https://abs.twimg.com/emoji/v2/... draggable="false" alt="🧵" title="Thread" aria-label="Emoji: Thread">
What if we had a better metric for communicating the risk COVID-19 currently poses to society than case numbers?

The R number and deaths also have big flaws (more https://abs.twimg.com/emoji/v2/... draggable="false" alt="👇" title="Rückhand Zeigefinger nach unten" aria-label="Emoji: Rückhand Zeigefinger nach unten">)

What if we could say from yesterday& #39;s cases, we can expect 4-5 fatalities? Or from the last week, 25-30?
1/

We now know more about the infection fatality rate (IFR) and how it impacts different ages.

We know if infected, the chance of dying is roughly in line with one& #39;s own all-cause mortality rate for the next year.
2/

Therefore, the overwhelming problem with cases is that the impact differs dramatically depending on the age of the person infected.

The risk to those under 50 is relatively small. The risk to those over 50 starts to become more significant and increases the older you are.
3/

I& #39;m constantly bewildered by how much press the R number receives for COVID-19.

How can we know the true R number when so many infections are asymptomatic? Too much reliance is placed on such an uncertain number.

A good explainer on the R number https://abs.twimg.com/emoji/v2/... draggable="false" alt="👇" title="Rückhand Zeigefinger nach unten" aria-label="Emoji: Rückhand Zeigefinger nach unten">
https://twitter.com/AdamJKucharski/status/1259441470618001409?s=20
4/">https://twitter.com/AdamJKuch...

Deaths, sadly, are the best way to understand the true impact of the pandemic. Unfortunately, they are also the most lagging indicator of all metrics.

By the time they reach noticeable levels, it& #39;s already too late.
5/

If we use what we know about the IFR along with case numbers by age band, we begin to get a pretty accurate view on the number of deaths we can expect.

And for Ireland currently, that& #39;s ultimately 4-5 fatalities based on yesterday& #39;s case numbers and 25-30 from the last week.
6/

For #NPHET, instead of setting a target of 100 cases a day, the target could be <X expected deaths per day or <Y per week.

X/Y are contentious figures. It& #39;s easier for people to talk about an acceptable level of cases than deaths. But they& #39;re effectively the same thing.
7/

In Ireland, we typically have 80-90 deaths occurring per day. Is there an acceptable number of COVID-19 related deaths per day/week for you?
8/

Link to the paper with the theoretical IFRs used
https://www.medrxiv.org/content/10.1101/2020.07.23.20160895v7

It& #39;s">https://www.medrxiv.org/content/1... currently preprint and awaiting peer review, but figures reported look to be appropriate for Ireland.
9/