Biased Surveillance Data:
Dismantling Canada's COVID-19 Mandate Evidence

"Biased Surveillance Data: Dismantling Canada's COVID-19 Mandate Evidence", Presentation by R.N. Watteel, Independent Researcher, Ottawa, Canada, BSc (Mathematics & Physics), MSc, PhD (Statistics)

Summary: 

The presentation argues that Canadian COVID-19 policy relied on fundamentally flawed vaccinated versus unvaccinated data. Early fatality estimates were quickly revised downward, yet restrictive measures escalated rather than eased. The speaker shows that clinical trials did not test key policy claims such as reduced transmission, while real-world data later contradicted official narratives. She argues that multiple built-in biases made surveillance data unsuitable for causal conclusions. When these biases could no longer support the narrative, authorities stopped reporting key metrics and shifted to simulation models. The conclusion is that evidentiary standards were abandoned and replaced with politically convenient modeling, raising serious accountability concerns.

Additional reading:

• Unreliable Evidence: Flawed Vaccinated vs. Unvaccinated Comparisons in Canada’s COVID-19  Vaccine Mandates, R.N. Watteel, November 22, 2025 

• Fisman's Fraud: www.fismansfraud.ca 

• Memorial page for Canadian children who suddenly died following COVID-19 vaccination: Answers4Sean (www.Answers4Sean.ca) 

• Evidence of Data Skewing by PHAC in its reporting of Cases Following Vaccination

• Evidence of Foreign and Corporate Influence on Canadian health policy Undermining Canadian National Interests and Constitutional Rights

The seven biases identified

The core contribution of the paper is the identification of seven major methodological biases embedded in mandate-era data presentation:

1. First, selection bias, often described as the “healthy vaccinee effect,” meant that vaccinated and unvaccinated populations differed systematically in baseline health, socioeconomic status, and behavior. These differences alone could produce outcome gaps unrelated to vaccine effects.

2. Second, misclassification bias arose from how vaccination status was defined. Infections occurring shortly after vaccination were frequently counted as “unvaccinated,” artificially inflating risk in that group and creating misleading impressions of vaccine effectiveness and later “waning.”

3. Third, testing bias reflected changing testing rules, differential testing by vaccination status, behavioral responses, and large temporal and provincial variations. These shifts distorted case rates and made comparisons unstable.

4. Fourth, misattribution bias blurred the distinction between being hospitalized or dying with COVID versus from COVID. Incidental hospitalizations and deaths involving multiple comorbidities were counted in ways that did not reliably measure disease burden.

5. Fifth, denominator errors stemmed from underestimating the unvaccinated population due to outdated census data, in some cases shrinking the unvaccinated denominator dramatically and inflating per-capita rates.

6. Sixth, misuse of age standardization amplified errors when applied to unstable or incorrect age-specific estimates, sometimes reversing observed trends and masking low risk in younger populations.

7. Seventh, cumulative time-based comparisons aggregated data across fundamentally different phases of the pandemic—different variants, coverage levels, and policy regimes—producing totals that had no clear causal meaning.

Details summary of this article and presentation are published on Substack