Discovering science

An invitation to a messy process

Science is hard. Moreover, it is bloody messy. Strange, confounding, woven with contradiction.

One upshot of that is that we never rely on a single study. We build conviction with networks of evidence. Different sorts of information from different sources combine to point us in a direction. If different data sets and methodologies and populations all say similar things, we are onto something.

We can’t marry any single source or study. We invite challenge. We do our best to falsify our own convictions.

I challenge Discovery to do this.

They have published data that show: “the Pfizer COVID-19 vaccine 78% effective at 28 days post-dose one and 94% effective 14 days post-dose two, in reducing the relative risk of hospital admission.”

Okay. Good. A valid data point.

Now we need to plug it into the network. I invite Discovery’s actuaries to position their study in light of the following. [I draw the following from prior posts.]

A Lancet pre-print, Effectiveness of Covid-19 Vaccination Against Risk of Symptomatic Infection, Hospitalization, and Death Up to 9 Months: A Swedish Total-Population Cohort Study, found that Pfizer vaccine effectiveness wanes from 92% during days 15-30, then down to 47% for days 121-180. After 7 months “no effectiveness could be detected”.

For AstraZeneca did even worse. Effectiveness “was generally lower and waned faster, with no effectiveness detected from day 121 [four months] and onwards”. In fact, the effectiveness they cite was negative 19%. So it caused worse symptoms after it wore off.

Another study has shown among countries and US counties no reduction in cases with increase in vaccination rate:

In fact, the trend line points upwards. They don't say if this is statistically significant.

I invite Discovery to discuss their study in light of this evidence.