SARS-CoV-2 seroprevalence in Nova Scotia blood donors

Abstract

Background:

SARS-CoV-2 seroprevalence monitors cumulative infection rates irrespective of case testing protocols. We aimed to describe Nova Scotia blood donor seroprevalence in relation to public health policy and reported data over the course of the COVID-19 pandemic (May 2020 to August 2022).

Methods:

Monthly random Nova Scotia blood donation samples (24,258 in total) were tested for SARS-CoV-2 infection antibodies (anti-nucleocapsid) from May 2020 to August 2022, and vaccination antibodies (anti-spike) from January 2021 to August 2022. Multivariable logistic regression for infection antibodies and vaccination antibodies separately with month, age, sex, and racialization identified independent predictors. The provincial nucleic acid amplification test (NAAT)-positive case rate over the pandemic was calculated from publicly available data.

Results:

Anti-N seroprevalence was 3.8% in January 2022, increasing to 50.8% in August 2022. The general population COVID-19 case rate was 3.5% in January 2022, increasing to 12.5% in August 2022. The percentage of NAAT-positive samples in public health laboratories increased from 1% in November 2021 to a peak of 30.7% in April 2022 with decreasing numbers of tests performed. Higher proportions of younger donors as well as Black, Indigenous, and racialized blood donors were more likely to have infection antibodies (p < 0.01). Vaccination antibodies increased to 100% over 2021, initially in older donors (60+ years), and followed by progressively younger age groups.

Conclusions:

SARS-CoV-2 infection rates were relatively low in Nova Scotia until the more contagious Omicron variant dominated, after which about half of Nova Scotia donors had been infected despite most adults being vaccinated (although severity was much lower in vaccinated individuals). Most COVID-19 cases were detected by NAAT until Omicron arrived. When NAAT testing priorities focused on high-risk individuals, infection rates were better reflected by seroprevalence.

Infection with the SARS-CoV-2 virus causes COVID-19. The COVID-19 pandemic was declared in March 2020. Public health monitored community infections using data from positive tests. Testing for antibodies that the body makes in response to infection, called seroprevalence, is another way to monitor infection rates. We measured antibodies from both infection and vaccination in Nova Scotia blood donors from May 2020 to August 2022 (24,258 samples). As vaccination was first rolled out over the winter/spring of 2021, the percentage of vaccination antibody positive samples increased to nearly 100% in older donors (aged 60 +), then progressively younger age groups consistent with prioritizing older people first. The seroprevalence of infection antibodies was low initially, reaching 3.2% by January 2022. At that time a new variant of the SARS-CoV-2 virus called Omicron was infecting people. Omicron was more infectious than previous SARS-CoV-2 variants, but symptoms were less severe, partly due to properties of Omicron, and partly because most people had vaccination antibodies by that time. Nevertheless, by August 2022 about half (50.8%) of blood donors had infection antibodies showing that they had been infected with SARS-CoV-2 at some point. The percentage of people in the general population who had a positive SARS-CoV-2 laboratory test was only 12.5%. This was because testing in public health laboratories had been scaled back when Omicron arrived to focus on high-risk people while many people were self-testing at home using rapid antigen tests. Higher proportions of younger donors had infection antibodies, and Black, Indigenous, and racialized donors were also more likely to have infection antibodies. In summary, SARS-CoV-2 seroprevalence in blood donors shows that infections were low in Nova Scotia until 2022 when the Omicron variant dominated after which about half were infected. Most Nova Scotia blood donors had vaccination antibodies before Omicron arrived.

Introduction

In Canada during the COVID-19 pandemic there have been seven distinct waves of SARS-CoV-2 infection as of June 2023 fuelled by new variants of concern and subvariants. However, Atlantic provinces, such as Nova Scotia, had comparatively fewer SARS-CoV-2 cases than the rest of the country prior to the emergence of the more infectious Omicron variant in late 2021 (1).

Throughout the pandemic, Nova Scotia had liberal access to testing. Early on, nucleic acid amplification tests (NAAT) such as RT-PCR were available for symptomatic individuals (2). Testing capacity was later enhanced using both NAAT and rapid tests to include contacts, entrants to the province, workers, school attendees, and asymptomatic individuals. Reported cases were a key metric of burden of disease, but despite broad access to testing, can underestimate the burden since not all infected people will present for testing. By January 2022 as the Omicron variant became dominant, infections increased dramatically but often with milder symptoms than previous variants, and access to NAAT testing was scaled back to hospital admissions and other high risk congregate settings, health care workers, and for those at risk of severe outcomes (3). While access to testing was greater than in other jurisdictions, the streamlining of testing eligibility including the elimination of most asymptomatic testing would further underestimate the true burden of infection.

Blood donor SARS-CoV-2 seroprevalence studies have been instrumental in informing public health policy around the world (4–8), from the early pandemic infection seroprevalence, to the mid-pandemic vaccination seroprevalence, and then infection seroprevalence in vaccinated populations. Blood donors are healthy adults and not fully representative of the general population but provide cost-effective and efficient seroprevalence sampling. In collaboration with the Canadian Federal Government COVID-19 Immunity Task Force (9), Canadian Blood Services has monitored SARS-CoV-2 seroprevalence in Canadians using monthly samples since May 2020. Canadian Blood Services collects blood donations in all provinces except Quebec. Although in Nova Scotia about 60% of donations are collected in Halifax, the remainder come from rural regions as far south as Yarmouth (380 km), Barrington Passage (300 km), and Saulnierville (330 km), as far north as Amherst (180 km) and as far east as Antigonish (180 km) and Port Hawkesbury (240 km), thus covering most of the province. SARS-CoV-2 seroprevalence has been reported in the literature from some Canadian provincial studies (10–13), but not for Nova Scotia. One national general population serosurveys reported Nova Scotia seroprevalence at one time point (14,15) but most (16–19) combined results of four Atlantic provinces, as have previous reports from our study (20–22). A recent publication modelled data from multiple studies, of which the Canadian Blood Services data was the largest contributer, and reported results for all provinces including Nova Scotia (23). This study evaluates both infection seroprevalence before vaccine deployment and infection and vaccination seroprevalence thereafter from May 2020 to August 2022 in Nova Scotia. There is emphasis on the relationship with public health policy and public health reported data.

Methods

Study design and population

At Canadian Blood Services all potential donors must be at least 17 years of age, afebrile, and must answer screening questions to ensure they are in good health and not at risk of blood transmissible infections. Additional questions to reduce risk of SARS-CoV-2 infection for donors and staff at the collection site were asked during the pandemic. All donors were deferred from donating blood for 2 weeks after contact with someone who was infected, or for 2 weeks after they had an infection (3-week deferral if hospitalized). The deferrals were applied in all Canadian Blood Services locations in Canada.

Canadian Blood Services collects blood donations and distributes blood products for the province of Nova Scotia. Blood donations are collected at a fixed site in Halifax and mobile sites in most towns at variable intervals. An extra EDTA blood sample is routinely collected from all donors at the time of donation in case any additional testing is required to qualify the blood product. About 80% of these samples are not used and were; therefore, available for SARS-CoV-2 serologic testing. Sample selection evolved over time. Samples were included from every week for the time periods from May 2020 to July 2020, and from January 2022 to August 2022. Samples were not tested in August and September in 2020 nor in February 2021. To reduce study costs, from October 2020 to December 2021, samples were selected from approximately the last 2 weeks of every month (see Appendix Figure A1). A random sample was used except from June 2021 to December 2021, where samples were stratified into age groups by region before randomly selected to reduce the total monthly sample size while maintaining a sample representative of important population demographics (ie, age and region). Donors could donate multiple times within this period with a chance of being selected for the study, thus there were some donors who had more than one sample tested. Information about the study was available on the Canadian Blood Services website. Donors read a pamphlet about donation which includes information about routine testing and provide consent for their donation. There was no specific consent for the study testing. Donors were not informed of their results because the assay was recently developed and clinical implications of the result were unknown. This study was approved by the Canadian Blood Services Research Ethics Board (REB Number 2020.028).

Serologic testing

Plasma samples were aliquoted and stored at ≤−20°C. All testing was conducted at the Canadian Blood Services COVID-19 serology laboratory in Ottawa, Ontario. All testing occurred monthly.

Infection seroprevalence testing: Samples collected from May 2020 to December 2020 were tested using the Abbott Architect SARS-CoV-2 nucleocapsid IgG assay (24). Manufacturer estimates for the Abbott anti-N assay at a signal/cut-off ratio of 1.4 were a sensitivity of 92.7% and specificity of 99.9% (24). From January 2021 onwards each sample was tested using the Roche Elecsys anti-SARS-CoV-2 qualitative immunoassay (Roche Diagnostics International Ltd., Rotkreuz, Switzerland) (25) which measures total antibodies (including IgA, IgM, and IgG) to SARS-CoV-2 recombinant protein, nucleocapsid antigen (anti-N). At a signal cut off index (COI) of ≥1.0, the anti-N assay was assumed to have a sensitivity of 99.5% and specificity of 99.8% (26). A positive anti-N result was considered evidence of prior SARS-CoV-2 infection. All assays were approved for testing plasma (24–26).

Vaccination seroprevalence testing: From January 2021 onwards each sample was also tested using the Roche Elecsys Anti-SARS-CoV-2 spike semi-quantitative immunoassay (Roche Diagnostics International Ltd., Rotkreuz, Switzerland) (22) which measures total antibodies (including IgA, IgM, and IgG) to the SARS-CoV-2 S1/receptor-binding domain (RBD) of the spike protein (anti-S). As of September 2021, all samples were tested for anti-S at a dilution of 1:400, and if negative were re-tested undiluted. Prior to that they were tested undiluted and at a 1:10 dilution if they exceeded the maximum concentration. At a concentration of ≥0.8 U/mL, the anti-S assay was assumed to have a sensitivity of 98.8% and specificity of 99.97% (22). A positive anti-S result was considered evidence of prior SARS-CoV-2 infection or vaccination.

Data management and statistical analysis

The following variables were extracted from the Canadian Blood Services donor database: donation date, Forward Sortation Area (FSA) from the residential postal code, sex, age, and self-reported race-ethnicity. Donors self-identified as either White, South Asian, Asian (East and other), Indigenous, Arabic, Black, Latin-American, or Other. Because the proportions in various non-White race-ethnicities were small, race-ethnicities were regrouped a priori as either White (the majority of donors) or Black, Indigenous, and racialized groups. Socioeconomic status was estimated by the Pampalon Material and Social Deprivation Indices (MSDI) (27,28). Material deprivation is associated with insecure job situation, insufficient income, and low education, while social deprivation refers to a fragile social network, characterized by living alone, being a single parent, or being separated, divorced, or widowed. MSDI was derived from the 2016 Statistics Canada census, aggregated from postal codes to the dissemination area (DA) level (the smallest geographic unit available in the Canadian census, considering 400–700 persons) and were categorized into quintiles; from least deprived (quintile 1) to most deprived (quintile 5) areas. Donors age groups were categorized as: 17–24, 25–39, 40–59, and 60+ years old.

Because case rates were relatively low in Nova Scotia prior to the emergence of the Omicron variant (2020 and 2021), seroprevalence was presented graphically for the full study period, but subsequent analyses limited to the period January 2022 to August 2022.

The number of NAAT-positive COVID-19 cases in the province (ie, confirmed cases) and proportion vaccinated (ie, vaccine coverage) were accessed from publicly available datasets (29,30). Total population counts were accessed from national census data (Catalogue # 82-402-x-2018001). The cumulative percentage of the general population SARS-CoV-2 NAAT positive for 2021 was the cumulative number of cases divided by the number of people for each month. The cumulative percentage of the general population vaccinated was the cumulative number vaccinated divided by the number of people for each month. Key interventions and dates were obtained from on-line records (31).

For univariate comparisons, data were weighted using iterative proportional fitting for the donor’s FSA, age group, and sex to make inference to the general population based on Statistics Canada data (catalogue # 98-400-X2016008). For FSAs with few donors, several adjacent FSAs were combined to include at least 500 donors. The weighted data were adjusted for sensitivity and specificity of the assay as reported by the manufacturer using the Rogan-Gladen Eq. (32). The seroprevalence was calculated as the number of positive samples divided by all samples tested. The Exact method was used to estimate 95% confidence intervals. SARS-CoV-2 seroprevalence was stratified by region, sex, age group, Black, Indigenous and Racialization, and MSDI.

Logistic regression models were constructed for each geographic region with the unique donor identifier to identify all repeated measures in a generalized estimating equation (GEE). Models were constructed to predict anti-N seroprevalence for the period of January 2022 to August 2022 because most infections occurred in this period, and to predict anti-S seroprevalence for the period of January 2021 to December 2021 when the first two doses of vaccine (the primary series) were rolled out. Models were first constructed for each variable to assess univariate predictors of seropositivity. Then a regression model was built which included all variables. Age and sex were retained in all model iterations. Missing values for particular variables were included as a ‘missing’ category.

All analyses were conducted using SAS (version 9.4, Cary, NC, USA) and STATA/MP 17 (Statacorp. 2021. College Station, TX, USA).

Results

There were 24,258 donations from 13,564 donors tested in total in Nova Scotia during the study period, with 11,057 being from January to August 2022. Figure 1 shows nucleocapsid (anti-N) and spike (anti-S) antibody seroprevalence from May 2020 to August 2022. Anti-N seroprevalence was very low over 2020 and 2021, and increased from about 0.5% in December 2021 to 50.8% in August 2022. Anti-S was not measured until January 2021. Anti-S seroprevalence increased from January (0.6%) and March (2.5%) to August 2021 (99.2%) and was 100% by the end of 2021 through to August 2022.

Figure 1:

SARS-CoV-2 anti-N and anti-S seroprevalence in blood donors from May 2020 to August 2022

Note: The Atlantic Bubble was an agreement between Atlantic provinces to permit inter-provincial travel of residents

Vaccination seroprevalence: Because anti-S seroprevalence increased over 2021, and due to the change in serology tests from an IgG to total antibody test, this period was analyzed separately. From January 2021 to August 2021 anti-spike increased initially in the oldest age groups, followed by younger age groups (Figure 2). Primary vaccine series coverage in adults also increased over this period (Figure 2). In the logistic regression model anti-S positivity increased over 2021. While all donors were anti-S positive by the end of 2021, the odds ratios were progressively higher by age group (Table 1). Neither Black Indigenous and Racialization or social deprivation index were significant predictors, and while material deprivation index levels varied, no difference between those in the least deprived level and the most deprived level was observed (Table 1). Over the fall of 2021 to August 2022 the concentration of anti-spike was very high median 3,769 U/mL (IQ range 2,100, 6,245) in September 2021, and peaking at a median of 22,136 U/mL (IQ range 11,049, 39,977) in February 2022 (see Figure 3). There was a gradual decrease in median anti-S concentration after both of those time periods.

Figure 2:

SARS-CoV-2 anti-S seroprevalence in blood donors in 2021, broken down by sex (panel A) and age group (panel B) and primary vaccine coverage in the adult general population by dose (panel C)

Table 1:

Output of multiple logistic regression model with anti-S positivity as the outcome (January to December 2021)

	Odds ratio (CIs)	(95% CI)
Sex
Female	Referent
Male	0.61	0.50 to 0.74
Age group
17–24	Referent
25–39	1.58	1.08 to 2.31
40–59	2.62	1.81 to 3.79
60+	6.52	4.38 to 9.69
Race
White	Referent
Racialized	0.99	0.71 to 1.40
Material deprivation index
1 (least deprived)	Referent
2	1.08	0.77 to 1.50
3	0.71	0.51 to 0.98
4	0.71	0.52 to 0.96
5 (most deprived)	0.77	0.55 to 1.07
Social deprivation index
1 (least deprived)	Referent
2	0.88	0.61 to 1.28
3	0.78	0.54 to 1.12
4	0.94	0.65 to 1.37
5 (most deprived)	0.80	0.55 to 1.18
Month of donation
January	Referent
March	3.13	1.59 to 6.19
April	15.85	9.13 to 27.52
May	101.77	59.35 to 174.53
June	>1000	609.21 to >1000
July	>1000	>1000 to >1000
August	>1000	>1000 to >1000
September	>1000	>1000 to >1000
October	>1000	>1000 to >1000
November	>1000	>1000 to >1000
December	>1000	>1000 to >1000

Figure 3:

Violin plots of SARS-CoV-2 anti-S concentration in blood donors from September 2021 to August 2022

Infection seroprevalence: Because anti-N seroprevalence was very low until the end of 2021, but increased in 2022 the January to August 2022 period was analyzed separately. Figure 4 shows the anti-N seroprevalence from January 2022 to August 2022 by age groups and Black, Indigenous, and Racialization. Figure A2 shows the breakdown of racial/ethnic groups. The logistic regression model showed that the highest anti-N seroprevalence was observed in the youngest age group (17–24-year-olds) followed by 25–39-year-olds, and 40–50-year-olds with the lowest seroprevalence in the oldest donors (Table 2). Black, Indigenous and racialized donors had higher seroprevalence than White donors after taking into account other variables in the model; there was no impact of living in a socially deprived neighbourhood, while material deprivation index levels varied, there was no difference between those in the least deprived level and the most deprived levels (Table 2).

Figure 4:

SARS-CoV-2 anti-N seroprevalence in blood donors from January 2022 to August 2022, broken down by racialization (panel A) and age group (panel B)

Table 2:

Output of multiple logistic regression model with anti-N positivity as the outcome (January to August 2022)

	Odds ratio (CIs)	(95% CI)
Sex
Female	Referent
Male	0.99	0.89, 1.09
Age group
17–24	4.80	3.88, 5.95
25–39	2.81	2.43, 3.26
40–59	2.07	1.81, 2,37
60+	Referent
Race
White	Referent
Racialized	1.41	1.18, 1.68
Material deprivation index
1 (least deprived)	Referent
2	0.88	0.75, 1.05
3	0.82	0.69, 0.98
4	0.81	0.69, 0.97
5 (most deprived)	0.95	0.78, 1.15
Social deprivation index
1 (least deprived)	Referent
2	0.98	0.80, 1.21
3	1.05	0.86, 1.29
4	1.05	0.86, 1.29
5 (most deprived)	0.91	0.74, 1.12
Month of donation
January	Referent
February	2.60	1.99, 3.4
March	4.08	3.20, 5.20
April	5.31	4.24, 6.65
May	10.69	8.54, 13.37
June	14.31	11.33, 18.08
July	16.28	12.96, 20.46
August	21.38	16.98, 26.93

The percentage of anti-N positive blood donations increased from 3.8% in January 2022 to 50.9% in August 2022, while the percentage of the general population testing positive by NAAT over the same period was initially similar at 3.5%, but increased at a more modest rate to 12.5%, only about a quarter of the seroprevalence rate (Figure 5). The number of NAAT tests carried out in laboratories across the province increased over 2021 with a peak in May of 2021 at 242,576 tests and another peak in December 2021 with 176,841 tests (see Figure 6, for comparison with other provinces see Appendix Figure A1). The percentage of NAAT positive for SARS-CoV-2 increased from about 1% in November 2021 to a peak of 30.7% in April 2022, with the number of NAATs performed decreasing after January 2022. Over the pandemic the population of Nova Scotia increased from 989,154 in 2020 to 1,006,805 in 2021 to 1,035,666 in 2022 (Statistics Canada Table 17-10-0009-01).

Figure 5:

SARS-CoV-2 anti-N seroprevalence in blood donors versus cumulative percentage of general population with reported positive NAAT test (using laboratory testing) from January 2021 to August 2022

Figure 6:

The frequency of SARS CoV NAAT testing and percentage of tests positive for SARS-CoV-2 from March 2020 to August 2022

Discussion

The infection-related seroprevalence in Nova Scotia blood donors remained very low (less than 0.5%) over 2020 and 2021. Vaccination-related seroprevalence increased to nearly 100% of donors between January 2021 and August 2021 in progressively younger age groups consistent with the initial vaccine roll out policy and high vaccine coverage in adults. Despite this, infection-related seroprevalence increased with the emergence of the more contagious Omicron variant (but with much lower severity and mortality in vaccinated individuals) in late 2021 to about 50% by August 2022 and was highest in younger donors. Our study contributed the bulk of the data for Nova Scotia in a comprehensive national modelling study by Murphy and colleagues that traversed the course of the pandemic (23). There were two other national studies that contributed Nova Scotia results, but their numbers of samples were small and time points limited, and there were no local studies. It is therefore not surprising that the Nova Scotia modelled results are similar to the results we report here, and underscores the important role that the national reach of blood donor seroprevalence has played in Canada over the pandemic.

Blood donors are a healthy adult population, thus not representative of some segments of the general population, notably children, elderly people, some Black, Indigenous and racialized groups, those with chronic conditions, and those at risk of blood transmissible infections who may be part of those at higher risk of getting SARS-CoV-2 infection. Nevertheless, blood donors have provided SARS-CoV-2 seroprevalence data to inform public health policy around the world and results are not dis-similar from other population samples (4,5,11,18,21,33,34).

In addition to our blood donor sero-surveillance, there have been three other national serosurveys over the pandemic which included Nova Scotia participants (15,17,18). These relied on self-collected mail-in blood spots and had the advantage of study questionnaires, but limitations included smaller sample size, point prevalence rather than continuous monitoring, and imperfect response rates. To date two have published results up to 2022 (15,18,19). Some prenatal sero-surveillance was carried out in Nova Scotia, but this would include only younger females (35). Some provinces carried out seroprevalence studies using residual samples from patient testing, notably British Columbia, Alberta, Saskatchewan, Manitoba, and Ontario (2020 only) (10–13); however, such testing was not carried out in Nova Scotia nor the other Atlantic provinces. Ours is the first report of SARS-CoV-2 population seroprevalence in Nova Scotia.

In Nova Scotia, similar to other Canadian provinces, vaccination was deployed in late December 2020 initially limited to high-risk individuals including persons in long- term care and health care workers, and then to older age groups followed by progressively younger age groups as supply allowed; these age cohorts are clearly seen in our 2021 spike antibody seroprevalence results. The higher odds ratio in females is likely related to vaccination prioritizing to those providing personal care. Only anti-spike antibodies are produced after spike protein-based vaccination (36). All Canadian approved vaccines are spike protein based (37). By the fall of 2021 the median anti-S concentration was very high, increasing in early 2022 consistent with recommendations for a booster dose of COVID-19 vaccine. There was also a decrease in anti-S concentration suggestive of waning antibody in the months after vaccine roll-out. By fall of 2021 two-dose coverage in the total Nova Scotia population was 72%, and coverage among adults 60 years of age and older was >90% (26,27). In contrast, nearly 100% of donors had antibodies consistent with vaccination perhaps reflecting greater acceptance of vaccination in donors. Higher donor COVID-19 vaccination than the general population was also noted in the United States (5,38). We note that there is no conclusive evidence of a specific anti-S threshold for immunity, and that the U/mL values are assay specific and cannot be easily compared with other assays.

Anti-N (infection-related) seroprevalence remained very low through to the end of 2021. This is similar to other Atlantic provinces, but contrasts with other provinces where both confirmed COVID-19 cases (29) and blood donor seroprevalence increased with successive waves of infection (20). Prior to Omicron, the number of confirmed COVID-19 cases in Nova Scotia was relatively low compared to other non-Atlantic provinces and access to testing was high. The low rates seen in the seroprevalence data are likely largely the result of public health measures put in place early in the pandemic, including severe restriction of travel into the region. Although Nova Scotia has an international airport, international travel was suspended early in the pandemic. Leaders, including Medical Officers of Health, from each of the Atlantic provinces met to discuss how to ensure COVID-19 safety in the region while encouraging economic strength and fluidity. As of July 3, 2020, travel between the four Atlantic provinces was permitted without quarantine, termed the ‘Atlantic Bubble,’ but travellers from outside those areas were tested upon arrival and underwent a 2 weeks quarantine. On November 26, 2020, with a second wave of COVID-19 in other provinces the Atlantic bubble was suspended and travellers between Atlantic provinces also required testing. The bubble was not re-opened until June 2021 after vaccination was deployed.

In comparison to ancestral virus and earlier variants of concern, Omicron was much more transmissible. It was more immune evasive to COVID-19 vaccines but also tended to produce milder illness, especially in vaccinated individuals (39). The cumulative confirmed COVID-19 case counts provided more concordant results to seroprevalence during periods when there was a high volume of testing which included both symptomatic and asymptomatic individuals. However, when the Omicron variant predominated testing capacity was overwhelmed and, similar to other jurisdictions in Canada, the NAAT based strategy was scaled back to include primarily those at high risk of severe disease and those residing or working in high-risk settings. Access to rapid antigen testing (RAT) continued and many were using RAT kits at home to exercise appropriate precautions as recommended by public health. With the omicron variant leading to higher numbers of infections and the NAAT testing volume decreased based on changes in testing policy, the test positivity rate increased accordingly. However, with the more restricted testing policy our data suggest that only about a quarter of cases were identified. Reliance solely on laboratory-confirmed cases to understand burden of disease, given the restricted access to testing, underestimated the true burden of illness.

The highest seroprevalence rates were seen in younger donors consistent with reported cases (29) and seroprevalence in other provinces (21,22) and countries (5,6,38) and a national modelling study (23) thought to be related to social behaviours of younger people. Older people are at risk of more severe outcomes (39), but also have higher vaccine coverage compared to younger people. Some may also be better able to reduce social contacts, for example, if retired or working from home. The higher odds ratio in racialized donors is also consistent with donor seroprevalence results from other provinces (20,21), other Canadian populations (40),(41). Unlike other provinces, living in a materially deprived neighbourhood was not associated with higher SARS-CoV-2 seroprevalence in Nova Scotia (20,21). It may be that the risk in Nova Scotia was less associated with socioeconomic status, at least for Omicron infection, that there was a challenge with the measure, or that neighbourhood material deprivation may not describe individuals who donate in that region. In summary, our study provides insight into SARS-CoV-2 infection rates in Nova Scotia with the emergence of Omicron. By August of 2022 our data suggest that about half of adult Nova Scotians had been infected with SARS-CoV-2 despite a universal vaccination program with high population coverage. Our results highlight the added value of seroprevalence, used in conjunction with surveillance date, to monitor burden of disease and inform public health policy.

Figure A1:

Timeline of sample collection

Figure A2:

Pie chart of ethnic/racial group of blood donors

Funding Statement

This project was supported by funding from the Government of Canada through the COVID-19 Immunity Task Force.

Contributors:

Conceptualization, SF O’Brien, S Deeks, T Hatchette, C Pambrun, SJ Drews; Funding Acquisition, C Pambrun; Methodology, SF O’Brien, SJ Drews; Writing – Original Draft, SF O’Brien; Writing – Review & Editing, SF O’Brien, S Deeks, T Hatchette, C Pambrun, SJ Drews.

Ethics Approval:

N/A

Informed Consent:

N/A

Registry and the Registration No. of the Study/Trial:

N/A

Data Accessibility:

Summary data are available on https://www.covid19immunitytaskforce.ca/seroprevalence-in-canada/. Data may also be made available upon request from Canadian Blood Services (contact person: Sheila O’Brien) subject to internal review, privacy legislation, data sharing agreements and research ethics approval.

Funding:

This project was supported by funding from the Government of Canada through the COVID-19 Immunity Task Force.

Disclosures:

This work originated at Canadian Blood Services.

Peer Review:

This manuscript has been peer reviewed.

Animal Studies:

N/A