Why was this study done?
Based on research that shows physical activity is linked with better health outcomes, adults are encouraged to do at least 150 minutes a week of moderate to high intensity leisure-time exercise. However, studies are increasingly showing that some forms of physical activity at work can be linked with higher risks of heart disease and diabetes. This study set out to examine the effects of leisure-time exercise on risk of diabetes among people in jobs with varying levels of physical demands.
How was the study done?
The study followed 7,026 workers aged 35-74 from Ontario, Canada from 2003 to 2017. These workers were respondents to the 2003 Canadian Community Health Survey (CCHS), which collected information on personal factors, health conditions, health behaviours and work conditions. Based on survey participants’ job titles, as well as job exposure information in the Human Resources and Skills Development Canada Career Handbook, the team estimated respondents’ minimum strength and body postures at work. The team then sorted survey participants into three groups:
- people who were mainly sedentary at work (either sitting or standing) and handled light loads of up to 10 kilograms (e.g. bank tellers, office workers);
- people who handled light loads but were frequently on the move (e.g. retail salespersons, grocery store workers); and
- people who handled heavier loads of 10 to 20 kilograms (e.g. construction workers, hospital orderlies, labourers).
Using participants’ answers about the amount of exercise they did outside of work, the team further sorted the participants into two groups: people who were physically active (i.e. who did more than 150 minutes of exercise a week) and those who were not (i.e. who did less than 150 minutes). After these two classification steps, the team ended up with six groups of workers (for example, those who mainly sat or stood at work and met physical activity guidelines in their leisure time, those who mainly sat or stood at work but did not meet physical activity guidelines off work, and so on).
The researchers then linked the CCHS information to administrative health records from the Ontario Diabetes Database, which collects diabetes history based on hospital discharge and physician/laboratory claims, to identify people who developed a new case of diabetes over 15 years (2003-2017). To make sure participants in the study did not have pre-existing diabetes at the time of the survey, all participants diagnosed with diabetes before 2005—i.e. up to two years of follow-up—were removed from the sample.
What did the researchers find?
Among people who worked in sedentary jobs, meeting physically activity guidelines reduced the risk of developing diabetes by about 37 per cent. Among people who did light loads and were on the move at work, an eight to 10 per cent reduction in diabetes risk was found, but this was not statistically significant and may have been due to chance. Among people who did heavy loads at work, a seven to 10 per cent increase in diabetes risk was found, but this was also not statistically significant.
The above results were found even after controlling for a wide range of variables, including those related to sociodemographic characteristics (e.g. household makeup, immigrant status, geography), health factors (e.g. physical or psychological disabilities, alcohol or cigarette use, body mass index) and work factors (e.g. shift schedule, hours of work a week).
What are the implications of the study?
The public health recommendation for people to do at least 150 minutes of exercise a week to reduce diabetes risk should be targeted mainly to people who are sedentary at work. Leisure-time exercise may not have any additional benefits for reducing diabetes risk in people in physically demanding work, nor is there strong evidence that physically demanding work can increase diabetes risk.
What are some strengths and weaknesses of the study?
A strength of this study was its long follow-up time. Another strength was its use of administrative data to measure the incidence and risk of diabetes in the sample. A third strength was the use of job titles to estimate the level of physical activity at work. This method, while an approximate measure of physical job demands, helps reduce some of the biases seen in previous studies that relied on self-reported physical job demands. A fourth strength was the team’s ability to take into consideration a broad range of potentially confounding variables.
A weakness of this study was its use of self-reported physical activity, which can be biased due to respondents overestimating their actual physical activity levels. Another weakness was its one-time measure of physical activity levels, which may change over the 15-year follow-up period.