Why are 40% of COVID hospitalizations in the UK among the vaccinated?

Data and Metrics Data Literacy Vaccines

Q: I heard 40% of COVID hospitalizations in the UK are among the vaccinated. What’s going on, this sounds bad?

A: This is a confusing statistic, for sure. This number is actually good news and reflects the really high vaccine coverage of the older population in the UK.

🙋‍♀‍ Wait, what now?

Let me explain! No, there is too much, let me sum up.

As vaccinations go ⬆️, total hospitalizations will decrease but the SHARE of hospitalizations who are vaccinated will increase.

This is NOT because the vaccines aren’t working, but because there are MORE vaccinated people and FEWER hospitalizations overall (remember to MIND THOSE DENOMINATORS!!).

The COVID-19 vaccines are REALLY good, but no vaccine is 100%. Some infections and even hospitalizations among the vaccinated will slip through, primarily among the oldest and most vulnerable groups whose immune systems may not respond as well. We are thus unlikely to drive COVID infections or hospitalizations down to zero.

This means:

💥 When *nobody* is vaccinated, 100% of COVID hospitalizations are in unvaccinated people.

💥 When *everyone* is vaccinated, 100% of (a much smaller number of) hospitalizations are in vaccinated people.

So somewhat counterintuitively, having a high % of hospitalizations among the unvaccinated is NOT a good thing.

The fact that a high % of COVID hospitalizations and deaths in the US are in unvaccinated groups is evidence that vaccine coverage is too low, and there are a lot of high-risk people who haven’t yet been vaccinated in the US compared to the UK.

Over 95% of those age 65+ in the UK have been fully vaccinated, compared to 79.8% in the US.

In this worked example from the Financial Times, you can see how this works out for a hypothetical population of 1 million people with either 92% or 70% of the population vaccinated.

With 92% of the population vaccinated, a higher % of those hospitalized are fully vaccinated, but the total number of people hospitalized is MUCH LOWER.

Or if it helps to think of the same phenomenon outside of vaccination, a different example:

Around two-thirds of people who die car accidents in the UK are wearing a seatbelt, despite nearly 99% seatbelt usage overall. This doesn’t mean that seatbelts don’t work. The 1% of non-seatbelt-wearers account for a disproportionately huge 33% of the deaths, and thus there would be MANY more deaths if seatbelt use was lower.

💥 BOTTOM LINE: The vaccines are working.

➡️ The share of hospitalizations or deaths who are fully vaccinated is not a good metric for tracking effectiveness of the vaccines.

➡️ What we care about is the hospitalization rate in vaccinated compared to unvaccinated. Our real world data still show that even for Delta, the risk of hospitalization in the vaccinated is 95% LOWER than the risk in the unvaccinated.

Those Nerdy Girls


Original tweet thread from John Burn-Murdoch from Financial Times

Financial Times article

Some helpful explainers of this concept:

Live Free or Dichotomize

You Can Know Things


Vaccine Efficacy against hospitalization (Delta Variant)

Link to Original FB Post