Scientists have been wondering why SA’s case fatality rates for Covid-19 are lower than those of its global counterparts, but now a major question is being asked: are our mortality statistics even accurate?
The Burden of Disease Research Unit at the South African Medical Research Council (SAMRC) brings out a weekly report on natural and unnatural deaths on the national population register.
From this they figure out “excess deaths” – the number of deaths over and above what one would normally expect at that time of year.
This week, the organisation said: “In the past weeks, the numbers have shown a relentless increase.”
By the second week of July “there were 59% more deaths from natural causes than would have been expected based on historical data. It also means that reported deaths have shown a pattern that is completely different to those indicated by historical trends.”
Prof Debbie Bradshaw, chief specialist scientist and a co-author of the report, said “the timing and geographic pattern” immediately raise the issue of Covid-19 deaths and the accuracy of our fatality statistics.
“The weekly death reports have revealed a huge discrepancy between the country’s confirmed Covid-19 deaths and numbers of excess natural deaths,” she said.
Excess deaths between May 6 and July 14 were at 17,090 for the country. This included high figures for the provinces we know have been hardest hit by the pandemic: excess deaths in Gauteng over that period were around 4,700, Eastern Cape around 4,900, Western Cape about 4,000 and KZN around 2,100.
“The rise in numbers of deaths from natural causes in July in Gauteng has confirmed that the epidemic has set in the province,” according to the report.
While many of the deaths are likely direct Covid-19 deaths, others could have resulted from other diseases not being managed optimally as the health system turns its attention to the pandemic.
According to council president Prof Glenda Gray, “the SAMRC has been tracking mortality for decades in South Africa, and this system has identified excess deaths associated with the Covid-19 epidemic”.
“These may be attributed to both Covid-19 deaths as well non-Covid-19 due to other diseases such as TB, HIV and non-communicable diseases, as health services are reorientated to support this health crisis.”
In the latest South African Medical Journal, Bradshaw, and colleagues at the council Pamela Groenewald, Oluwatoyin Awotiwon and Lyn Hanmer, called for more accurate registrations of Covid-19 deaths.
That is because the stats have been skewed by variations in death certificate practices.
For example, if someone with diabetes dies of Covid-19 as a result of their comorbidity placing them at much higher risk of death in the first place, some officials would register the death as being due to diabetes, and others as a Covid-19 death.
“With increasing mortality from Covid-19, it is essential to undertake proper counting of the deaths to better understand the impact of the epidemic. Information from death certificates can be used for epidemic surveillance, outbreaks and emergencies such as the Covid-19 pandemic. Accurate and complete reporting of cause of death ensures that surveillance is effective,” they said.
Prof Fikile Nomvete from the Nelson Mandela University health sciences faculty said global statistics on Covid-19 are, by their nature, tricky because they are “a big function of highly subjective reporting”. He said it was hard to “interrogate” any statistics when one didn’t know how reliable the data was.
This potential under-reporting on Covid-19 deaths might shed new light on the country’s perceived low case fatality rate. It was pegged at 1.4%, whereas in China it was just above 2% and in Italy, startlingly, above 7%.
Based on official statistics rather than what the SAMRC report might be telling us, epidemiologists had been wondering why our fatality rates were low.
Ridhwaan Suliman, a senior researcher at the Council for Scientific and Industrial Research (CSIR), said if the case fatality rate was as low as it seemed, it could be “due to a younger median population age, the early lockdown and subsequent delay in the spread, as well as hospitals mostly being able to cope thus far”.
Prof Shabir Madhi, who is heading Covid-19 clinical trials in SA, has said for some weeks that comparisons across countries are not valid because of variability in data.
“The virus wasn’t introduced simultaneously in all countries, and other factors, possibly including climate, result in countries being at different stages,” he said.