How many excess deaths have there been in Australia?

The answer depends on the model you (or the ABS) use to predict expected deaths

Recently I was asked to check a figure for excess deaths since the beginning of 2021.

The quick answer is that the Australian Bureau of Statistics (ABS) now say that there have been 29,601 excess deaths since 2021.

But not long ago the ABS’ s data was showing much higher total excess deaths and we need to question why the numbers have changed.

As we went through the first three and a half years of the pandemic we relied on the Provisional Mortality Statistics (PMS) published monthly by the ABS.

To work out excess deaths you have to compare actual numbers of deaths with expected numbers of deaths.

How you decide on the expected number is arbitrary and can make a huge difference to the number of excess deaths you calculate.

With the PMS, for 2020 and 2021, the expected number of deaths in each week was calculated as the average number of deaths in the corresponding week of the five pre-pandemic years, 2015-2019. The ABS called this expected number of deaths ‘baseline’.

Their approach made a lot of sense, and, laudably, the method was chosen in advance of significant numbers of excess deaths being reported. One could argue that it might have been better to use death rates rather than deaths¹ but the model had many good points².

After these first two years, the ABS changed their baseline for the PMS for 2022 and 2023 to one that I and many others feel was less easy to justify.

The new ‘official’ measure of excess deaths

Things took a step change, however, in July 2023 when the ABS released a new model for expected deaths during the pandemic, which has higher expected deaths on average, and thus lower excess deaths.

The ABS tells us that this new model now provides the “official” excess mortality statistics for the pandemic. It will be released six-monthly with the most recent release being on 18th December 2023.

Now wait a minute! We need to look at this new model very closely, because as you will see from the table below it presents a very different picture of the pandemic.

According to the PMS, there have been 50,555 excess deaths in Australia since January 2020, the year in which the pandemic began.

But the new official statistics more than halve the number of excess deaths over the pandemic to date - to 24,351.

The bottom line of the table shows how many fewer excess deaths there are with the new model compared to the PMS.

The fact that the new model was created retrospectively - some three years after the pandemic began - raises concerns. It was presumably made knowing that the PMS had first shown runs of excess deaths in Autumn and Spring 2021, and that deaths had been unusually high throughout 2022 and remained so as we went into 2023.

Another concern is that the new model results in numbers of excess deaths which, one imagines, would be more palatable to the government.

Given these concerns that there may be a risk of bias, it is important that the ABS should be transparent about how their model works.

We also need to know why they selected this particular model above others they considered - including simply improving and refining their original baseline used for the PMS in 2020 and 2021.

We must understand the causes of the excess deaths

So, let’s go back to the question of how many excess deaths there have been since 2021 (which was he year that the covid vaccines were rolled out, of course). According to the PMS: 48,954. According to the new model: 29,601. But either way, that’s a lot of earlier-than-expected deaths.

Someone - one would hope a responsible government - should be carefully and thoroughly investigating to understand the cause(s) of these unexpected deaths, and investigating with an open mind.

Some excess deaths will have been caused by covid, but probably not as many as we are being led to believe, given that many of those dying from the disease were the very old (median age 85.8 in 2022) and so might be expected to die in that period anyway.

Other potential causes of the excess deaths need to be properly investigated too. And that includes the hypothesis that the genetic covid vaccines have caused some of the excess deaths.

That’s a hypothesis that sounds reasonable to me. These vaccines were using a completely new technological approach of injecting genetic instructions to get the human body to make a part of the coronavirus. And they were unleashed with only two months of safety data. In situations like that, things might go wrong.

I believe it is the Australian government’s duty to carefully test this unwelcome hypothesis.

Notes on the table

A personal communication from the ABS has made it clear that Provisional Mortality Statistics for earlier years are continually being revised. Following ABS advice, the figures for the earlier years of the PMS in the table have been calculated using data for actual and baseline deaths downloaded as the ‘Deaths by month of occurrence, 2015-2022’ spreadsheet from the December 18th 2023 PMS report, which is the most recent available at the time of writing. For this reason some of the figures will differ from earlier PMS reports, which have been widely circulated. For example, a figure of 15.3% excess deaths in 2022 was reported in the March 2023 PMS, but calculations using the latest information give this figure as 15.5%, so this is shown in the table.

For the new model, the data is all taken from the report Measuring Australia’s excess mortality during the COVID-19 pandemic until August 2023

If you would like a copy of the spreadsheet underlying the table please ask in the comments.

---

Thank you for reading Clarity on Health. All our posts are free. Please share them. If you feel moved to support our work, consider becoming a paid member. You can cancel a monthly subscription at any time. Joining for one month gives a $5 donation, but you do need to remember to cancel!

1. Using death rates (number of deaths per population helps to adjust the data for changes in the size of the population. This would be important from late 2021 onwards, when Australia reopened its international border, but it didn’t matter much during 2020 and most of 2021, when the population size remained fairly steady due to the closed borders.

2. The model was logical and defensible: it used an average of a good number of years (5) thus smoothing out any abnormal years; all the years used to predict expected deaths were before the pandemic and thus ‘normal’; weekly data was used which automatically produced a weekly forecast with seasonality. A model like this has been used to calculate excess deaths by other organisations such as the Human Mortality database (www.mortality.org) and the OECD.