Covid-19 modellers have had a tough task, but one group in particular seems to have hit the nail on the head: the National Covid-19 Modelling Consortium said 40,000 people would die from the disease and it looks as if the number for this year will be 36,000.
Prof Gesine Meyer-Rath, a modeller at the Health Economics and Epidemiology Research Office in Johannesburg, recently disclosed on a Sunday TimesLIVE Dialogues that the consortium of which she is a member had got it right.
“We take no joy in this,” she said, “but we will probably end up being close to the 40,000 that we projected in mid-May, if you include excess deaths.”
“Excess deaths” are those above what one would normally expect from a certain time of year, based on historical data.
The Burden of Disease Research Unit at the South African Medical Research Council (SAMRC) had been bringing out a weekly report on excess deaths.
Prof Meyer-Rath said: “The only way of knowing how many deaths are Covid-19-related is to look at the pattern of how those excess deaths move through space and time, and we saw that in every single place where excess deaths show up, Covid-19 cases also increased.”
She said it looked as if the final number would be about 36,000 deaths, which would include about 80% of excess deaths, leaving a margin for those who had perhaps needed hospital care for other reasons, but had not sought it due to fear or food insecurity, which she said was a “nice way of saying starving”.
She added: “We hope that by the beginning of November only a few deaths will be added, but we will never know with absolute certainty.”
We hope that by the beginning of November only a few deaths will be added, but we will never know with absolute certainty
— Covid-19 modeller Prof Gesine Meyer-Rath
To help lay people understand how modelling works, she gave a simple example.
“Think of when you’re deciding what to wear based on the weather. You make assumptions based on data before you actually know how the weather will be for the entire day. That is exactly what we as modellers do. We rely on the data that is available, but there is also a great deal of mathematics and less intuition than you would use in an everyday situation.”
When the pandemic struck, modellers had to decide which type of model to use to try to figure out what their country would need.
She said most opted for a “standard modelling system” that had “stood us in good stead”, especially considering the uncertainties of “how many people would get Covid-19, how severely they would be ill and how many would die”.
Some of the “biggest unknowns” were “how long people would stay in hospital” and how many risk factors would affect each of the other variables, such as infections and deaths, she said.
This meant there was an absence of “perfect data”, but as the department of health and National Institute of Communicable Diseases (NICD) collected data, the model was updated, though this was not “done in front of the public” so “perhaps that could have been done better”.
Dr Nicholas Crisp, a public health specialist who was deployed to oversee the department’s response to the pandemic in the Western Cape, explained how the province was at the “spearhead” of learning about the “unknown” and responding to what the modelling was telling us.
“We didn’t know how many people would be sick. We didn’t know how many additional beds would be needed,” he said.
Also, there were surprises in store regarding how people would respond to the governmental approach.
“As community transmission picked up, we tried to offer people options for isolation, particularly for people from informal settlements where we were expecting large numbers of people to need assistance,” he said, “but many people did not want to leave their homes, even with space in three- or four-star hotels on offer.”
Many refused as they were “afraid their goods would be stolen or their houses would be occupied”, he said, “so these are some of the social things we learnt about community behaviour. You cannot simply, as a councillor or medical practitioner, decree how people are going to behave.”
Prof Shabir Madhi, who leads the SAMRC’s vaccines and infectious diseases unit, said “the number of infections is waning because of non-pharmaceutical interventions”, but that there is “also another hypothesis being explored”.
He said in very densely populated communities, such as Soweto or Khayelitsha, people had been intensely exposed to common coronaviruses before Covid-19 came along and that this had possibly induced an immune response that now stands them in good stead during the pandemic.
“This would provide cross-protection against severe forms of Covid-19 disease because underlying immunity has been induced by other coronaviruses,” said Madhi.
Would you like to comment on this article?
Sign up (it's quick and free) or sign in now.
Please read our Comment Policy before commenting.