Tag: vaccine efficacy

Observational studies of COVID vaccine efficacy are riddled with bias/ Not counting cases 14 days after dose 2 is a problem

Mary Pat Campbell

Link:https://www.drvinayprasad.com/p/observational-studies-of-covid-vaccine

Graphic:

Excerpt:

In the experiment, he says, what if we compare the control arm of the Pfizer study against an imaginary vaccine arm. And for the thought experiment assume the vaccine is useless. As the table above shows, both groups have identical numbers of covid cases— just what you would expect from a useless vaccine. A straight forward analysis shows no benefit (second to last row)

But in the ‘fictional vaccine observational study’ cases are excluded for 36 days. When this is done the useless vaccine, looks like it reduces infections by 48%!!

Doshi makes a very good point in his paper that the solution is to subtract the 36 day infection rate from the observational control arm. Sadly most investigations don’t do that.

This is one of several biases Doshi discusses, and it plagues the vaccine literature.

Author(s): Vinay Prasad

Publication Date: 14 May 2024

Publication Site: Vinay Prasad’s Observations and Thoughts, substack

Weekly COVID-19 age-standardised mortality rates by vaccination status, England: methodology

Mary Pat Campbell

Link:https://www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/deaths/methodologies/weeklycovid19agestandardisedmortalityratesbyvaccinationstatusenglandmethodology#age-standardised-mortality-rates

Graphic:

Excerpt:

Age-standardised mortality rates are calculated for vaccination status groups using the Public Health Data Asset (PHDA) dataset. The PHDA is a linked dataset combining the 2011 Census, the General Practice Extraction Service (GPES) data for pandemic planning and research, and the Hospital Episode Statistics (HES). We linked vaccination data from the National Immunisation Management Service (NIMS) to the PHDA based on NHS number, and linked data on positive coronavirus (COVID-19) Polymerase Chain Reaction (PCR) tests from Test and Trace to the PHDA, also based on NHS number.

The PHDA dataset contains a subset of the population and allows for analyses to be carried out that require a known living population with known characteristics. These characteristics include age-standardised mortality rates (ASMRs) by vaccination status and the use of variables such as health conditions and census characteristics.

Author(s): UK government, ONS

Publication Date: Accessed 27 Sept 2021

Publication Site: Office of National Statistics

Israeli data: How can efficacy vs. severe disease be strong when 60% of hospitalized are vaccinated?

Mary Pat Campbell

Link: https://www.covid-datascience.com/post/israeli-data-how-can-efficacy-vs-severe-disease-be-strong-when-60-of-hospitalized-are-vaccinated

Graphic:

Excerpt:

These efficacies are quite high and suggests the vaccines are doing a very good job of preventing severe disease in both older and young cohorts. These levels of efficacy are much higher than the 67.5% efficacy estimate we get if the analysis is not stratified by age. How can there be such a discrepancy between the age-stratified and overall efficacy numbers?

This is an example of Simpson’s Paradox, a well-known phenomenon in which misleading results can sometimes be obtained from observational data in the presence of confounding factors.

Author(s): Jeffrey Morris

Publication Date: 17 August 2021

Publication Site: Covid-19 Data Science

On the Interpretation of Vaccine Efficacy Rates

Mary Pat Campbell

Link: https://www.yengmillerchang.com/post/on-the-interpretation-of-vaccine-efficacy-rates/

Graphic:

Excerpt:

With COVID-19 vaccines now being widely available in the U.S., I’ve seen various interpretations of vaccine efficacy rates. As one example, the paper disseminating the study on the efficacy of BioNTech and Pfizer’s vaccine BNT162b2 states in its results section:

BNT162b2 was 95% effective in preventing Covid-19

The intent of this post is to clarify the interpretations of such numbers.

Author(s): Yeng Miller-Chang

Publication Date: 15 July 2021

Publication Site: Math, Music Occasionally, and Stats

J&J and Delta update

Mary Pat Campbell

Link: https://yourlocalepidemiologist.substack.com/p/j-and-j-and-delta-update

Graphic:

Excerpt:

Neutralizing antibodies in a lab is difficult to map to real world efficacy. Thankfully, some scientists figured out a mathematical model (here). Using their model, this means the efficacy of J&J would be around 55-60% against symptomatic disease. It will still work well against severe disease.

In the same update, J&J said their vaccine continues to work over time, with strong responses for up to 8 months. This is because there’s only 8 months of data; we are optimistic it will last longer.

Author(s): Katelyn Jetelina

Publication Date: 2 July 2021

Publication Site: Your Local Epidemiologist on substack

Vaccine Efficacy and the Immunity Time Spans

Mary Pat Campbell

Link: https://polimath.substack.com/p/vaccine-efficacy-and-the-immunity

Graphic:

Excerpt:

Let’s refer back to the Pfizer study submitted to the FDA. In that study, 18,555 people were vaccinated and 18,533 people received the placebo injection. In these groups, 7 days after the second dose was administered, we saw that the vaccinated group got infected at only 5% the rate that the placebo group was infected.

Furthermore, this is the number of cases we see over the course of a two month study. So those 9 people out of 18,555 were not symptomatic and infectious that whole time, but only for a few weeks.

So, to take CNN’s example and re-imagine it for the reality we have with this data.

Let’s say 1 million people are travelling. If everyone is unvaccinated (and the window of infection is roughly one week), there will be about 1,100 infected travelers.

If, however, everyone is vaccinated, there will be about 60 infected travelers and their chance of infecting you (my dear vaccinated friend) is reduced substantially.

Author(s): PoliMath

Publication Date: 13 April 2021

Publication Site: Marginally Compelling on substack