I don't think there should be much difference in the categories between the unvaccinated and vaccinated groups as you suggested above since the vaccines are given free of charge in the U.K. and there is no reason to believe "Younger, Healthier, Richer, and More Educated people" would be more likely to be unvaccinated.
In the UK, vaccines given free of charge to OLDER people first. That matters.
Thank you for proving that Unvaccinated group is far younger and healthier than Vaccinated group.
UK Vaccination Roll-out:
Phase 1: Age 50+ older
Phase 2:
a) Age 40-49 first,
b) then Age 30-39,
c) then Ages 18-29 (LAST)
So yes, Unvaccinated group is younger and healthier. The vast majority of vaccinations in UK are among the older group first based on how the vaccines are rolled out.
Thanks for proving it.
Also, without knowing the baseline demographics, you can't assume they be normally distributed across the vaccination or unvaccinated groups.
But why do we have to go into all these confounding variables?
Because older people are more likely to die compared to young people, with or without vaccines.
Because poorer people are more less likely to have good nutrition (which is associated with mortality), with or without vaccines.
Because uninsured people are less likely to have access to healthcare (which is associated with mortality), with or without vaccines.
Because Obese/Overweight people are more likely to die compared to healthier people, with or without vaccines.
Because comorbidity such as cardiovascular disease/diabetes/high BP are positively associated with death, with or without vaccines.
If you do not make the two groups comparable, you are not isolating the solitary effect of vaccine exposure alone. That is called a systematic bias in your results. Do you like biased results?
Even the vaccine manufacturers would not break down the efficacy rate to different age groups, let alone all the other categories.
Pfizer and Moderna's large RCT does adjust for age via randomization so the two cohorts have similar age compositions.
It's not about reporting efficacy rates BY age groups, it's about eliminating the confounding effects of age by ensuring both groups are equal is age distribution. Randomization also ensures ALL other confounding variables, including residual confounding variables that is not measurable is equally distributed in both arms. Randomization takes care of all of the confounding variables.
Why do you expect us to do it here, assuming we have the required data, which we do not?
Don't use low-quality observational data to make conclusions, because large well-designed RCT's (already published by Pfizer/Moderna) are superior in level of evidence than observational data?
In the hierarchy of quality data, there is:
#1: Meta-analysis of RCTs
#2: RCTs
#3: Cohorts with control arms.
#4: Case-control studies
#5: Case series observational data (This is where you are getting your data from)
The level of evidence that you are estimating the crude mortality rate is basically case-series from public active surveillance database based on self-reported data. There is no attempt at long-term collection of side-effects or adverse events between vaccine or unvaccinated in a systematic manner like in a prospective cohort clinical trial.
The general efficacy/protection rate for the sample pool of population as a whole (with its usual mix of age groups, income level, education level, health status, etc) would already suffice for our discussion in this type of forum, even if the sample pool, and therefore the analysis, only represents most closely the population it was taken from.
Wow, using your logic, FDA should just go ahead with population-level vaccination without requiring large well-designed RCTs.
You are completely clueless if you think just active surveillance self-reporting on population-level can replace the rigorous FDA process of requiring RCT study designs.
Also, vaccination roll-out is targeted to older people first, so the vaccinated group is definitively OLDER than the unvaccinated group. So there is an inherent systematic bias in the way you are interpreting the results.
