- On Monday, 22 November, health officials reported an additional and historic 17 718 Covid-19 infections, creating a sharp spike in the country’s epidemic curve.
- Days later, scientists announced the discovery of the B1.1.529 variant which was named Omicron, which had a worrying number of mutations that raised concerns over its potential for increased transmissibility.
- While cases in Gauteng are now increasing dramatically, the timing of the publication of the data gave rise to impressions that SA had seen a major surge in cases just as the world’s attention turned to the country.
An untimely dump of 17 718 historic Covid-19 cases on 22 November created the impression that South Africa had seen a dire spike in coronavirus infections - just as the country told the world about the Omicron variant.
Within hours of the announcement of the discovery on Thursday, the UK and other countries introduced travel bans which have been slammed by President Cyril Ramaphosa as "unjustified".
However, at the exact moment the announcement was made, coronavirus data available showed that South Africa had seen a major spike in cases – from 312 on 21 November, to 18 586 on 22 November.
In reality there were only 868 new cases while 17 718 of the cases introduced were historic cases – some as old as April 2020, according to data reviewed by News24.
While cases are now increasing quickly, particularly in Gauteng which reported more than 6 000 new infections on Wednesday, it was the picture shown to the world on 23 November that has created additional alarm.
News24 has undertaken a data analysis to correct the misleading spike in cases on 23 November, based on data provided by the department of health. The department was asked to clarify the decision-making behind the publication of the historic data, but had not responded at the time of writing.
In the graph below, daily cases and the seven-day moving average is shown as the data has been reported, including the addition of the retrospective cases. The significant spike in cases on 23 November is clearly visible, and this was the picture available to most of the world when the discovery of Omicron was announced.
However, based on data shared with News24 by the department of health, this is what the epidemic trajectory should actually look:
While the increase remains sharp, it was not as significant on 25 November, the day of the announcement. Contextually, the current rise in national daily cases is due to a sharp increase in infections in only one province – Gauteng – while other provinces are showing early signs of increased infections.
The crude positivity rate, calculated by simply determining how many tests were positive as a percentage of all the tests done, was also skewed. On 23 November, the crude positivity rate was 15.7%, compared to 2.3% the previous day. This was because the health department had also reported roughly 75 000 historic tests.
On Thursday, Western Cape health officials declared the province was now in a fourth wave.
This is calculated based on infection numbers in previous days, and while technically true, the number of new cases is yet to rise as sharply as in Gauteng elsewhere in the country.
A closer look shows more clearly the dramatic impact of the introduction of the historic cases.
When in fact, the curve looks like this:
Professor Tom Moultrie, director of the Centre for Actuarial Research (CARe) at the University of Cape Town, works closely with testing data and told News24 that he did not believe the introduction of the historic tests were mendacious.
"It is an effort to be more transparent, and it should be commended. It's just a case of really bad timing," he said.
"It is apparent from the NICD weekly reports that antigen tests have been included for some time. But there are issues where some facilities have not been reporting all the negative antigen test results, which then skews the positivity rates at a facility level," he added.
Moultrie said the impression created by the reporting of the cases just as the announcement was made regarding the discovery of a worrying new variant was "deeply unfortunate".
Professor Alex van den Heever, a public health systems expert and the chair for Social Security Systems Administration and Management Studies at Wits University, echoed calls made from early in the local epidemic that health officials needed to report Covid-19 data by date of testing or sample collection.
"This also talks to the issue of not reconciling infection data back to the date of the test. If they did this with the publicly reported data, this problem would not occur. I don't think it 'aids transparency' to allocate infection data to the wrong time periods. It is merely more transparent at the absolute level," he said.
Professor Glenda Gray, president of the South African Medical Research Council, meanwhile said it was not so much about the cases at the time, but rather the high number of mutations in the variant that caused concern.
"Everyone started watching the cases, that's what you do after the you learn about a variant like this. At one stage the positivity rate was 1% and then it was 2%, and then it just went up. And we realised this was the beginning of the fourth wave," she said.
"An increase on one day could be an aberration, you must look at it over time."
When Imperial College virologist Tom Peacock, who was the first to pick up on sequences uploaded to a global repository, GISAID, by scientists in Botswana, South Africa and Hong Kong posted about the worrying level of mutations on 23 November on public forums, the picture shown to the world was an increase of more than 18 000 cases in a single day.
South Africa had recently entered a protracted lull in the local epidemic, with low case figures first seen at the very start of the outbreak of Covid-19 locally, a picture which is quickly changing as cases increase rapidly in Gauteng.
By lunch time on Wednesday, 24 November, UK news site iNews had broken the story, based on Peacock’s observations.
Reporting on the announcement of the variant’s discovery by Professor Tulio de Oliveira on 25 November, the day after iNews first broke the story, UK-tabloid The Daily Mail – one of the most widely read online news sites in the world - published graphs based on SA having reported 18 000 cases in a single day.
The Daily Mail article headline referred to the Omicron variant, as yet unnamed then, as "worst-ever" and "super mutant".
A screengrab of the graph included in a 25 November Daily Mail article reporting on the announcement of the new Omicron variant, showing a skewed picture of South Africa’s epidemic curve, caused by the publication of historical cases by the department of health on Monday, 22 November.
"Through ongoing efforts to ensure the best available surveillance data for decision-making, the department is aware of a number of COVID-19 antigen tests from various sources that have not been incorporated into the laboratory information system," spokesperson for the department, Foster Mohale, said in a statement on 23 November.
Mohale explained that the addition of these historic cases was not unusual and was part of "data cleaning, quality checks and endeavours for completeness".
"As of 8 November 2021, we have identified approximately 75 000 antigen tests that need to be captured into the database, of these tests about 20 813 were diagnosed as positive for SARS-CoV-2," Mohale's statement said.
It is unclear why, if 20 813 historic cases were identified, only 17 718 were reported. According to weekly testing reports by the National Institute for Communicable Diseases (NICD) however, antigen test results have been reported to some degree for months.
"As we report test data and case numbers for the past 24 hours through our outbreak reporting system, we will observe an increase in a single day, which will create a distortion of the seven-day moving average and an unusual spike on the reporting epidemic curve," Mohale said.
Scientists are still studying the Omicron variant which has raised concern for its high number of mutations, roughly 50, and the potential for to escape immunity provided by vaccines and natural immunity brought on by prior infection.
While it remains unclear whether the Omicron variant is more transmissible, early signs have concerned researchers. More data from ongoing research is only likely to be available in the coming weeks.
If you come across Covid-19 vaccination information that you do not trust, read Covid-19 vaccine myths debunked: Get the facts here. If you can't find the facts you're looking for, email us at the address mentioned in the article and we will verify the information with medical professionals.