Graham Neary Profile picture

Aug 24, 2020, 11 tweets

Time for another quick look at Irish mortality. This time, I'm going all the way back to 1989.

Do you remember the 1990s? Remember Oasis, Green Day and Blink-182?

Do you remember being afraid of death back then?

Thread.

I need to use estimates and assumptions, but I'll be upfront about them. Let me know what errors I've made. The picture for 2020 is still emerging and uncertain.

Sources - CSO (provisional from 2017), GRO (provisional for 2020), HIQA estimates for April-May 2020.

Let's begin!

Below are GRO death registrations as of end-July 2020.

Comparing the end-July and end-June GRO tables, I think total deaths from Jan to May 2020 were probably around 14,400.

HIQA's April-May estimates are a little higher than mine (a different data source, and also uncertain).

Let's naively annualise my January to May total without making any seasonal adjustment.

This naive estimate lets us compare 2020 mortality versus all years to 1989.

Conclusion: 2020 is more dangerous than the 2010s, about the same as 2004-2005.

Less dangerous than the 1990s.

Now let's do crude seasonal adjustment.

January to May (GRO) are on average c. 45% of annual CSO totals (cold weather, flu).

Applying an average seasonal effect gives a full-year 2020 estimate of 32,186 deaths.

Hopefully an over-estimate (if Jan to May are >45% of 2020 total).

Now let's draw those charts again.

Result:

At this early stage and using highly provisional data, it looks like 2020 is shaping up to be the most dangerous year since... 2016.

If true, does it matter? How many people realise that overall mortality this year is so normal?

Now for a long-term perspective - let's compare 2020 vs. the average of prior decades.

Results:

The 2010s were, on average, just 2% safer than the 2020 estimate.

The 2000s were 9% more dangerous.

The 1990s were a massive 36% more dangerous.

Despite Covid-19 in 2020. Amazing!

Many people will take this as proof that lockdown, social distancing and masks have been very effective.

If there was no national catastrophe, it must be because we prevented it.

Why would we have been so afraid, if the threat wasn't real? We are safe because of actions taken.

My question is: were you afraid in the 1990s?

Suppose we had simply returned to the 1990s, instead of having lockdown/Covid-19 (1,072 excess deaths over 97 days).

How many excess deaths would there be, in 2020, at the 1990s mortality rate?

You sure you want to know?

12,102.

In other words, the 1990s saw more than TEN TIMES as many extra deaths, year in and year out, as seen during lockdown/Covid-19.

If 100 days of lockdown/peak Covid-19 were replaced by 100 days in the 90s, it would have been three times more dangerous than what actually happened.

Conclusions:

2020 mortality is likely to be within the range of the 2010s, e.g. lower than 2016.

Progress since the 1990s/2000s in reducing mortality has been maintained.

If it didn't make sense to live in fear in the 1990s, it doesn't make much sense to live in fear in 2020.

Share this Scrolly Tale with your friends.

A Scrolly Tale is a new way to read Twitter threads with a more visually immersive experience.
Discover more beautiful Scrolly Tales like this.

Keep scrolling