relationship between chronic conditions and WRIs and RSIs.
It is possible that chronic conditions, in particular back problems, may be the result of a WRI or RSI. To address the possibility of reverse causation in this situation, we removed all WRIs and RSIs where the back was the part of body injured when examining the relationship between chronic back conditions and WRI and RSI. An additional series of logistic models examined differences in risk of WRI and RSI across types of arthritis and the length of time with diabetes. A final set of descriptive analyses examined the types of WRIs and causes of WRIs across conditions with an elevated risk of injury in our logistic models. As work tasks and work injuries differ between men and women, even within similar occupations18,19, all logistic models were stratified by sex. We also examined potential moderation of the effects of chronic conditions on WRI and RSI by occupational characteristics (eg, does the association between arthritis and WRI differ across occupations with different strength requirements). These analyses did not find any consistent evidence of moderation by occupational characteristics in our sample for any chronic condition. To account for the complex sample design of the CCHS, in line with guidelines from Statistics Canada, the confidence intervals around each point estimate have been adjusted using a bootstrap technique.20 All analyses were weighted for the original probability of selection into the sample, and for nonresponse.
Table 1 presents descriptive information about the sample and the percentage of respondents reporting WRIs and RSIs in the previous 12 months. A total of 2.5% of our sample reported WRIs, and 7.2% reported RSIs. Focusing on chronic conditions, rates of WRI were higher among respondents with chronic conditions, compared with the rate in the total sample, with the highest rates observed among respondents reporting chronic back problems (4%) followed by respondents with arthritis (3.6%) and diabetes (3.5%). Similar patterns were observed for RSIs with the highest rates observed among respondents with arthritis and back problems (14.1%), followed by respondents with heart disease (10.2%).
TABLE 1-a. Descripti...Image Tools
Table 2 presents the odds ratios and 95% confidence intervals for each of the five chronic conditions examined with adjustment for confounders, other chronic conditions, and finally occupational characteristics. Elevated risks of WRI were observed for all conditions, with statistically significant elevated risks for arthritis and back problems among both men and women. Among men, in our final model, the risks associated with heart disease and diabetes were higher than those observed for back problems (1.39 and 1.34 vs 1.24), although the smaller number of respondents with these conditions resulted in confidence limits that crossed one. With the exception of men with diabetes, similar elevated risks were observed for chronic conditions and RSIs. The highest risk of RSIs was observed for arthritis and back problems, with a statistically significant risk of injury also observed among women with diabetes.
Table 3 presents additional information about the risk of injury associated with types of arthritis and length of time with diabetes. With the exception of unknown arthritis type among women, elevated risk of WRI and RSI was observed for all other types of arthritis. Patterns of risk were less clear for length of time with diabetes. For each injury type, there appeared to be a protective effect for recent diabetes diagnoses. Among women, there were statistically significant elevated risks of RSI for those with between 1 and 5 years, and those with more than 10 years, between the survey and their diabetes diagnosis, although the odds of RSI among women with diabetes for 5 to 9 years was close to one.
Table 4 compares the types and causes of WRIs among respondents with and without arthritis and with and without back problems. A lower percentage of respondents with arthritis and back problems reported their cause of injury contact with objects, while a higher percentage of these respondents reported injury due to a fall. A higher percentage of injured respondents with chronic back problems reported the nature of their injury was sprain and strain, or multiple injuries, concussions, or internal injuries.
The objective of this article was to examine the relationship between five medically diagnosed chronic conditions associated with aging (arthritis, hypertension, heart disease, diabetes, and back problems) on risk of WRI and RSI. We found an elevated risk of both WRIs and RSIs for all chronic conditions examined, with the exception of RSIs among men with diabetes. Statistically significantly elevated risks were observed for arthritis and back problems for both WRIs and RSIs, with diabetes also being associated with an elevated risk of RSIs among women. We further found similar risks of both WRI and RSI across arthritis types, although pattern of risk across length of time since diabetes diagnosis was less clear. Finally, a higher percentage of injured respondents with arthritis and back problems reported that the cause of their WRI was a fall compared with injured respondents without these conditions. These findings fill an important gap in knowledge about the impact of chronic conditions on work injury within the context of the aging Canadian workforce.
The results of this studyshould be interpreted given the following limitations. This study is cross-sectional, which allows for the possibility of reverse causation between both WRI and RSI and chronic conditions. To address this possibility, we removed WRI and RSIs where the back was the affected body part from analyses examining chronic back pain and work injury, which we believe to be the chronic condition most open to reverse causation. As such, we feel that reverse causation is unlikely to explain the findings presented in this article. We were also unable to examine the independent impact of medication use associated with chronic conditions on risk of WRI and RSI. Unfortunately, questions about medication use are only optional content in the CCHS and including these questions would have substantially reduced both the size and representativeness of our sample.
This study also has a number of strengths. The medical diagnosis of each condition, although self-reported, is an improvement on previous studies in this area, as is the ability to control for many potentially important confounders in the chronic disease and work injury relationship, with the exception of job tenure.12 In addition, these analyses used a representative population sample of the Canadian labor market.
The risks observed in this study for osteoarthritis are higher than those observed by Voaklander and colleagues,21 who reported no excess risk among male farmers older than 66 years. However, previous studies by Sprince and colleagues22 (also among agricultural workers) and Zwerling and colleagues23 (using a US population sample) have also both reported a statistically significant elevated risk for arthritis and work injury, albeit slightly weaker risks than those reported here. In addition, our finding of an elevated percentage of falls among respondents with arthritis is consistent with the research by Sprince and colleagues24 who have reported a higher odds of fall injuries associated with arthritis, compared with injuries in general22 among Iowa farmers. The studies by Sprince and colleagues22 and Voaklander and colleagues21 both reported no excess risk of injury associated with cardiovascular disease, whereas this study found an elevated, but not statistically significant, risk for heart disease and both WRIs and RSIs.
Our reported odds of WRI for diabetes (odds ratios of 1.34 among men and 1.31 among women) are smaller than those previously reported by Zwerling and colleagues23 and Voaklander and colleagues,21 but slightly higher than those reported more recently by Sprince and colleagues.25 Both Sprince and colleagues25 and Voaklander and colleagues21 reported an elevated risk of injury for respondents with longer duration of diabetes, although this group was only more than 6 months in the study of Voaklander and colleagues, whereas it was more than 13 years in the study of Sprince and colleagues. We did find elevated risks associated with duration of diabetes between 1 and 10 years among men and women for WRI (in each case not statistically significant), and among women with a length of time between 1 and 5 years, and more than 10 years, but not 5 to 9 years, with RSIs. These findings do not make a definitive statement about duration and diabetes and injury risk possible.
As outlined in the introduction the prevalence of chronic conditions within the Canadian labor market is increasing, mainly due to the aging of the Canadian population. The results of this study suggest that the increasing prevalence of chronic conditions, in particular arthritis, back problems, diabetes, hypertension, and heart disease, will have important implications for the prevention of work place injuries.
The Toronto Region Statistics Canada Research Data Center provided access to the micro-data used for this study. Access to the public use files for the National Population Health Survey and Canadian Community Health Survey was made available through the Statistics Canada's Data Liberation Initiative via the University of Toronto.
1. Statistics Canada. 1986, 1996 and 2006 Canadian Census Public Use Microdata Files. Ottawa, Ontario, Canada: Statistics Canada; 2007.
2. Ilmarinen JE. Aging workers. Occup Environ Med. 2001;58:546–552.
3. Savinainen M, Nygard C-H, Korhonen O, Ilmarinen JE. Changes in physical capacity among middle-aged municipal employees over 16 years. Exp Aging Res. 2004;30:1–22.
4. Costa G, Sartori S. Ageing, working hours and work ability. Ergonomics. 2007;50:1914–1930.
5. Schultz S, Kopec JA. Impact of chronic conditions. Health Rep. 2003;14:41–53.
6. Health Council of Canada. Why Health Care Renewal Matter: Learning From Canadians With Chronic Health Conditions. Toronto, Ontario, Canada: Health Council; 2007. available at: www.healthcouncilcanada.ca
7. Lipscombe LL, Hux JE. Trends in diabetes prevalence, incidence, and mortality in Ontario, Canada 1995–2005: a population-based study. Lancet. 2007;369:750–756.
8. Tu K, Chen Z-L, Lipscombe LL. Prevalence and incidence of hypertension from 1995 to 2005: a population-based study. CMAJ. 2008;178:1429–1435.
9. Perruccio AV, Power JD, Badley EM. Revisiting arthritis prevalence projections—it's more than just the aging of the population. J Rheumatol. 2006;33:1856–1862.
10. Statistics Canada. 2007 Canadian Community Health Survey Cycle 4.1. Ottawa, Ontario, Canada: Statistics Canada. Health Statistics Division [producer]; Statistics Canada. Data Liberation Initiative [distributor]; 2009.
11. Statistics Canada. 1994 National Population Health Survey. Ottawa, Ontario, Canada: Statistics Canada. Health Statistics Division [producer]; Statistics Canada. Data Liberation Initiative [distributor]; 2000.
12. Palmer KT, Harris EC, Coggon D. Chronic health problems and risk of accidental injury in the workplace: a systematic literature review. Occup Environ Med. 2008;65:757–764.
13. Statistics Canada. User Guide to the 2003 Canadian Community Health Survey. Ottawa, Ontario, Canada: Statistics Canada; 2005.
14. Statistics Canada. User guide to the 2005 Canadian Community Health Survey. Ottawa, Ontario, Canada: Statistics Canada; 2006.
15. Human Resources Development Canada. National Occupational Classification: NOC Training Tutorial. Ottawa, Ontario, Canada: Human Resources Development Canada; 2001:1–25.
16. Human Resources and Skill Development Canada. National Occupational Classification Career Handbook. Ottawa, Ontario, Canada: Government of Canada; 2011.
17. Agresti A. Categorical Data Analysis. New York, NY: John Wiley & Sons; 1990.
18. Messing K. Women workers and their working conditions. One-eyed Science: Occupational Health and Women Workers. Philadelphia, PA: Temple University Press; 1998:1–11.
19. Smith PM, Mustard CA. Examining the associations between physical work demands and work injury rates between men and women in Ontario, 1990–2000. Occup Environ Med. 2004;61:750–756.
21. Voaklander DC, Kelly KD, Rowe BH, et al. Pain, medication, and injury in older farmers. Am J Ind Med. 2006;49:374–382.
22. Sprince NL, Zwerling C, Lynch CF, et al. Risk factors for agricultural injury: a case-control analysis of Iowa farmers in the Agricultural Health Study. J Agric Saf Health. 2003;9:5–18.
23. Zwerling C, Whitten PS, Davis CS, Sprince NL. Occupational injuries among workers with disabilities - The National Health Interview Survey, 1985–1994. JAMA. 1997;278:2163–2166.
24. Sprince NL, Zwerling C, Lynch CF, et al. Risk factors for falls among Iowa farmers: a case-control study nested in the Agricultural Health Study. Am J Ind Med. 2003;44:265–272.
25. Sprince NL, Pospisil S, Peek-Asa C, Whitten PS, Zwerling C. Occupational injuries among workers with diabetes: the National Health Interview Survey, 1997–2005. J Occup Environ Med. 2008;50:804–808.
©2012The American College of Occupational and Environmental Medicine