Excess deaths will soon be a thing of the past: Medical Research Council

26 August 2020 - 15:15 By Dave Chambers and dave chambers
subscribe Just R20 for the first month. Support independent journalism by subscribing to our digital news package.
Subscribe now
Estimated excess deaths from natural causes reached 39,087 for the period between May 6 - when statistics began to depart sharply from the prediction - and August 18. In the same period, the health department reported 12,112 deaths from Covid-19.
Estimated excess deaths from natural causes reached 39,087 for the period between May 6 - when statistics began to depart sharply from the prediction - and August 18. In the same period, the health department reported 12,112 deaths from Covid-19.
Image: 123RF/DOLGACHOV

After watching the number of excess deaths decline for the fourth successive week, SA Medical Research Council (SAMRC) experts said on Wednesday that the end is in sight.

The SAMRC team has issued weekly reports about the number of deaths in excess of a prediction based on patterns of mortality in 2018 and 2019, combined with a lower number of deaths as a result of the national lockdown.

On Wednesday, estimated excess deaths from natural causes reached 39,087 for the period between May 6 — when statistics began to depart sharply from the prediction — and August 18. In the same period, the health department reported 12,112 deaths from Covid-19.

However, the SAMRC scientists said excess deaths were declining so rapidly in some provinces and cities, or had been declining for so long, that they were starting to approach the predicted number.

“After the number reaches or falls below the baseline, we will cease to accumulate excess deaths,” they said. “We will, however, continue to monitor the relationship between the total and predicted number of deaths.

“Not only will continued monitoring be necessary to confirm a possible ‘second wave’ but it may also signal [if the actual deaths remain consistently below the predicted number over time] the need for a revision of the baseline we used.”

The team from the SAMRC burden of disease research unit which compiles the weekly reports — Debbie Bradshaw, Ria Laubscher, Rob Dorrington, Pam Groenewald and Tom Moultrie — said deaths from all causes had exceeded the prediction by 2,408 in the week ending August 18.

In the same week, there were 1,513 confirmed Covid-19 deaths, according to the health department.

After beginning to climb in early May, the number of excess deaths from natural causes among people aged one to 59 peaked in mid-July and is now approaching predicted levels.
After beginning to climb in early May, the number of excess deaths from natural causes among people aged one to 59 peaked in mid-July and is now approaching predicted levels.
Image: SA Medical Research Council

“The number of deaths from natural causes has continued to drop from the peak experienced during the week starting July 15. The number remains significantly higher than the predicted number for people 1-59 years and for people 60+ years,” the SAMRC team said.

“In the period May 6 to August 18, there has been an excess of 39,087 deaths from natural causes of people 1+ year old when using the revised base accounting for lower mortality during lockdown. For people 1-59 years, the excess is 12,771 and 26,292 for people 60+ years.”

The number of excess deaths among over-60s is declining steadily.
The number of excess deaths among over-60s is declining steadily.
Image: SA Medical Research Council

Cape Town continues to be the metro with the most excess deaths (3,741) but Johannesburg's excess deaths increased to 183 in the week ending August 18, bringing its total to 3,665.

How excess deaths from natural causes have affected provinces and metros since they started to depart from predicted levels in each area.
How excess deaths from natural causes have affected provinces and metros since they started to depart from predicted levels in each area.
Image: SA Medical Research Council

“The number of natural deaths in Nelson Mandela Bay metro and eThekwini have reached the upper prediction bound while the number in the city of Cape Town has fallen below the upper bound and is approaching the predicted value,” said the council.

“The numbers of weekly deaths from natural causes declined in all provinces but are all significantly higher than predicted, except for Limpopo and the Western Cape.”

Compared with the predicted number of natural deaths in the week ending August 18, the increases were:

  • Free State, 74%;
  • Gauteng, 44%;
  • North West, 36%;
  • Mpumalanga, 35%;
  • Northern Cape, 33%;
  • KwaZulu-Natal, 31%;
  • Eastern Cape, 26%;
  • Western Cape, 12%; and
  • Limpopo, 6%.

The council explained on Wednesday that estimating the number of Covid-19 and collateral deaths — those arising as a result of the affect of the epidemic and its effect on the provision of health care — was not straightforward.

“In developed countries this is measured using the extent that deaths due to all causes exceed what might be expected had the epidemic not occurred. For various reasons this is not the best way of estimating Covid and collateral deaths in SA, particularly in the early weeks of the epidemic,” they said.

“For a start, unnatural deaths are a higher proportion of all deaths in SA than in developed countries, and reduced significantly as economic and social activity became more limited, particularly in lockdown level 5. Thus, it was decided to limit our focus to the numbers of deaths from natural causes.

“Further, it was observed that during lockdown the number of deaths due to natural causes fell well below the predicted number. The weekly numbers tracked the lower prediction interval, probably due to the affect of lockdown and non-pharmaceutical interventions on the spread of non-Covid communicable diseases.

“Thus, using the predicted value as the base would understate the affect of the Covid-19 epidemic and identify no Covid or collateral deaths until the week starting June 17, when there were already more than 2,000 confirmed Covid deaths.

“It was therefore decided to take as the expected number of deaths a proportion of the predicted number, that matched the proportion in the week before week when there was a clear rapid increase in numbers [the week starting on May 6 for Cape Town, Western Cape and nationally]. The number of cumulative Covid and collateral cases was set to the number of confirmed cases for that week.

The 10 countries with the most Covid-19 infections on August 26.
The 10 countries with the most Covid-19 infections on August 26.
Image: worldometers.info

“Continuing to use this adjusted base until late June is supported by the observation that the number of natural deaths in the provinces less affected by Covid-19 continued to track lower than predicted numbers through to the end of June.

“However, after the emergence of Covid-19 in most of the provinces, and with uncertainty about the counterfactual after the seasonal winter peak in numbers of deaths, the baseline is transitioning to the predicted number of deaths by next week.”

Though most excess deaths were probably due to Covid-19, “it is unclear at this stage what proportion are Covid deaths and what proportion are collateral deaths”.

SA's official Covid-19 death toll, announced on Tuesday night by health minister Zweli Mkhize, is 13,308. The country's mortality rate from the virus is 2.17%, compared with a global average of 3.42%.

Over the past 14 days, an average of 183 South Africans a day have died from Covid-19.

SA has the 13th highest number of Covid-19 deaths. Chile, with 10,958, is in 14th position. The US remains in first place with 182,430 deaths and a mortality rate of 3.06%.

TimesLIVE


subscribe Just R20 for the first month. Support independent journalism by subscribing to our digital news package.
Subscribe now