• https://www.youtube.com/watch?v=K3odScka55A

      ... “efficacy rates,” of around 95%. But the third vaccine introduced in the US, from Johnson & Johnson, has a considerably lower efficacy rate: just 66%. Look at those numbers next to each other, and it’s natural to conclude that one of them is considerably worse. Why settle for 66% when you can have 95%? But that isn’t the right way to understand a vaccine’s efficacy rate, or even to understand what a vaccine does. And public health experts say that if you really want to know which vaccine is the best one, efficacy isn’t actually the most important number at all.

      #covid-19 #sars-cov2 #statistique #vaccine

    • https://en.wikipedia.org/wiki/Vaccine_efficacy

      [...]

      Vaccine efficacy formula

      The outcome data (vaccine efficacy) generally are expressed as a proportionate reduction in disease attack rate (AR) between the unvaccinated (ARU) and vaccinated (ARV), or can be calculated from the relative risk (RR) of disease among the vaccinated group.[3][4][5]

      The basic formula[6] is written as:
      VE= (ARU-ARV): ARU x 100%

      VE = Vaccine efficacy,
      ARU = Attack rate of unvaccinated people,
      ARV = Attack rate of vaccinated people.

      An alternative, equivalent formulation of vaccine efficacy
      VE=(1-RR) x 100%
      where RR is the relative risk of developing the disease for vaccinated people compared to unvaccinated people.

      Testing for efficacy

      Vaccine efficacy differs from vaccine effectiveness in the same way that an explanatory clinical trial differs from an intention to treat trial[clarification needed]: vaccine efficacy shows how effective the vaccine could be given ideal circumstances and 100% vaccine uptake; vaccine effectiveness measures how well a vaccine performs when it is used in routine circumstances in the community.[7] What makes the vaccine efficacy applicable is that it shows the disease attack rates as well as a tracking of vaccination status.[jargon][7] Vaccine effectiveness is more easily tracked than the vaccine efficacy considering the difference in environment;[vague] however, the vaccine efficacy is more expensive and difficult to conduct. Because the trial is based on people who are taking the vaccination and those not vaccinated, there is a risk for disease, and optimal treatment is needed for those who become infected.

      The advantages of a vaccine efficacy have control for all biases that would be found with randomization, as well as prospective, active monitoring for disease attack rates, and careful tracking of vaccination status for a study population there is normally a subset as well, laboratory confirmation of the infectious outcome of interest and a sampling of vaccine immunogenicity.[7][failed verification] The major disadvantages of vaccine efficacy trials are the complexity and expense of performing them, especially for relatively uncommon infectious outcomes of diseases for which the sample size required is driven up to achieve clinically useful statistical power.[7]

      It has been proposed that standardized statements of efficacy be parametrically expanded to include multiple categories of efficacy in a table format. While conventional efficacy / effectiveness data typically shows ability to prevent a symptomatic infection, this expanded approach could include prevention of outcomes categorized to include symptom class, viral damage minor/serious, hospital admission, ICU admission, death, various viral shedding levels, etc. Capturing effectiveness at preventing each of these “outcome categories” is typically part of any study and could be provided in a table with clear definitions instead of being inconsistently presented in study discussion as is typically done in past practice. Some 2021 era COVID-19 studies appear to be implementing similar methods and presentation. Improved methods and presentation remain desirable.[8][9]

      [...]

    • #efficacité_clinique vs.
      #efficacité_sérologique

      https://fr.wikipedia.org/wiki/Vaccin#Efficacité

      [...]

      L’efficacité clinique d’un vaccin se mesure par la réduction de la fréquence de la maladie chez les sujets vaccinés (taux de protection effectif de la population vaccinée). Elle est parfois estimée par des marqueurs de substitution (taux d’anticorps connus protecteurs)2, mais l’efficacité sérologique (mesurée en laboratoire) ne concorde pas toujours avec l’efficacité clinique (mesurée en épidémiologie de terrain).

      [...]

    • La vidéo est très claire ; elle indique pourquoi on ne peut pas comparer des résultats d’essais conduits dans des contextes différents (en termes d’incidence du virus et des variants présents), et en conclusion rappeller que les 6 vaccins évalués offrent 100% de protection contre un covid qui conduirait à une hospitalisation.