Excess deaths during the COVID-19 pandemic in Alberta, Canada.

Objectives: To determine if there was excess mortality in Alberta, Canada during the coronavirus disease 2019 (COVID-19) pandemic, to confirm if excess mortality affected all age groups equally, and to determine what proportions of excess deaths were directly related to COVID-19 and non-pharmaceutical drug poisoning.
Methods: Weekly all-cause data used to estimate excess mortality were modelled against the pre-pandemic period (January 2015-February 2020). Age-adjusted weekly mortality rates for March 2020 to December 2021 were compared with the preceding 5 years.
Results: From March 2020 to December 2021, there was an 11% excess mortality rate, corresponding to an average of 265 monthly excess deaths (maximum >30%). COVID-19-related deaths (n=3202) accounted for 54.9% of total excess deaths (n=5833) that occurred in the 22-month period. The increase in all-cause excess deaths was proportionately higher, and with significantly greater numbers, in younger age groups. Significant increases in monthly drug poisoning deaths occurred from March 2020 to April 2021, with a total of 1819 deaths. Eight hundred and 25 excess drug poisoning deaths, representing 25.4% of total all-cause excess deaths, occurred, mainly among those aged 25-60 years. Overall, 54.9% of all excess deaths were directly related to COVID-19 and 25.4% were related to drug poisoning. Conclusions: There was a significant increase in all-cause mortality during the COVID-19 pandemic. Although older adults are more likely to die of COVID-19, a massive increase in non-COVID-19-related mortality was observed among younger people. These factors should be considered in public policy decisions on epidemic/pandemic management.

Introduction

Since the emergence of SARS-CoV-2 in December 2019, with high case fatality rates reported in China (3.8-14.6% from January 1 to January 17, 2020), the international public health community reacted promptly by implementing a suite of major non-pharmaceutical interventions (NPIs) (Mi et al. 2020) . In the Western world, these measures were initially accepted as necessary, but over time skepticism emerged (Melnick and Ioannidis. 2020) and some questioned whether drastic measures that led to economic losses, adverse mental health outcomes and infringement on personal liberties were proportional to disease severity (Bowcott. 2020). In some jurisdictions, health services such as elective surgeries were deferred to prepare the health system to manage potential surges in hospitalizations and intensive care unit admissions (CIHI. 2020). Fear of COVID-19 deterred some patients from accessing vital medical services. For example, more than 40% of American adults chose not to access medical care due to concerns about risks of contracting COVID-19 in health care settings (Czeisler et al. 2020). The same effect has been found in the Canadian health system, including Alberta, where overall emergency department visits dropped by 35 % (IRR 0.65, p<0.001) and surgical and medical admissions through the Emergency Department, also declined by 18% (IRR 0.82,  p<0.001) and 14% (IRR 0.86, p<0.001), respectively (Rennert-May et al. 2021).

Although policy makers are subjected to a daily onslaught of massive amounts of data and advice from scientists, modelers, economists, disaster management professionals and legal experts, public health decision making usually relies on case numbers, test positivity rate, clinical severity of disease, hospital admissions, intensive care unit admissions and case fatality rates. Comparing absolute number of deaths or mortality rates across different jurisdictions can be problematic as the numbers can be biased due to differences in case definitions, SARS-CoV2 testing criteria, and reporting of deaths as being “due” to COVID as opposed to “with” COVID. To assess the real impact of the pandemic, all-cause mortality, which compares mortality over a given period to historical baseline, is emerging as a reliable, objective and gold standard metric (Weinberger et al. 2020; Simonsen and Viboud. 2021; Lee et al. 2022; McGrail. 2022). With the exception of a few countries, such as Australia, Singapore, Iceland, and Norway, excess mortality was documented in most jurisdictions impacted by the COVID-19 pandemic (Wang et al. 2022). Although in the beginning it was unclear whether the excess mortality was solely attributable to COVID-19, recent reports are confirming major mortality impacts during the pandemic. For example, based on US data collected from August 2014 to April 2021, Lee et al. reported higher excess mortality for diabetes, heart disease, Alzheimers, cerebrovascular disease, accidents and injuries, drug overdoses and assaults and homicides during the pandemic. (Lee et al. 2022). Likewise, DiGennaro et al, found significant excess drug-overdose mortality during the pandemic (DiGennaro et al. 2021). In Canada, the deaths due to opioid poisonings significantly increased during the pandemic. Increased toxicity of the non-pharmaceutical drug supply in Alberta, particularly related to fentanyl, contributed to the increased number of accidental drug poisoning deaths during 2021 (PHAC. 2022).

In this work, we aim to confirm whether the pandemic was accompanied by excess all-cause mortality in Alberta, Canada, which age groups were most impacted, and attempt to tease out subtle factors driving the mortality trends.

Methods

We acquired death data from the Alberta Vital Statistics database, accessed through Alberta Health Services, Data and Analytics. The Vital Statistics database tracks all of the births and deaths in the province of Alberta, and is maintained by Service Alberta. Population estimates were obtained from the Alberta Population Health Statistics database which contains estimates for the population of Alberta stratified by various attributes including sex, age, and postal code and is derived from the Provincial Registry which tracks registration with the Alberta Health Care Insurance Plan. Deaths related to COVID-19 were obtained from the Communicable Disease Outbreak Management (CDOM) system which is used for contact tracing and follow-up with people having confirmed COVID-19. Deaths were considered to be related to COVID-19 if the cause of death was recorded as being ‘from the disease’ or the disease ‘contributed to the death (secondary cause).’ Additionally, we utilized the publicly available data from Alberta's substance use surveillance system (Alberta Health-A. 2022), to look at the number of deaths associated with substance abuse by age and sex in the same time frame.

All-cause deaths were summarized weekly between 2015 and 2021 and stratified by pre-specified 10-year age groups. The weekly data were used to estimate excess mortality using an over-dispersed Poisson model where the expected mortality is based on the pre-pandemic period (January 2015 to February 2020). Total cumulative excess mortality between March 2020 and December 2021 was estimated as well as the proportion increase relative to the expected mortality, overall and for each age group. All statistical analyses were conducted using SAS version 9.4 (Cary, NC) and R version 4.1.2 (Vienna, Austria) including the excessmort package for estimation of excess mortality (Acosta et al. 2022).

Results

The province of Alberta has a population of ∼4.4M people and an observed COVID-19 case rate of 8411 per 100,000 people in Alberta from March 15, 2020 through Dec 31, 2021 (Alberta Health-B. 2022). As can be seen in Figure 1 , there was statistically significant excess observed deaths compared to expected deaths for 22 months from March 2020, through December 2021. In Alberta, the first COVID-19-associated fatality was reported in early March 2020. From March 2020 to December 2021, we estimated a total of 5833 excess deaths (95% CI 5373 to 6293) which equates to an average of approximately 265 excess deaths per month. The highest excess mortality relative to the expected death rate was in late 2020 and late 2021, with maximum weekly increases of 33.1% (95% CI 27.4% to 28.8%) for the week ending September 24, 2021 and 30.4% (95% CI 25.6% to 35.3%) for the week ending December 25, 2020 (Figure 2 ). There are only three periods where a statistically significant increase in excess mortality was not observed: a) prior to the week ending March 20, 2020, b) between the weeks ending May 8, 2020 to July 10, 2020, and c) between the weeks ending February 12, 2021 to April 9, 2021 (Figure 2). COVID-19 related deaths (n=3202) account for 5.7 % of all deaths since March 2020 (n=56010), but 54.9% of the total excess deaths (n=5833) that occurred over the 22 months. There was a strong correlation between the weekly estimated excess deaths and COVID-19 related deaths (Pearson correlation, r = 0.75).

Figure 1

Cumulative number of covid-19 related deaths and total excess deaths from March 2020 to Dec 2021

Figure 2

Excess Mortality across all ages in From March 2020 to Dec 2021

Our analysis of weekly data stratified by age groups suggest there was no significant excess mortality in the 0 to 9 year age group and generally very little in the younger age groups. The older age groups followed the pattern observed in the overall population with peaks in late 2020 and late 2021, however the 30-39 age group shows a steadily increasing excess mortality throughout the pandemic (Figure 3 ). Alberta data show that most of the COVID-19 fatalities were in the older age groups, notably in those over 80 years of age.

Figure 3

Excess Morality by 10-year Age Groups.

Based on Alberta substance use surveillance system data (Alberta Health-A. 2022) there was an increase in the number of deaths associated with non-pharmaceutical drug poisoning. The surveillance group found there were a total of 1379 and 1818 deaths in 2020 and 2021 respectively, attributed to drug poisoning related to any substance (opioids, methamphetamines, cocaine, alcohol, benzodiazepines, or others). These numbers correspond to an excess of 534 and 946 drug poisoning deaths in 2020 and 2021 respectively when compared to a mean of 825 drug poisoning death in each of the four preceding years. The 1480 excess drug poisoning deaths accounts for 59% of the 2514 non COVID-19 imputed excess deaths and 25.4 % of the total all-cause excess mortality between March 2020 through December 2021 and occurred primarily in those aged 25-60. Of the 1545 non-pharmaceutical opioid drug poisoning deaths in 2021, the highest proportion of deaths occurred in young men between 25-44 years (41.4%, n=639) (Alberta Health-A. 2022). Overall, 80.3 % of all-cause excess mortality can be accounted for by COVID-19 related deaths (54.9 %) and drug poisoning excess deaths (25.4%) (FIGURE 4 ). The remaining 19.7 % of excess deaths are likely due to a constellation of factors including, but not limited to, delay in accessing urgent and critical medical care, suboptimal chronic disease follow-up and potential adverse outcomes associated with delay in some surgeries.

Figure 4

Drug poisoning Deaths by age group and sex from 2016 to 2021

Discussion

Despite both the death rate (75/100,000) and the overall case fatality rate (0.8 %) being lower in Alberta than in more heavily populated Canadian Provinces such as Ontario and Quebec, all-cause excess mortality was higher in Alberta (McGrail. 2022). Our study confirms that the COVID-19 pandemic has been associated with excess mortality in Alberta and permits a deeper exploration of possible causes for this. Importantly, nearly half of the excess deaths in the study period do not appear to be COVID-19-related. Given the robust infectious disease surveillance infrastructure in Alberta, it is unlikely that the 43.1% of excess deaths attributed to non-COVID-19 causes are exclusively due to under-reporting, though it is possible that some COVID-19 deaths may have been attributed to other underlying co-morbidities.

In recent months several papers have been published documenting potential associations between non-Covid- 19 imputed excess mortality (NCIEM) and diabetes, Alzheimer's, cerebrovascular disease (Glei. 2022), heart disease, (Brant et al. 2020) which are possibly due to avoidance or inaccessibility of healthcare (Czeisler et al. 2020). Furthermore, recent data show an increase in homicides (Kegler et al. 2022), suicides (Kaggwa et al. 2022, Orellana and de Souza 2022, Rogalska and Syrkiewicz-Świtała. 2022) and drug poisoning deaths (Faust et al. 2021, Mason et al. 2021, Hawkins and Phan. 2022, Schleihauf and Bowes 2022) in other jurisdictions during the pandemic. Alberta has a robust surveillance system for drug poisoning deaths (Alberta Health. 2022) which was used to verify an increase in the number of drug poisoning deaths from March 2020 to December 2021 (Figure 4). Much of this increase can be attributed to an increased toxicity of the non-pharmaceutical drug supply, especially in the last 6 months of 2021. For example, the Supervised Consumption Service in Calgary, Alberta, reported the average number of daily client visits dropped significantly as public health measures affected site capacity. In Dec 2019, there were an average of 210.9 visits per day, compared to 100 in Dec 2021. The proportion of overdoses per number of time drugs consumed increased from 1.22% in Dec 2019 to 7.95% in 2021, reflecting this increased drug supply toxicity. (Alberta Health Services-B 2022)

There are limitations to what can be surmised from the available data. Overall mortality rates are highly accurate but monthly and yearly fluctuations can impact any precise estimates of excess deaths. Importantly, discerning between deaths attributable to acute COVID-19 infection and non-COVID-19 related deaths requires that testing be available, broadly applied, and accurately reported. This example can be clearly seen in Eastern Europe and Latin America in the first wave of COVID-19 where high rates of excess mortality were observed prior to the widespread availability of testing (Ritchie et al. 2021), resulting in a large gap between COVID-19 attributable deaths and the overall increased number of observed deaths over expected. Access to testing was readily available at no cost and with minimal eligibility restrictions to the individual as part of Alberta's publicly-funded healthcare system. With a testing rate of 111 059 per 100 000 population, Alberta had one of the highest per capita testing rates in the country (PHAC 2021). However, there may well have been differences in testing rates across age/sex strata (Sundaram et al. 2021). Implementation of public health measures to control the spread of COVID-19 may lessen mortality below expected levels (and underestimate excess mortality) through reduction in other communicable diseases (such as influenza), work-related accidents, and road traffic fatalities, or may increase mortality, for example, due to substance use and mental health related deaths (Moriarty et al. 2021). A deeper exploration of cause of death and contributing factors would help clarify how these factors may contribute to changes in observed mortality. Though standardized definitions were applied throughout, some misclassification of either COVID-19 or non-COVID-19 deaths is still possible. It is unlikely that COVID-related deaths were missed prior to the first recognized case in the province on March 5, 2020 (Kanji et al. 2021).

Conclusion

This cross-sectional study confirms earlier findings in other jurisdictions that the COVID-19 pandemic led to substantial excess mortality, and highlights the differences in effect across age groups. We found that there were disproportionately higher relative death rates among young Albertans between the ages 20 -59, which were not linked directly to a proven COVID-19 infection. Analysis of official acute drug poisoning death data during the pandemic strongly suggests that the latter made a substantial contribution to the overall non-COVID-19 deaths during our study period. The contributions and interrelations of medical, social and economic factors to explain the high number of non-COVID-19 related excess deaths are important to understand for making policy decisions during future pandemics. This work highlights the importance of looking beyond case counts and case-fatality rates in the understanding the effects of a pandemic in society.

Acknowledgements

We would like to acknowledge Zoe Hsu, from Provincial Research Data Services at Alberta Health Services and Alberta SPOR Data and Research Services Team for assistance with data extraction and linkage for the study.

Ethics

Ethics approval for the current study was obtained from the Health Research Ethics Board of the University of Alberta, Canada (Pro00112035).

R EFERENCES

Acosta, R.J. and R.A. Irizarry, (2022). “A Flexible Statistical Framework for Estimating Excess Mortality.” Epidemiology 33(3): 346-353.

Alberta Health-A. 2022. Alberta Substance Use Surveillance System. Acute substance related deaths by age and sex. Available from: https://healthanalytics.alberta.ca/SASVisualAnalytics/?reportUri=%2Freports%2Freports%2F1bbb695d-14b1-4346-b66e-d401a40f53e6§ionIndex=0&sso_guest=true&reportViewOnly=true&reportContextBar=false&sas-welcome=false. Accessed June 1 2022

Alberta Health-B. 2022. COVID-19 Alberta Statistics; Available from: https://www.alberta.ca/stats/covid-19-alberta-statistics.htm. Accessed on June 1, 2022.

Alberta Health Services-A. 2022. Safeworks. Calgary's Supervised Consumption Service at the Sheldon M. Chumir Health Centre. Available at https://www.albertahealthservices.ca/info/Page15433.aspx. Accessed on June 1 2022.

Alberta Health Services-B 2022. Safeworks Monthly Report – December 2021. Supervised Consumption Services. Available at https://www.albertahealthservices.ca/assets/info/amh/if-amh-sup-con-chumir-2021-12.pdf. Accessed on June 1 2022.

Bowcott Owen. Covid measures will be seen as 'monument of collective hysteria and folly' says ex-judge. The Guardian. 2020: UK. Available from: https://www.theguardian.com/law/2020/oct/27/covid-measures-will-be-seen-as-monument-of-collective-hysteria-and-folly-says-ex-judge. Accessed on July 7, 2021.

Brant, L. C. C., B. R. Nascimento, R. A. Teixeira, M. A. C. Q. Lopes, D. C. Malta, G. M. M. Oliveira and A. L. P. Ribeiro (2020). "Excess of cardiovascular deaths during the COVID-19 pandemic in Brazilian capital cities." Heart 106(24): 1898-1905.

Czeisler, M., K. Marynak, K. E. N. Clarke, Z. Salah, I. Shakya, J. M. Thierry, N. Ali, H. McMillan, J. F. Wiley, M. D. Weaver, C. A. Czeisler, S. M. W. Rajaratnam and M. E. Howard (2020). "Delay or Avoidance of Medical Care Because of COVID-19-Related Concerns - United States, June 2020." MMWR Morb Mortal Wkly Rep 69(36): 1250-1257.

DiGennaro, C., G.-G. P. Garcia, E. J. Stringfellow, S. Wakeman and M. S. Jalali (2021). "Changes in characteristics of drug overdose death trends during the COVID-19 pandemic." International Journal of Drug Policy 98.

Faust, J. S., H. M. Krumholz, C. Du, K. D. Mayes, Z. Lin, C. Gilman and R. P. Walensky (2021). "All-Cause Excess Mortality and COVID-19-Related Mortality Among US Adults Aged 25-44 Years, March-July 2020." Jama 325(8): 785-787.

Glei, D. A. (2022). "The US Midlife Mortality Crisis Continues: Excess Cause-Specific Mortality During 2020." American journal of epidemiology.

Hawkins, D. and A. T. Phan (2022). "Changes in drug poisoning mortality before and after the COVID-19 pandemic by occupation in Massachusetts." Am J Ind Med.

Kaggwa, M. M., G. Z. Rukundo, E. K. Wakida, S. Maling, B. M. Sserumaga, L. M. Atim and C. Obua (2022). "Suicide and Suicide Attempts Among Patients Attending Primary Health Care Facilities in Uganda: A Medical Records Review." Risk Manag Healthc Policy 15: 703-711.

Kanji JN, Diggle M, Bulman DE, Hume S, Taylor S, Kelln R et al. Retrospective testing of respiratory specimens for COVID-19 to assess for earlier SARS-CoV-2 infections in Alberta, Canada. Journal of the Association of Medical Microbiology and Infectious Disease Canada. 2021 March (6). pp. 10-15

Kegler, S. R., T. R. Simon, M. L. Zwald, M. S. Chen, J. A. Mercy, C. M. Jones, M. C. Mercado-Crespo, J. M. Blair, D. M. Stone, P. G. Ottley and J. Dills (2022). "Vital Signs: Changes in Firearm Homicide and Suicide Rates - United States, 2019-2020." MMWR Morb Mortal Wkly Rep 71(19): 656-663.

Lee, W.-E., S. W. Park, D. M. Weinberger, D. Olson, L. Simonsen, B. T. Grenfell and C. Viboud (2022). "Direct and indirect mortality impacts of the COVID-19 pandemic in the US, March 2020-April 2021." medRxiv: the preprint server for health sciences.

Mason, M., P. Arukumar and J. Feinglass (2021). "The Pandemic Stay-at-Home Order and Opioid-Involved Overdose Fatalities." Jama 325(24): 2495-2496.

Melnick E.R. and J.P.A. Ioannidis. Should governments continue lockdown to slow the spread of COVID-19? Br Med J. 2020. 369: p. m1924.

Mi YN, Huang TT, Zhang JX, Qin Q, Gong YX, Liu SY, et al. Estimating the instant case fatality rate of COVID-19 in China. Int J Infect Dis. 2020 Aug 97:1-6.

Moriarty T, Boczula AE, Thind EK, Loreto N, McElhaney JE. Excess All-Cause Mortality During the COVID-19 Epidemic in Canada. Royal Society of Canada. June 2021

Orellana, J. D. Y. and M. L. P. de Souza (2022). "Excess suicides in Brazil: Inequalities according to age groups and regions during the COVID-19 pandemic." Int J Soc Psychiatry: 207640221097826.

PHAC. 2021. Public Health Agency of Canada; March 2022. COVID-19 in Canada. Available at https://www.canada.ca/content/dam/phac-aspc/documents/services/diseases/2019-novel-coronavirus-infection/surv-covid19-weekly-epi-update-20210813-en.pdf Accessed on June 1 2022.

PHAC. 2022. Special Advisory Committee on the Epidemic of Opioid Overdoses. Opioid- and Stimulant-related Harms in Canada. Ottawa: Public Health Agency of Canada; March 2022.

Rennert-May, E., J. Leal, N. X. Thanh, E. Lang, S. Dowling, B. Manns, T. Wasylak and P. E. Ronksley (2021). "The impact of COVID-19 on hospital admissions and emergency department visits: A population-based study." PLoS One 16(6): e0252441.

Ritchie H, Esteban Ortiz-Ospina, Diana Beltekian, Edouard Mathieu, Joe Hasell, Bobbie Macdonald, Charlie Giattino, Cameron Appel, Lucas Rodés-Guirao and Max Roser (2020) - "Coronavirus Pandemic (COVID-19)". Published online at OurWorldInData.org. Retrieved from: 'https://ourworldindata.org/coronavirus' [Online Resource]. Accessed on July 7 2021.

Rogalska, A. and M. Syrkiewicz-Świtała (2022). "COVID-19 and Mortality, Depression, and Suicide in the Polish Population." Front Public Health 10: 854028.

Schleihauf, E. and M. J. Bowes (2022). "Suicide and drug toxicity mortality in the first year of the COVID-19 pandemic: use of medical examiner data for public health in Nova Scotia." Health Promot Chronic Dis Prev Can 42(2): 60-67.

Simonsen, L. and C. Viboud (2021). "Mortality: A comprehensive look at the COVID-19 pandemic death toll." Elife 10.

Sundaram, M. E., A. Calzavara, S. Mishra, R. Kustra, A. K. Chan, M. A. Hamilton, M. Djebli, L. C. Rosella, T. Watson, H. Chen, B. Chen, S. D. Baral and J. C. Kwong (2021). "Individual and social determinants of SARS-CoV-2 testing and positivity in Ontario, Canada: a population-wide study." Canadian Medical Association Journal 193(20): E723-E734.

Wang et al. (2022). "Estimating excess mortality due to the COVID-19 pandemic: a systematic analysis of COVID-19-related mortality, 2020-21." Lancet 399(10334): 1513-1536.

Weinberger, D. M., T. Cohen, F. W. Crawford, F. Mostashari, D. Olson, V. E. Pitzer, N. G. Reich, M. Russi, L. Simonsen, A. Watkins and C. Viboud (2020). "Estimating the early death toll of COVID-19 in the United States." medRxiv: the preprint server for health sciences.