Provisional Mortality Statistics

Provisional deaths data for measuring changes in patterns of mortality during the COVID-19 pandemic and recovery period

• 44,079 doctor certified deaths occurred between 1 January 2020 and 28 April 2020 and were registered by 31 May. This compares to a baseline average of 42,507 over the past 5 years.
• The highest numbers of deaths were recorded in the weeks ending 24 March (2,657), 31 March (2,776) and 7 April (2,746).
• Counts of deaths in those weeks exceeded historic averages by 624 in total, indicating that there may have been excess deaths during those weeks.
• Deaths from pneumonia peaked in the week ending 31 March and were above historic maximums for the four weeks from 18 March to 14 April. Pneumonia deaths decreased in the last two weeks of April.
• Deaths from diabetes increased in the week ending 31 March and have remained high throughout April, with a total of 115 deaths above the historic averages over those five weeks.
   

Counts of deaths for the period from 1 January to 31 March 2020 will differ from those in the previous report, with additional deaths that occurred in that period being registered and reported between the end of April and the end of May. Changes will occur as each new report is released, with data for the most recent weeks being most affected.

This report will be updated monthly throughout 2020. Reports will only include data in full weeks. As such, data for April includes information for the four weeks from 1 April to 28 April but data for 29 and 30 April are not included. The next report will include deaths that occurred up until 26 May and were registered by the end of June.

Excess mortality is an epidemiological concept typically defined as the difference between the observed number of deaths in a specified time period and the expected numbers of deaths in that same time period. Estimates of excess deaths can provide information about the burden of mortality potentially related to the COVID-19 pandemic, including deaths that are directly or indirectly attributed to COVID-19.

Throughout this report, counts of deaths for 2020 are compared to an average number of deaths recorded over the previous 5 years (2015-2019). These average or baseline counts serve as a proxy for the expected number of deaths, so comparisons against baseline counts can provide an indication of excess mortality. The minimum and maximum counts from 2015-19 are also included to provide an indication of the range of previous counts. Minimums and maximums for any given week can be from any of the five years from 2015-19.

Tracking the number of doctor certified deaths against historic averages provides an indication of when excess deaths may occur. In 2020, this is of particular relevance because of the many potential public health impacts of the COVID-19 pandemic.

The number of COVID-19 infections by week in Australia is highlighted alongside total deaths to enable a comparison of the timelines for the pandemic with changes in numbers of deaths. Over the period to end-April, peak numbers of COVID-19 infections in Australia occurred in the four weeks from 18 March to 14 April. These weeks are therefore of particular interest in mortality analysis. Changes in patterns of mortality beyond those four weeks also provide insights into the delayed effect of COVID-19 infections (including for the small proportion of people who died from COVID-19) and other indirect effects of measures in place to reduce infections (i.e. the temporary cessation of elective surgery, changes in access arrangements for medical services).

• 44,079 doctor certified deaths occurred between 1 January 2020 and 28 April 2020 and were registered by 31 May. This compares to a baseline average of 42,507 over the past 5 years.
• Over the four weeks from 18 March to 14 April there were a total of 10,842 deaths, 709 more than the historic average for 2015-19.
• Counts of deaths over the last two weeks of April were closer to historic averages, although these counts will increase as additional registrations are finalised and reported next month. This will also be the case for most causes of death.

Note: This analysis does not include coroner referred deaths. Any changes in patterns of coroner referral could affect counts of doctor certified deaths. Some conditions have higher coroner referral rates (ischaemic heart disease, cerebrovascular diseases and to a lesser extent respiratory diseases and diabetes) so counts for those conditions would be more likely to be affected by such changes.

The graph below shows the number of provisional doctor certified deaths that occurred on a weekly basis in the four months of 2020 (red line). A baseline representing the average of 2015-2019 deaths for the same period is also shown (orange line). The minimum and maximum range of deaths that occurred in this same period is shown as a shaded outline. A dark blue line represents the number of COVID-19 infections in Australia during the time period. The graph shows the number of deaths is above the 5 year average for all weeks other than the week ending 21 April when it was slightly below. In the week ending 28 January and the weeks ending 24 March, 31 March and 7 April, the number of deaths was above the maximum range. The graph shows that the peak number of COVID-19 infections during this time period occurred in the week ending 31 March and coincided with the peak number of deaths for the same period. Data for mortality presented in this graph are available in the data cubes in the Data downloads section of this publication.

Ischaemic heart disease is the leading cause of death in Australia. The category includes acute conditions such as myocardial infarction and chronic conditions such as coronary atherosclerosis.

• 4,057 deaths occurred from ischaemic heart disease between 1 January 2020 to 28 April 2020.
• The average number of deaths recorded for the same period between 2015 and 2019 was 4,456.
• Weekly counts of heart disease deaths have remained below historic averages throughout April but are very close to the average in the last week of April.
   

The graph below shows the number of provisional doctor certified deaths due to ischaemic heart diseases that occurred on a weekly basis in the four months of 2020 (orange line). A baseline representing the average of 2015-2019 deaths for the same period is also shown (dark blue line). The minimum and maximum range of deaths that occurred in this same period is shown as a shaded outline. The graph shows the number of ischaemic heart disease deaths is below the 5 year average for all weeks of 2020 except in the week ending 18 February when deaths were slightly above the 5 year maximum. Data presented in this graph are available in the data cubes in the Data downloads section of this publication.

Cerebrovascular diseases refers to a number of conditions such as stroke, cerebral aneurysms and stenosis that affect blood flow and circulation to the brain.

• 2,734 deaths from cerebrovascular diseases (including strokes) occurred between 1 January and 28 April 2020.
• The average number of deaths recorded for the same period between 2015 and 2019 was 2,982.
• The highest number of deaths from cerebrovascular diseases was recorded in the week ending 7 April (197). This is lower than the maximum recorded in the previous 5 years but higher than the baseline average.
   

The graph below shows the number of provisional doctor certified deaths due to cerebrovascular diseases that occurred on a weekly basis in the first four months of 2020 (orange line). A baseline representing the average of 2015-2019 deaths for the same period is also shown (dark blue line). The minimum and maximum range of deaths that occurred in this same period is shown as a shaded outline. The graph shows the number of cerebrovascular disease deaths is below the 5 year average for all weeks 2020 except in the week ending 11 February where deaths were slightly above the 5 year average, and the week ending 7 April when deaths were above the 5 year maximum. In the majority of weeks, deaths due to cerebrovascular are below the minimum range for the last five years. Data presented in this graph are available in the data cubes in the Data downloads section of this publication.

Respiratory diseases include causes of death such as pneumonia, influenza, and chronic lower respiratory diseases (including emphysema and chronic bronchitis). Deaths due to COVID-19 are not included in this category. COVID-19 deaths are captured using WHO issued emergency codes U07.1 and U07.2 and are not currently grouped with any other diseases, though they are included in the total number of deaths.

• There were 3,937 deaths from respiratory diseases between 1 January 2020 and 28 April 2020.
• The average number of deaths for the same period between 2015 and 2019 was 3,711.
• Over the four weeks from 18 March to 14 April there were a total of 1,019 respiratory disease deaths, 128 more than the historic average for 2015-19.
• Respiratory disease deaths decreased in the last two weeks of April with counts falling below historic averages.
   

The graph below shows the number of provisional doctor certified deaths due to respiratory diseases that occurred on a weekly basis in the first four months of 2020 (orange line). A baseline representing the average of 2015-2019 deaths for the same period is also shown (dark blue line). The minimum and maximum range of deaths that occurred in this same period is shown as a shaded outline. The graph shows the number of respiratory disease deaths is above the five year average for all weeks of 2020 other than the weeks ending 11 February, 10 March, 21 April and 28 April. The number of respiratory disease deaths are above the maximum range in the weeks ending 21 January, 4 February, 3 March, 24 March and 31 March. In the week ending 11 February it is slightly below the minimum range. Data presented in this graph are available in the data cubes in the Data downloads section of this publication.

Chronic lower respiratory diseases include chronic bronchitis, emphysema, asthma and chronic obstructive pulmonary diseases (COPD). Chronic lower respiratory diseases are a sub-group of respiratory diseases (J00-J99).

• There were 2,158 deaths from chronic lower respiratory diseases from 1 January to 28 April 2020.
• The average of number of deaths for the same period from 2015 to 2019 was 2,057.
• The number of deaths from chronic lower respiratory diseases decreased during April.
   

The graph below shows the number of provisional doctor certified deaths due to chronic lower respiratory diseases that occurred on a weekly basis in the first four months of 2020 (orange line). A baseline representing the average of 2015-2019 deaths for the same period is also shown (dark blue line). The minimum and maximum range of deaths that occurred in this same period is shown as a shaded outline. The graph shows the number of chronic lower respiratory disease deaths fluctuates for the first six weeks of 2020 when compared with the five year average. From the week ending 18 February until the week ending 7 April it is above the five year average before falling below average for the following two weeks and finishing slightly above in the week ending 28 April. Deaths due to chronic lower respiratory diseases were above the 5 year maximum in the weeks ending 21 January, 4 February, 18 February, 3 March and 24 March. Chronic respiratory disease deaths were below the 5 year minimum in the weeks ending 28 January, 11 February and 21 April. Data presented in this graph are available in data cubes in the Data downloads section of this publication.

Influenza and pneumonia are acute respiratory diseases often grouped together when compiling leading causes of death. They are two distinct diseases: influenza is a viral infection, while pneumonia can be caused by a virus, bacteria or fungi. Influenza is a common cause of pneumonia.

Influenza deaths are highly seasonal and most commonly occur in Australia between the months of May and September. The timing, length and severity of the influenza season varies considerably from year to year. Influenza and pneumonia are a subset of respiratory diseases (J00-J99).

• There were 805 deaths due to influenza and pneumonia recorded from 1 January to 28 April 2020.
• The average number of deaths recorded over the same period from 2015 to 2019 was 731.
• Counts of influenza deaths have remained very low throughout 2020 so deaths in this category are almost all from pneumonia.
• Over the four weeks from 18 March to 14 April there were a total of 261 influenza and pneumonia deaths, 88 more than the historic average for 2015-19.
• Numbers of influenza and pneumonia deaths decreased rapidly towards the end of April, falling below historic averages. of deaths from influenza and pneumonia were above baseline maximums from the week ending 24 March until the week ending 14 April.
   

The graph below shows the number of provisional doctor certified deaths due to influenza and pneumonia that occurred on a weekly basis in the first four months of 2020 (orange line). A baseline representing the average of 2015-2019 deaths for the same period is also shown (dark blue line). The minimum and maximum range of deaths that occurred in this same period is shown as a shaded outline. The graph shows that the number of influenza and pneumonia deaths is relatively stable against the baseline for the first 8 weeks of the year. In the week ending 3 March, and from the week ending 24 March until the week ending 14 April it is above the five year maximum range, before falling below average for the week ending 21 April and below the minimum for the week ending 28 April. Data presented in this graph are available in data cubes in the Data downloads section of this publication.

Cancer includes malignant neoplasms encompassing carcinomas, sarcomas and lymph and blood cancers.

• 15,305 deaths from cancer occurred between 1 January and 28 April 2020.
• The average number of deaths for the same time period from 2015-2019 was 14,903.
• Cancer deaths have been generally higher than baseline averages through 2020.
   

The graph below shows the number of provisional doctor certified deaths due to cancers that occurred on a weekly basis in the first four months of 2020 (orange line). A baseline representing the average of 2015-2019 deaths for the same period is also shown (dark blue line). The minimum and maximum range of deaths that occurred in this same period is shown as a shaded outline. The graph shows that the number of deaths due to cancer fluctuates during 2020 when compared with the five year average but is above average more often than not. The graph shows in weeks ending 18 February, 25 February, 31 March and 7 April the number of deaths are shown to be above the maximum range for the last five years. In the week ending 10 March it is below the minimum range. Data presented in this graph are available in data cubes in the Data downloads section of this publication.

Diabetes includes both Type 1 diabetes and Type 2 diabetes. The majority of diabetes deaths are from Type 2 diabetes.

• 1,482 deaths from diabetes occurred between 1 January to 28 April 2020.
• The average number of deaths over the same period from 2015 and 2019 was 1,331.
• Over the four weeks from 18 March to 14 April there were a total of 390 diabetes deaths, 77 more than the historic average for 2015-19.
• Deaths from diabetes remained high in the last two weeks of April with a further 36 deaths above historic averages recorded in those two weeks.
   

The graph below shows the number of provisional doctor certified deaths due to diabetes that occurred on a weekly basis in the first four months of 2020 (orange line). A baseline representing the average of 2015-2019 deaths for the same period is also shown (dark blue line). The minimum and maximum range of deaths that occurred in this same period is shown as a shaded outline. The graph shows that the number of diabetes deaths is above the maximum range in the week ending 7 January, before falling below minimum range for the following two weeks, remaining below average until the week ending 4 February. From the week ending 11 February until the week ending 24 March it is generally between the average and maximum range, and from the week ending 31 March until the week ending 28 April it is above the maximum range. Data presented in this graph are available in the data cubes in the Data downloads section of this publication.

Dementia, including Alzheimer disease, is the second leading cause of death in Australia and counts of deaths from dementia have increased steadily over the past 20 years. This increase should be taken into consideration when comparing 2020 counts against baseline averages.

• 4,702 deaths from Dementia, including Alzheimer disease occurred between 1 January and 28 April 2020.
• The average number of deaths for the same period from 2015 to 2019 was 3,959.
• Deaths due to dementia have remained above historic averages throughout 2020.
• Over the four weeks from 18 March to 14 April there were a total of 1,195 deaths, 220 more than the historic average for 2015-19.
• During the 11 weeks from 1 January to 17 March, there were an average per week of 39 additional dementia deaths compared to historic averages. Over the six weeks from 18 March to 28 April, this increased to 52 deaths above historic averages.
   

The graph below shows the number of provisional doctor certified deaths due to dementia including Alzheimer disease that occurred on a weekly basis in the first four months of 2020 (orange line). A baseline representing the average of 2015-2019 deaths for the same period is also shown (dark blue line). The minimum and maximum range of deaths that occurred in this same period is shown as a shaded outline. The graph shows that the number of deaths in 2020 due to dementia including Alzheimer disease was above the maximum range for the last five years in every week except the weeks ending 18 February and 17 March. Data presented in this graph are available in the data cubes in the Data downloads section of this publication.

Each death registration in the national mortality dataset has 3 dates:

• The date on which the death occurred.
• The date on which the death was registered with the jurisdictional Registry of Births Deaths and Marriages (RBDM).
• The date on which the death was lodged with the ABS.
   

The time between the occurrence of a death and registration can vary although in general deaths certified by a doctor are registered sooner. Coroner certified deaths undergo extensive investigative processes which can delay registration times, and for this reason they are excluded from the provisional mortality reports.

When looking to measure change over time, the completeness of data for the most recent period is important. When data are received each month by the ABS, the lag between the date of death and date of registration means that only 40-50% of reported registrations are of deaths that occurred in the month being reported. The remainder are deaths that occurred in earlier months. After a second month of reporting, approximately 95% of doctor certified registrations have been received. This is considered sufficiently complete to enable meaningful comparison with historic counts, noting that the level of completeness will be higher for the start of any given month than the end of that month.

This pattern of registration and reporting is highlighted in the table below, which also shows the slight variation in reporting timelines by cause of death. This should be considered when comparing 2020 data to the 2015-2019 baseline data.

Updates to this report will be released on a monthly basis. Tracking of all doctor certified deaths and deaths by specified causes will continue throughout 2020.

Each new report will include those deaths that have been registered and reported to the ABS since the previous report. As some of those deaths will have occurred in time periods covered in previous reports, the counts of deaths for those time periods will change from one report to the next.

Numbers of deaths cannot be aggregated across months to calculate a quarterly or annual count of deaths, as additional deaths for any given reference period can be registered and reported many months after the date on which they occurred.