Thread by @GidMK

Our big new study on the infection fatality rate of COVID-19 in developing countries is out

The news is pretty grim 🧵

medrxiv.org/content/10.110…

Throughout the pandemic, there has been an apparent contradiction - low-income places appear to have lower death rates from COVID-19 than higher-income areas

This makes no sense on the face of it

So, we looked at the infection fatality rate (IFR) of COVID-19 in every place that we could find, using antibodies to estimate the number of infections in each area and the number of registered deaths as our numerator

The results?

Firstly, developing countries had much higher seroprevalence than higher-income areas. They had more infections by a long way after the first wave than most developed countries

Worst still, they had relatively uniform seroprevalence across ages. In high-income places, edlerly people were mostly protected from infection, but in developing countries this does not seem to have happened

What about IFR?

Well, developing countries had a much higher IFR than high-income areas. This graph and table show this issue - at age 20 the risk of dying in a developing country is nearly triple as high as in a higher-income place

Now, this was not true across the board. There are some places with extremely low IFRs - an example is some locations in India

How do we explain this?

The TL:DR is actually pretty simple. If you look at death reporting systems in these places FROM BEFORE THE PANDEMIC there were very serious reporting issues, with some of these places actually certifying less than 30% of deaths properly!

If you compare places that had adequate death reporting with those that didn't the IFR is TEN TIMES HIGHER

TEN. TIMES. HIGHER.

So what we can say with a great deal of certainty is that the risk of COVID-19 is much higher in developing countries than in high-income countries for similar ages

For those locations where the death rate is very low, by far the most likely explanation is that they are simply not recording deaths very well. We knew this was an issue before the pandemic, so it's not surprising!

Ultimately, the message from this work is simple - COVID-19 has been devastating to low-income places, and we must act urgently to improve vaccine equity across the globe

This has been a mammoth piece of work, taking well over 9 months, and included a whole cavalcade of amazing researchers - @ikashnitsky, @baileyfosdick, @lonnibesancon, @anup_malani, @HEARDatUNSW, @satejsoman and more!

It turns out that the apparent contradiction I mentioned above is explained in two basic ways:

1. Developing countries are often relatively young
2. Deaths often go unreported (this includes non-COVID deaths too)

It's also worth noting that the higher IFR in developing nations is despite these areas having few nursing homes. Indeed, nursing homes seem to be quite PROTECTIVE for COVID-19 because living in the community makes it harder to prevent elderly people from getting infected