Summary In 2012, employment injuries with permanent physical or mental impairment (PPMI) constituted 12% of all traumatic accidents (TA) and musculoskeletal disorders (MSD) accepted, but they resulted in 50% of days compensated and disbursements made by the CNESST for these types of injuries. Yet there is limited knowledge about the groups of workers who are at the greatest risk of incurring this type of injury in Québec and the associated factors. Using data from 2010–2012, this study aimed to identify the subgroups of workers with the highest risk of suffering a TA or MSD with PPMI, and to analyze the temporal evolution of this risk among various subgroups, from 2003 to 2012. Analyses were performed according to age, gender, occupational category and industry. The study also made it possible to determine, among the 4 factors analyzed, those that are the most strongly associated with the advent of a PPMI, by controlling the effect of the other variables analyzed. The risk was estimated using full-time equivalent (FTE) frequency rates. These were calculated by dividing the number of TA and MSD with PPMI accepted by the Commission des normes, de l’équité, de la santé et de la sécurité du travail (CNESST) with the numbers of FTE workers, using data from the National Household Survey (NHS) between 2010 and 2012 and those of the Labour Force Survey (LFS) for analyses performed between 2003 and 2012. For the analysis of recent data (2010–2012), the research team used negative binomial regression models to evaluate the associations between the FTE frequency rate and age group, gender, occupational category and industry. The effect associated with each independent variable is presented in the form of relative risk (RR), measured by using frequency rate reports. All these analyses were carried out separately for the TA and for the MSD with PPMI. In 2010–2012, analyses based on age revealed that workers aged 55 or over have an FTE frequency rate of accepted TA with PPMI that is twice as high as those aged 15 to 24. This association between age group and FTE frequency rate differs in a statistically significant manner depending on gender. For instance, among women, the RR of those aged 55 or over compared to those aged 15 to 24 is over 3, while it is below 2 among men. This increase in risk according to age also holds true for accepted MSD with PPMI, but the effect is amplified when the oldest workers have an RR above 6 compared to those aged between 15 and 24. With respect to the MSD risk according to age, there are no significant differences between the 2 genders. The gender-based analysis indicates that the risk of TA with PPMI among women is 76% of that of men in 2010–2012. However, this difference in risk between men and women varies according to age group, occupational category and industry. With regard to the effect of gender as it relates to age group, the analysis found that the risk faced by women aged 15 to 24 is equivalent to half of that faced by young men, while no statistically significant difference was noted between genders among those aged 55 and over. There is also no statistically significant gap between men and women among non-manual workers in the majority of service industries, while in other occupational categories and industries there is statistically higher risk among men. With respect to risk of MSD with PPMI, overall, there is no statistically significant difference between men and women. However, the effect of gender varies according to the industry. In 2 goods-producing industries, women’s risk is statistically lower than that of men, while it is higher in 2 service industries. The occupational category was the independent variable with the strongest effect associated with FTE frequency rates, for both accepted TA and MSD, between 2010 and 2012. For TA, compared to the non-manual category, the RR of manual workers is approximately 9, and 4 for those in the mixed category. With respect to MSD, relative risks are 13 for manual workers and 4 for mixed occupations. These RR differ according to the industry for both TA and MSD, but in each industry, the RR of manual workers is the highest, followed by those in mixed occupations, and all are significantly different statistically when compared with non-manual workers. Industry-related RR sometimes varies by a factor of 2 among the 9 groups of industries, for both the risks of TA and MSD accepted in 2010–2012. For TA, no industry has a statistically higher risk than that of manufacturing and repair and maintenance services, the reference category for the industry variable. With respect to MSD, however, health and social assistance services are the only industries in which the risk is significantly higher statistically than that of manufacturing and repair, and maintenance services (RR equals 1.4). In addition, gender and the occupational category are variables modifying the effect of industry for both TA and MSD. The study also made it possible to identify 23 industries/occupational categories among men, and 22 among women, for which the FTE frequency of TA accepted with PPMI is twice as high as the average rate in 2010–2012 (men: 3.8‰; women: 1.5‰). They include 17% of workers, but 48% of TA with PPMI for men, while the proportions are 12% and 39%, respectively, for women. With respect to FTE frequency rates of MSD accepted with PPMI, 18 industries/occupational categories for men, and 9 for women, have a rate that is twice as high as the average rate in 2010–2012 (men: 1.4‰; women: 0.9‰). These industries/occupational categories correspond to 16% of workers and 45% of MSD with PPMI for men. Those proportions are 8% and 37%, respectively, for women. The analysis of changes in FTE frequency rates between 2003 and 2012 is also based on negative binomial regression. It appears that the average annual FTE frequency rate has decreased by 4.1% for TA and 5.4% for MSD. This difference is statistically significant. The decrease was not as substantial among older workers, women and non-manual workers; the differences are statistically significant. However, in the case of MSD, the decrease was less for men, but the difference with women is not statistically significant. We also determined 9 industries for the TA and 6 industries for the MSD in which the changes were less favourable and statistically different than the industry with the greatest decrease in its FTC frequency rate. Finally, changes to the composition of Québec’s industrial sector and, to a lesser extent, demographic changes between 2003 and 2012, appear to have contributed to accelerating the decrease in the overall FTC frequency rate of TA accepted with PPMI. With regard to MSD, industrial and demographic changes appear to have opposing effects (the first accelerating the decrease and the second slowing it down), which, to all intents and purposes, cancel each other out.