Australia is doing brilliantly in terms of excess mortality during the pandemic... or is it?

The Australian Bureau of Statistics pandemic period mortality data (2020-2021) presents Dr. certified deaths. Coroner certified deaths are excluded and their contribution to total data is unknown.

In any population a typical amount of people die year; referred to as “all cause mortality”. This is normal, expected, and part of life. Barring a mass natural disaster, a pandemic, or war, population all cause mortality is typically quite stable with annual year to year variation within not more than a couple of percentage points.

“Excess Mortality” reflects on a population level people who would not have been expected to die under normal circumstances, but who died anyway. It can be a good reflection of societal health and conditions.

Excess Mortality = Actual People Dead - Expected People Dead.

The covid-19 pandemic of the last 2 years has been quite unique because typically, not to disrespect the loss of human lives and loved ones, it has taken primarily the very elderly and those who were already vulnerable with multiple comorbidities. These were people with reduced life expectancy who could have been expected to have died within months, and thus would usually fall within the expected deaths within any population.

Further, when vulnerable people die, the remaining population is less vulnerable and less likely to die. Thus, in normal times, any peak in deaths maybe related to extreme weather or illness outbreaks, would be followed by a period of lower deaths until a more vulnerable population built up again; resulting in a fairly flat average line, though necessarily with some up and down deviations.

Theoretically I thought that the older age and extreme vulnerability of most of the victims should have translated to relatively few excess deaths. However, the actual data is quite different, perhaps reflective of lockdown impacts, or perhaps other as yet undefined factors.

Death patterns since the start of the pandemic have not been typical. There were death peaks associated with infective waves but these were not followed by much, if any, counter balancing negative death peaks; some countries reported lower than usual deaths in at least some of 2020; and many reported high excess deaths in 2021 even as the anticipated-to-be-lifesaving vaccines were rolled out. I selected 10 countries to show what this looks like. 0% means no excess deaths.

Australian Excess Mortality, only available up to November 28, 2021, seems to show mostly lower than usual all cause mortality since January 2020.

I looked further, and here is the Australian Bureau of Statistics (ABS) Death Chart for the same period, which seems to match the Our World in Data plot. These are Doctor Certified Deaths plotted against the 2015-2019 range.

Between 1st and 8th of August there was a notable sudden step drop in deaths, and thereafter the amplitude of the line flattened. Is this reflective of some change in reporting criteria? It would be interesting if this could be explained.

Doctor Certified deaths are certified using the Medical Certificate of Cause of Death (for general deaths) or the Medical Certificate of Cause of Perinatal Death (for perinatal deaths). The ABS stated that approximately 86 to 89% of deaths are typically certified by a doctor, and reflect natural causes of death.

The other category of deaths are reportable deaths, which are referred to the coroner for investigation. These are not included in this ABS data. Reportable deaths are defined in the Coroners Act of 2009 - Section 6. They generally include the following circumstances :

• where the person died in a violent (assault) or unnatural (injury, accident, suicide) manner

• where the person died unexpectedly and the cause of death is unknown

• where the person died under suspicious or unusual circumstances

• where the person died in circumstances where the death was not the reasonably expected outcome of a health related procedure (for example, as the result of an anesthetic)

• where the person was ‘held in care’ or custody immediately before they died

• where the identity of the person who died is unknown

Note that Covid-19 related deaths are not classed as reportable.

I found further explanation from the Australian Institute of Health and Welfare Website which stated that Coroner certified deaths take months or years to finalize, and may take several years to be closed and reflected in the final Cause of Death Record Files (CODURF). They provided an example of the 2017 data which had preliminary and revised versions before the final CODURF was released in March of 2021.

I have demonstrated that the current data from ABS is not complete. It may not be taken as reflective of the overall death rates in Australia for these pandemic years. no conclusions can be drawn. Total deaths may be considerably higher when the reportable deaths are finally accounted.

The next provisional population Dr. certified mortality data is due for release by the ABS on 30 March 2022.