The simple numbers every government should use to fight anti-vaccine misinformation

Ever since governments in Australia started sharing the vaccination status of those people unfortunate enough to have died due to Covid, or to require treatment in an intensive care unit, this information has been taken out of context.

The numbers often show that of the people who died, or are in the ICU, a significant number had been vaccinated.

Social media accounts of people who routinely share anti-vaccine misinformation, including some politicians, have taken this information out of context to present a false narrative that vaccines are ineffective.

When the number of vaccinated and unvaccinated people who have died, or ended up in ICU, is shown as a proportion of the total number of people in the state we can clearly see the protective effect of vaccines.

This also means that total number of severe outcomes is potentially going to be larger. However, when these numbers are put into context as proportions of the total number of people in the state, the protective effect of vaccines is obvious.

When you plot the data as a proportion of cases by age group, you can see that although the risk goes up with age, it also shows vaccinated people of all age groups are less likely to die or require treatment in intensive care if they catch Covid:

Notes and methods

Showing these rates is only possible as the NSW government publishes much more detailed information than the federal government or other jurisdictions in its weekly surveillance reports, and makes a hospitalisation and ICU rate for vaccinated and unvaccinated people available in the Covid-19 Critical Intelligence Unit report.

Cases, deaths and ICU numbers by age and vaccination status are from table 7 of the NSW weekly surveillance report. These were aggregated to the 10-plus and 50-plus age groups for the above graphics.

Vaccination numbers for the 12-plus and 50-plus groups were averaged from federal health vaccination statistics between 26 November and 1 January.

There is a mismatch in the age brackets between the two datasets, but the impact of this is likely to be negligible as there were only five serious outcomes in total in the 10-19 age bracket.

Thanks to Gideon Meyerowitz-Katz from the University of Wollongong for providing feedback on an earlier version of this graphic, and any errors remain my fault. You can contact me at nick.evershed@theguardian.com to report any issues