Seeing the evidence of COVID vaccine efficacy

Because we are epidemiologists, we often have people asking us questions about COVID-19 in general and about the vaccines in particular. Like many others, we have been awed and grateful for the speed at which safe and efficacious vaccines have been developed and rolled out, and the enthusiasm of our friends and neighbours to line up for a shot! However, like many other regions around the world, our home of British Columbia is showing some signs of vaccines hesitancy amongst a minority of the population – with the uptake of first doses slowing down substantially at around 80% of the eligible population. Given what we have seen internationally regarding the ongoing spread of the Delta variant,1 we wanted to share some of the work that we have been doing to visualize COVID public health data across North America. We have been looking at vaccine efficacy as part of the development of our COVID-19 data visualization that we presented at the recent ISPOR meeting (see below) and thought we would do our bit here to reinforce the message that the vaccines work. COVID cases versus vaccinations in four jurisdictions

Covid App

We know that these vaccines have been proven efficacious under the highly standardized conditions of a clinical trial, but the important question is how does this translate to the real world? We really like the explanation provided by Oxford’s Peiro Olliaro in a recent letter to the Lancet. To begin with, 95% efficacy does not mean that 95% of people are protected from COVID-19 by the vaccine. The protection is actually much higher because of course not everyone is exposed to or contracts the virus (in fact the majority of people won’t). What it means is that in a population of 100,000, instead of the 1,000 (for example) COVID-19 cases which might occur if the population is unvaccinated, we would expect 50 cases (5%), ie. 950 people (95%) are protected. This means 99·95% of the population would be disease-free, at least initially after vaccination.2

The exact results of the COVID-19 vaccine trials are hard to compare because of differing time periods of assessment and case definitions between the trials. Nonetheless, the WHO and other public health thought leaders confidently recommend vaccination with any of the around 20 approved agents as effective protection against COVID.

It should also be noted that vaccine efficacy found during a trial does not always predict the protection attributable to a vaccine administered in the real world. There are a number of reasons for this, these include:

  • The characteristics of the study population: socioeconomic factors, age, sex, ethnicity, the presence of comorbidities, or geographic location of participants in a trial may affect whether the findings can predict effectiveness if the vaccine is more widely deployed. The Pfizer and Moderna trials occurred at a time where no vaccines were available and they enrolled relatively high percentages of participants ≥65 years. Whereas, in the AstraZeneca trial, which included both standard and low dose arms, the low dose group consisted of 18-55 year olds.
  • Another potential contributing factor is the particular variants circulating during the time and place of the clinical trial.3 For example, efficacy in clinical trials in South Africa was lower for all three vaccines when tested in the midst of outbreaks of the Beta variant which is associated with high transmissibility and higher in-hospital mortality, when compared with efficacy in trials conducted in countries with other, less virulent, circulating variants.4

The good news is the vaccines really do appear to be working in the real world.  Our app visualizes trends over time in cases, hospitalizations, and vaccination rates, across states and provinces throughout North America – and while there are lots of subtleties and variability, a consistent association emerges. Areas with widespread vaccination have reductions in cases and hospitalizations.  This data supports the increasing evidence that the vast majority of COVID hospitalizations and deaths occur in unvaccinated individuals.5,6  So the key now is to get as many people vaccinated as possible. We invite you to join us in contributing to that cause with a donation to the UNICEF or Go Give One vaccination campaigns.

Using publicly available data to calculate the Real-Time Reproduction Number (Rt) for Covid-19
In addition to compiling public health data on COVID by jurisdiction, our app calculates a real-time approximation of the basic reproduction number using two different methodologies. We invite you to check it out.


  1. Zimmer CC, Jonathan; Wee, Sui-Lee. Coronavirus Vaccine Tracker. New York: New York Times, 2021.
  2. Olliaro P. What does 95% COVID-19 vaccine efficacy really mean? The Lancet Infectious Diseases. 2021.
  3. Ledford H. Why COVID vaccines are so difficult to compare. Nature. 2021; 591: 16-7.
  4. Abdool Karim SS and de Oliveira T. New SARS-CoV-2 variants—clinical, public health, and vaccine implications. New England Journal of Medicine. 2021; 384: 1866-8.
  5. Roser M, Ritchie H, Ortiz-Ospina E and Hasell J. Coronavirus pandemic (COVID-19). Our world in data. 2020.
  6. Polack FP, Thomas SJ, Kitchin N, et al. Safety and efficacy of the BNT162b2 mRNA Covid-19 vaccine. New England Journal of Medicine. 2020; 383: 2603-15.
  7. Baden LR, El Sahly HM, Essink B, et al. Efficacy and safety of the mRNA-1273 SARS-CoV-2 vaccine. New England Journal of Medicine. 2021; 384: 403-16.
  8. Voysey M, Clemens SAC, Madhi SA, et al. Safety and efficacy of the ChAdOx1 nCoV-19 vaccine (AZD1222) against SARS-CoV-2: an interim analysis of four randomised controlled trials in Brazil, South Africa, and the UK. The Lancet. 2021; 397: 99-111.
  9. AstraZeneca PLC. AZD1222 US Phase III primary analysis confirms safety and efficacy. 25 March 2021.
  10. Sadoff J, Gray G, Vandebosch A, et al. Safety and efficacy of single-dose Ad26. COV2. S vaccine against Covid-19. New England Journal of Medicine. 2021.