The COVID‑19 pandemic had a tremendous impact on many facets of human life. Although it affected countries around the world, these impacts can be difficult to compare since the data available differs from country to country. One way to assess the data is by considering excess mortality.

Excess mortality, or excess deaths, is the number of deaths exceeding expected levels in a specific period. Excess mortality provides valuable insight into the broader impact of the Covid pandemic because it captures the direct impact of the virus on deaths, as well as potential indirect impacts, such as overwhelmed healthcare systems, delayed medical treatments, and other effects. Excess mortality also captures any effects that may reduce mortality, such as reduced deaths associated with motor vehicle accidents during periods where stay-at-home orders were in place or reduced deaths due to decreased circulation of influenza viruses. Finally, excess mortality includes Covid deaths that were not captured by official Covid death statistics.

In this article we focus on the impact of the Covid pandemic on mortality in Canada using the measure of excess mortality. More specifically, we look at excess mortality relative to the expected number of deaths within the general and insured populations of Canada, we identify areas for further monitoring, and we discuss potential causes of excess mortality for lives below age 45.

Population Excess Mortality

Statistics Canada introduced weekly excess death tracking as part of its Covid monitoring.¹ Excess mortality for the Canadian population was recorded between January 2020 and August 2023, at which point, Statistics Canada stopped the official tracking.

Statistics Canada defined excess mortality as the difference between the number of observed deaths and the number of expected deaths over a certain period of time. This definition may have underestimated or overestimated excess mortality during the Covid pandemic because it is difficult to collect actual death counts across provinces and territories in a timely manner. As such, the actual deaths are based on provisional death counts, which were adjusted for reporting delays where possible. Additionally, this approach relies on developing an expected basis had there been no pandemic. Statistics Canada used recent trends to estimate expected deaths. We cannot know what mortality levels would have been had there been no pandemic, so the expected basis could also underestimate or overestimate excess mortality.²

Figure 1 – Excess mortality in Canada as a % of expected deaths (13‑week smoothing)

Early in the pandemic, excess mortality in Canada spiked in successive waves associated with the emergence of different variants, however, excess mortality remained elevated from April 2021 onwards. Over time, direct Covid deaths don’t fully explain the total excess mortality, which may partially be due to reduced Covid testing, but the elevated excess mortality also indicates other contributing factors to increased excess mortality.⁶

Excess mortality remains high in the three years from the onset of the pandemic, with seasonal highs in the winter (Figure 1, Table 1). While relative excess mortality appears to have dropped in 2023 (2.4% for the year), it is important to note that seasonal deaths for Covid- and flu-related deaths at the end of 2023 are excluded since the data for 2023 ends in August.

Table 1 – Excess Mortality in the Canadian Population

Population Excess Mortality by Age

Relative excess mortality varied across age groups (Figure 2), and while the number of Covid-related fatalities is higher at older ages,⁸ the proportional impact that these additional deaths have on mortality is larger at younger ages. This suggests that there are factors beyond just the fatality of the Covid virus driving excess mortality. We will explore some of these factors later in this article.

Figure 2 – Excess mortality as a percentage of expected deaths in the Canadian population by age

Relative excess mortality increased substantially for younger ages from 2020 to 2021 and increased notably for all ages in 2022. The overall relative excess mortality is more heavily weighted to older ages, which reflects the fact that Canada’s general population is aging. The relative excess mortality appears to drop for 2023, but 2023 data ends in August and thus excludes months where seasonal deaths related to Covid and the flu are more likely to be recorded. As such, the 2.4% relative excess mortality for all ages in 2023 is likely understated.

Figure 3 – Excess mortality by age in Canada as a percentage of expected deaths (13‑week smoothing)

Figure 3 shows that relative excess mortality for younger ages is consistently trending higher and has not experienced the same spikes in the winter months as the other age groups. The differences in data for younger ages is unexpected and leads to concern that additional factors may influence excess mortality in the younger cohort.

Using Statistics Canada weekly provisional death counts into 2024, we can estimate the 2023 excess mortality for the Canadian population using recent trends. While we are unable to replicate Statistics Canada’s reporting exactly (some accuracy is lost as ages are grouped), this approach gives us a good indication of relative excess mortality for the completed 2023 calendar year. The estimated relative excess mortality in 2023 is 4.8% overall, which supports the expectation that 2.4% relative excess mortality (January-19 August 2023) from Statistics Canada weekly Covid monitoring is understated (Table 2).

Table 2 – Estimated excess mortality as a percentage of expected deaths in 2023 by age

Insured Population Excess Mortality by Age

Using the raw data from the Canadian Institute of Actuaries’ Canadian Individual Life Experience report¹³ (June 2024), we compared the industry experience against an industry expected mortality basis (CIA2014 base mortality tables) for insured lives, adjusted for mortality trends. We calculated the relative excess mortality for the insured population by comparing the actual-to-expected (AvE) ratios for the 2020‑2021 and 2021‑2022 policy years against five years of historical AvE ratios.

The following analysis considers the impact by attained age for the first 22 policy years. The pattern of results by age group mirrors what we see in the population, with higher relative excess mortality at younger ages, which generally declines with age (Table 3). Overall, the relative excess mortality is higher for the insured group, however this is driven by differences in age distributions between the insured and general populations. The insured population is weighted to the younger age groups, where there is a higher proportion of relative excess mortality. By age group, the relative excess mortality in the insured group is comparable to the Canadian population, suggesting that the insured group is not insulated from the broader trend.

Note that the insured relative excess mortality is on a policy year basis, meaning that the recorded death could fall in one of two calendar years. For example, for a policy issued on 1 July, the policy year is 1 July 2020, to 30 June 2021. The population relative excess mortality is on a calendar year basis, so when comparing the two populations, it is important to consider the overlapping time periods.

Table 3 – Excess mortality as a percentage of expected deaths for insured population by age

It will be important to monitor insurance trends as these become available for 2022 to 2023 and beyond to assess relative excess mortality in the insured population compared to the general population.

Excess Mortality in Younger Age Groups

Figure 4 shows a significant increase in opioid-related deaths since 2020, with an increase in mortality rates of between 50% and 133% compared to pre-pandemic averages. The opioid crisis impacts younger ages (Table 4), with approximately 70% of the opioid-related deaths occurring in ages under 50, and 50% of the total opioid-related deaths occurring in ages 30 to 49. While the rise in opioid-related deaths hasn’t impacted any age group more than another since 2016, the increase in deaths due to the opioid crisis is greater for younger individuals as they typically have a better expected mortality rate compared to older age groups.

Figure 4 – Total apparent opioid/stimulant toxicity deaths in Canada

Table 4 – Total apparent opioid/stimulant toxicity deaths in Canada by age

The opioid crisis in Canada is a severe public health issue predating and exacerbated by the Covid pandemic. Some factors that may have contributed to the above increase include the following:

• Increasing overdoses – Potentially due to increased isolation, stress, and disruption of support services, more overdoses have been recorded during the pandemic.^19,20
• Worsening mental health – Mental health deterioration due to increased isolation, anxiety, and access to care issues.²¹
• More volatile and unregulated drug supply including polysubstance use – Fentanyl, which is 50 to 100 times more potent than morphine and cheaper, is becoming common in the drug supply, and accounts for the majority of opioid toxicity deaths in recent years.^22,23

Future Outlook

Figure 5 shows that age-standardized mortality rates in 2024 remain elevated compared to pre-pandemic (2019) rates, averaging about 4% higher. It is essential that we continue to monitor the evolving excess mortality trends to appropriately allow for additional mortality in cashflow projections.

Figure 5 – Provisional weekly age-standardized mortality rates as a percentage of the 2019 age-standardized mortality rates

Other Considerations

• Uncertainty in estimating excess mortality – Excess mortality relies on the expected mortality basis and assumptions about mortality improvement. It is difficult to know for certain what mortality experience would have been in the absence of the pandemic and therefore there will always be uncertainty about the true number of excess deaths due to Covid.
• Regional variations in mortality – Mortality rates and excess mortality can vary across different provinces and territories.²⁵ This is important to consider when applying pricing adjustments or conducting experience analyses.
• Variations by gender, age, and socioeconomic groups – Mortality rates and excess mortality vary by gender and socioeconomic group. When considering opioid-related mortality, analyses of narrower age buckets, gender, and socioeconomic groups would be useful.
• Healthcare delays – In 2023, the longest delay between a referral from a general practitioner and receipt of treatment was recorded.²⁶ This illustrates an ongoing struggle within the healthcare system, which no doubt has a part to play in the continued excess mortality.
• Excess mortality and projections – Excess mortality experience must be factored into the expected basis when it indicates a significant shift in long-term mortality trends, impacting the actuarial assumptions and projections for an insured population.

Conclusion

The analysis of excess mortality offers valuable insights into the Covid pandemic. Excess mortality was originally attributed to the Covid pandemic; however, it has persisted beyond the height of the Covid pandemic, particularly in younger age groups. Potential contributing factors include indirect Covid effects, exacerbation of pre-pandemic trends, and healthcare resource and access issues.

As shown in this article, general and insured populations are not insulated from the recent excess mortality trends. The growing drug toxicity crisis is likely relevant to excess mortality trends in younger ages in the general population, and it will be important to monitor industry data as it becomes available to compare trends going forward.

To assess how quickly mortality in Canada can return to a pre-pandemic state, it will also be essential to monitor all causes of death and broader public health indicators. What the future will look like and how long it will take to reach it remains unknown.