Yet another three scenarios for COVID-19 deaths in South Africa

I was invited to work with the modelling team at The Singular Group to develop three scenarios for the COVID-19 pandemic in South Africa. It is always interesting to work with a team that has been grappling with the same problem but from a different starting point and armed with a different tool set.

The three Singular scenarios are only intended to forecast the pandemic to the end of August, although I have shown the forecast for COVID-19 deaths unto the end of the year in the graph below:


The three scenarios differ in terms of three input assumptions:
  • What proportion of the population will be infected with the virus once the first wave of infections comes to an end in the Western Cape (input assumptions here were 13%, 16% and 20% for the best, base and worst case scenarios)
  • What proportion of the population in the other provinces will be infected with the virus once the first wave of infections comes to an end in those provinces (input assumptions here were 5%, 10% and 20% for the best, base and worst case scenarios)
  • What Infection Fatality Rate (IFR) should be used to fit the reported deaths in the model (average IFR's of 0.47%, 0.51% and 0.74% were used for the best, base and worst case scenarios)
The most important of the input assumptions is the proportion of the population in the other provinces that will be infected in the first wave.  The Singular team pointed out how the experience in Europe shows that the experience in other regions of a population falls far short of the experience in the region that is hardest hit by the pandemic.  

I hunted around for an analysis showing this and found the following graph by The Health Foundation.  The graph compares the 'excess deaths as a share of usual deaths' for the different regions in the UK, France, Italy and Spain.


The next few weeks will not only give us an indication of the impact of the first wave of COVID-19 infections in the Western Cape, but more importantly will tell us how similar (or not) the shape of the first wave of infections in some of the other provinces will be.

The NMG model was last updated on Wednesday, 24 June, and can be found at https://nmg-covid-19.sctechnology.co.za/.  The web tool will allow you to model each province separately, and shows the forecast for South Africa as an aggregate of the forecasts for each of the nine provinces. 


29 June 2020



Comments

Post a Comment

Popular posts from this blog

Update on COVID 19 tracking in SA

Image
An analysis of the weekly deaths from natural causes as reported by the SA Medical Research Council (SAMRC) shows evidence of elevated excess deaths over the winter months of 2022. The graph below shows the weekly deaths reported by SAMRC in excess of the average weekly deaths reported between 1 January 2020 and 15 March 2020.  The COVID-19 hospital deaths as reported by the National Institute for Communicable Diseases (NICD) would appear to follow a similar pattern to that shown by the SAMRC excess deaths but with a more muted 5th wave. Anecdotally, the number of COVID-19 cases appear to be on the rise in South Africa although COVID-19 cases are no longer reported on a regular basis (outside of a hospital setting).  The above graphs show that the COVID-19 disease severity is certainly not near the levels experienced in the first three waves. 4 November 2022
Read more

COVID-19 death toll increases in Gauteng

Image
The third wave shows no sign yet of reaching a peak in Gauteng and, equally as worrying, the reported COVID-19 deaths have gone past the second-wave peak in this province. The gradient of the third wave in the Western Cape would appear to indicate that the Delta variant has not yet spread beyond Gauteng. Reported COVID-19 deaths have gone past the second-wave peak in Gauteng which is of concern Reported COVID-19 hospital admissions in Gauteng are showing signs of plateauing at a level equivalent to the second-wave peak in that province. This is almost certainly due to having reached hospital capacity. 7 July 2021
Read more

The third wave in the Western Cape

Image
The daily reported COVID-19 infections in the Western Cape have overtaken those in Gauteng, and show no sign of having reached a peak. This pattern is also evident in the comparison of reported COVID-19 hospital admissions: We have adjusted the duration of the third wave modeling in the Western Cape to better fit the data.  4 August 2021
Read more