Are men really more than nine in ten workplace deaths, even allowing for hours worked?

In the United States, the Bureau of Labor Statistics counts every fatal work injury it can verify from multiple sources. In 2023 there were 5,283: 4,832 men and 447 women, so women were 8.5%. The count covers injuries, not deaths from long-term occupational disease.

In Canada, the Association of Workers’ Compensation Boards reports deaths accepted for compensation by provincial and territorial boards. In 2024 there were 1,046: 987 men, 55 women and 4 of unknown gender. These include deaths from occupational disease, such as illness from exposures years earlier, and only cover workers insured by a compensation board.

Men work more paid hours in total, so some difference in deaths would be expected even if every job were equally dangerous. BLS publishes rates based on hours worked. In 2023 men worked 56% of all hours, but the fatal injury rate was 5.7 per 100,000 full-time equivalent workers for men and 0.7 for women, about eight times higher.

So hours worked explain very little of the difference. Canada does not publish an equivalent rate by gender, so this adjustment can only be shown with US data.

Non-fatal injuries are much more evenly split. In Canada in 2024, men made up about 54% of accepted lost-time injury claims, compared with about 94% of accepted deaths. The large gap appears in the most severe outcomes, not in injury overall.

Supported: men account for more than nine in ten workplace deaths in both countries, and in the US the gap remains large per hour worked.

Not established by these figures: the reasons. The obvious candidates are differences in occupation and industry, such as construction, transportation and resource extraction, and in the tasks done within an occupation. The count tables alone cannot say how much each contributes, or whether factors such as risk-taking norms, training or enforcement play a part.

The US figure counts injury deaths in the year they happen. The Canadian figure counts compensated deaths, including occupational disease, in the year the claim is accepted, and leaves out workers not covered by a board, such as many self-employed people. The direction is the same in both, but the totals and shares are built differently.

The next step is rates by sex within the same occupation and industry, which would show how much of the gap is about which jobs men and women do. Where the data allow it, a future version of this explainer will add them.

Small counts are suppressed in the Canadian tables, and gender is occasionally unknown in both countries.