Work Absence: What the 2026 Data Actually Says About Disability in Canada

Work absence has stopped growing. Disability prevalence hasn’t.

For anyone managing claims, that gap is the story of the decade.

If you manage disability claims in Canada, the most useful number released this year may be the one that didn’t move. According to Statistics Canada’s work absence table (14-10-0463-01, updated January 2026), the total incidence rate of work absence among full-time employees was 9.1% in 2023, 9.1% in 2024 and 9.1% in 2025. That follows a sharp post-pandemic step up from 8.4% in 2021 to 9.3% in 2022.

Three consecutive flat years is a plateau, not a trend. And it is easy to read that plateau as reassurance: the surge is over, absence has normalized, the pressure is off.

We think that reading is wrong, and the reason it’s wrong is what makes the rest of the 2026 data worth studying carefully.

Table of Contents hide

2 Geography that tracks institutions, not health

3 The cliff moved to week 26

4 Injuries down, occupational disease up

5 What the data still won’t tell us

6 Sources

Prevalence is rising while absence sits still

The 2022 Canadian Survey on Disability found that 27.0% of Canadians aged 15 and over, or 8.0 million people, had at least one disability. That is an increase of 4.7 percentage points since 2017. Every province recorded an increase. The steepest rises were in New Brunswick, up 8.6 points to 35.3%, and Nova Scotia, up 7.6 points to 37.9%.

The composition of that increase matters more than the headline. Mental health-related disability rose 3.2 percentage points to 10.4% of the total population, the largest single contributor to the overall increase. Among youth aged 15 to 24, the disability rate reached 20.1%, up roughly 7 points from 2017, and among young people with a disability, mental health-related conditions were the most common type at 68%.

So prevalence is climbing, its fastest-growing component is psychological, and its fastest-growing age group is the one at the start of its working life. Meanwhile, weekly absence incidence is flat. Those two facts can only be reconciled in a few ways: more conditions are being accommodated inside the workplace rather than converted into absence, more are presenting as reduced capacity rather than full absence, or the pressure is accumulating and simply hasn’t surfaced yet in the absence statistics.

For claims and return-to-work practice, all three possibilities point the same direction. Incidence is a lagging, low-resolution indicator. It tells you how many people were away last week. It tells you nothing about who is struggling at work, how long they will stay away when they do leave, or whether they will come back.

Geography that tracks institutions, not health

The provincial spread in 2025 is wide enough to be interesting on its own:

Province	2021	2025
Saskatchewan	9.3	10.8
Manitoba	8.6	10.4
Quebec	9.8	10.1
Prince Edward Island	9.6	10.1
Nova Scotia	9.7	10.0
New Brunswick	9.0	9.9
Newfoundland and Labrador	9.2	9.8
British Columbia	8.2	9.3
Alberta	7.9	8.3
Ontario	7.5	8.4

Total incidence rate, full-time employees, percent. Source: Statistics Canada Table 14-10-0463-01.

Ontario sits lowest at 8.4% and Saskatchewan highest at 10.8%, a 2.4-point spread. Manitoba and Saskatchewan posted the largest four-year increases, at 1.8 and 1.5 points respectively.

The most revealing comparison, though, is inside a single labour market. In the Ottawa-Gatineau census metropolitan area, the Quebec part recorded 13.1% in 2025 against 10.7% on the Ontario part. Same weather, same commuting shed, same employers in many cases, and a 2.4-point gap that has persisted across all five years in the table. Toronto, at 6.6%, is the lowest CMA in the country, well below the 8.8% CMA average and the 10.3% non-CMA average.

We would not claim that provincial borders cause illness. But when absence differs this much across a line that health status cannot plausibly explain, the sensible inference is that leave entitlements, benefit design, health system access, sectoral mix and workforce composition are doing substantial work. Absence rates are behavioural and institutional measures as much as clinical ones, which is exactly why a claim should be assessed on its own evidence rather than against a regional benchmark.

The cliff moved to week 26

The Employment Insurance Monitoring and Assessment Report for 2024-25 contains the year’s most consequential single statistic for long-term disability coordination. Sickness benefits continued to represent almost two-thirds of the 691,600 EI special benefit claims established in 2024-25. And 17.9% of sickness claims completed that year used the maximum 26 weeks of benefits, compared with about a third of claims that used the maximum 15 weeks before the December 2022 extension.

Read that carefully. The extension did not eliminate the exhaustion problem; it relocated it. Roughly one in six claimants still runs out of entitlement, but now at week 26 rather than week 15. The practical effect is that the handoff point between public wage replacement and private long-term disability has shifted by eleven weeks, and the population arriving at that handoff is a more selected, longer-duration, harder-to-resolve group than the one that used to arrive at week 15.

That has direct implications for when independent assessments are most useful. An assessment commissioned at week 20 of a 26-week entitlement is a different exercise, with different leverage over outcome, than one commissioned at week 12 of a 15-week entitlement. Return-to-work planning windows, capacity evaluations and vocational assessments all need to be re-timed against the new schedule rather than the old one.

For context on the broader program: EI regular claims reached 1.40 million in 2024-25, up 2.0%, with $14.3 billion paid, an 11.5% increase and, excluding pandemic years, a historical high. The increase in claims was driven mainly by an additional 23,400 claims from Ontario, while every Atlantic province and territory declined. The average weekly benefit rate for claims established in 2024-25 rose 3.9% to $581, per Chapter 2 of the report. Cost growth, in other words, is coming from benefit levels and beneficiary duration rather than from a volume surge.

Injuries down, occupational disease up

The workers’ compensation picture points in two directions at once. The 2026 Report on Work Fatality and Injury Rates in Canada, by Sean Tucker and Anya Keefe using AWCBC data, reports 1,042 accepted work-related death claims in 2024: 381 traumatic injury cases and 661 attributed to occupational disease. Nationally there were 267,199 accepted lost-time injury claims, with Quebec’s 75,642 the largest provincial count.

Lost-time injury rates fell in every jurisdiction in 2024 relative to the 2021 to 2023 average, with Quebec down 24% and Alberta down 22%. Yet occupational disease fatalities now outnumber traumatic ones by nearly two to one, and the authors note that in most jurisdictions the disease fatality rate has surpassed or is approaching the injury fatality rate. They attribute this to better recognition of long-latency conditions and to policy changes such as presumptive coverage for firefighters. They also caution that the true total of work-related fatalities is likely far higher than compensation statistics capture, and that coverage ranges from an estimated 98% of the workforce in Prince Edward Island to 73% in Nova Scotia. A summary of the report is available from Canadian Occupational Safety.

Traumatic injury is a shrinking share of a caseload increasingly defined by conditions with long latency periods, contested causation and multi-factorial exposure histories. That is a fundamentally harder assessment problem, and it rewards specialist input.

The British Columbia data illustrates the operational end of this. WorkSafeBC accepted 50,914 new short-term disability claims in 2023 and recorded 4.15 million days of work missed. The most common accepted injury types were strains other than back strains at 20,509 and back strains at 10,465, with occupational disease third at 8,123. Where payments were made, the initial payment was issued an average of 19.1 days after the worker first became eligible. Nineteen days of uncertainty at the front end of a claim is a meaningful amount of time for a recovery trajectory to set in one direction or the other.

What the data still won’t tell us

Honest analysis means naming the blind spots.

The Canadian Survey on Disability’s most recent cycle is 2022, which means the prevalence picture is four years old while the absence picture is current.

The plateau in absence incidence is real. The stability it appears to describe is not. Prevalence is rising, its centre of gravity is moving toward mental health and toward younger workers, the exhaustion cliff in public sickness benefits has moved to week 26, and the fatal end of the compensation system is now dominated by disease rather than injury.

Every one of those shifts changes what a well-timed, well-scoped independent assessment needs to look at.