Change the world


“When is the peak?” This is a question being continuously explored, with various sets of data used to model predictions of when South Africa’s COVID-19 infections will reach their peak.

Since the COVID-19 outbreak, initially in China and spreading throughout the world, there have been efforts to model and visualize the geographic spread of the novel coronavirus.

Some of the models have been slightly more accurate in their predictions, while some not. Models tend to give a ‘best guess estimate’ of the future, while actual data, although it has its limitations, is based on what is actually known, seen and lived.

As at 22 June 2020, South Africa has 101 590 confirmed COVID-19 cases, of which 48 146 remain active. The Eastern and Western Cape provinces have respectively recorded 16 895 and 52 554 cases, of which 8 860 and 16 596, respectively, remain active.

Contributing to the ongoing data modelling discussions, Nelson Mandela University is hosting a webinar to unpack available COVID-19 models and analyse the data used therein. This, in a bid to get closer to answering the question “when is the peak?”

Predictions made through data modelling have been key to the national response to the pandemic. They have helped provide predictions of the scale of the pandemic’s impact relative to the country’s health system capacity, while also being used to gauge the success of government’s intervention strategies.

Webinar panellists include the University’s executive deans of science and health sciences, Professors Azwinndini Muronga and Lungile Pepeta, respectively, and director for the institution’s Centre for Community Technologies, Prof Darelle van Greunen.

“When dealing with COVID-19, South Africa needs to be guided by models in its planning, and should rely on actual data in implementing its response to the pandemic,” says Prof van Greunen.

“Models are based on assumptions about what should be included and excluded from reality.  The question is what is our reality?  Do we know enough about the behaviour of COVID-19?

“Models are often used as a scare tactic or a means to change behaviour. Rather than debating the accuracy of the model, we need to ask whether the model in use is good enough for the specific scenario.”

It has become apparent that some of the available models are based on specific regional contexts and may not be applicable to every country.

“There has been an emergence of different types of epidemiological models reported on the web, media and academic journals worldwide, trying to model the coronavirus pandemic using data primarily from John Hopkins University,” says Prof Muronga.

“Some people have been making predictions; some trying to extrapolate models developed for China to other countries, making claims as to how long the epidemic was going to last and what the peak was going to be.”

You are invited to join the webinar as the professors, moderated by the institution’s Research Director Dr Kwezi Mzilikazi, unpack the country’s COVID-19 models.

Date: 24 June 2020

Time: 15:00 – 17:00

View the webinar recording and presentations

Contact information
Mrs Allison Olivier
Deputy Director: Marketing
Tel: 041 504 2142