Excess Deaths Reveal Vulnerabilities Driving the COVID-19 Burden in South African Provinces

Photo: World Bank Photo Collection

South Africa is currently grappling with a painful second wave of COVID-19, heightened by the discovery of a highly contagious variant in December. As we attempt to assess the distribution of this burden in South Africa, we must consider how reported COVID-19 statistics — total cases and deaths — likely underestimate the true impact of the virus due to limited continent-wide testing. As a result, we should base our analysis of the pandemic in South Africa on alternative measures, such as excess deaths.

As cases surge and the number of excess deaths peaks, dropping only recently, South Africans face fear and uncertainty. This feeling was heightened after the recent passing of Jackson Mthembu, the public face of South Africa’s fight against COVID-19, due to virus complications. To help South African stakeholders — national policymakers, provincial government officials, business owners — in their efforts to combat the virus, we examined the relationship between South Africa’s excess deaths and Surgo’s Africa COVID-19 Community Vulnerability Index (Africa CCVI) to find out what may be driving these deaths, so that they can develop more precise responses. Our key takeaway: provinces that are fragile (more likely to host refugee sites or face food insecurity) or home to older age populations are more likely to experience fatalities.

Let’s dig deeper into this finding.

Connecting Observed Data to Vulnerability

South Africa is the only African country currently reporting excess deaths across its provinces. To understand why some provinces have suffered more than others and assess which ones are being hit hard during the second wave of the pandemic, we examined relationships between the South African Medical Research Council’s (SAMRC) reported excess deaths per 100K and province-level vulnerability from the Africa CCVI. The Africa CCVI ranks 48 countries on seven types of vulnerabilities, shown here for South Africa:

Figure 1: Overall and Sub-theme Vulnerabilities for South Africa, accessible on the Africa CCVI website.

At this point in the pandemic, we only see a consistent relationship between vulnerability and excess deaths per 100K for Age and Fragility factors. This is congruent with the widespread notion that older individuals are more vulnerable to the virus, which is further compounded when a region is fragile.

In Figure 2, we observe that provinces with higher Age vulnerability (proportion of population aged 65 or older) reported higher cumulative excess deaths in the first wave of the pandemic (by October 11, 2020), as seen in the Eastern Cape, Free State, and KwaZulu-Natal. This data suggests that provinces highly vulnerable on the Old Age theme might experience higher COVID-19 related deaths during the second wave, which appears to be occurring — most notably in Limpopo, the Eastern Cape, and KwaZulu-Natal — provinces that score highly with respect to old age vulnerability: 1.00, 0.875, and 0.50 (on a scale of 0 to 1), respectively.

Figure 2: Time Series of SAMRC Weekly Excess Deaths per 100k from May 3 to February 6, 2021, stratified by province. Line colors are determined by Age Vulnerability. The dotted line at October 11, 2020 separates the first and second waves.

Separately, Figure 3 indicates that in the first wave of the pandemic, provinces with higher Fragility (refugee camps, food insecurity) reported higher cumulative SAMRC excess deaths per 100K, as observed in the Eastern Cape, Northern Cape, Free State, and Gauteng. Among the provinces of South Africa, the number of UNHCR population of concern sites and percentage of children stunted (below -2 standard deviations of height for age according to the WHO) were the main drivers of fragility in the first wave.

As of late, some of these regions experienced an increase in excess deaths again, such as in the Eastern Cape. This surge reflects the discovery of the new, highly contagious COVID-19 variant identified by South African scientists in the state. The discovery, coupled with the fact that the Eastern Cape houses three refugee urban locations with over half a million people, could explain why the province has experienced a dramatic peak in excess deaths recently.

This data indicates that if the virus spreads to areas that are fragile, they are more likely to experience higher COVID-19 fatality rates. Such provinces are burdened by multiple crises, which complicates efforts to allocate resources and promote the right care-seeking behaviors. For example, South Africa Police Minister Bheki Cele recently identified Eastern Cape and Gauteng among provinces that have experienced high levels of both crime and COVID-19 infections during the pandemic.

Figure 3: Scatter Plot of Cumulative South African Medical Research Council’s (SAMRC) Weekly Excess Deaths per 100k versus Age Vulnerability by October 11, 2020. Point colors are determined by Fragility Vulnerability.

Fragile regions are likely to experience secondary impacts from COVID-19 which need to be mitigated. In Free State, South Africa, for example, not only is there a refugee urban location in Bloemfontein, but the National School Nutrition Program also came to a halt for three months during South Africa’s stringent lockdown. It was reinstated in July following legal action. Since then, uptake of meals has improved; however, 1.2 million learners (of 9.6 million) were not receiving their meals as of November 2020, according to Section27 communications officer Julia Chaskalson. This province, housing a UNHCR population of concern site and facing food insecurity, experienced the second-highest peak of excess deaths in the first wave, suggesting that they will once again be susceptible during the second wave.

Taking these observations into account, we must consider how to best support the provinces housing individuals who fall into these categories — individuals aged 65 or older, refugees without easy access to healthcare who cannot social distance, and children who face food insecurity — during the second wave, in an attempt to ease the burden of this pandemic.

Identifying Vulnerable Regions

The Africa CCVI does not predict how and where the virus will spread. Rather, it aims to capture the impact of the pandemic on regions when it does reach those areas. As such, there may be regions that are vulnerable that have been spared in the first wave. They should not fall off our radar when we plan ahead. In Figure 4, we highlight South African provinces that are vulnerable due to old age and fragility. Among those that did not experience the greatest number of excess deaths in the first wave, Limpopo is the most age-vulnerable, and Gauteng and Mpumalanga are the most fragile.

Figure 4: Age and Fragility Vulnerability in South Africa, according to the A-CCVI.

Recent news reports suggest that Gauteng is now the second-most affected province in South Africa in terms of active cases after KwaZulu-Natal. Over 30% of people who get tested for COVID-19 in Gauteng test positive, and the province has already surpassed its peak number of infections from the first wave. Gauteng houses over 2.3 million people in informal settlements and two refugee urban locations, in Johannesburg and Pretoria, suggesting that its excess deaths could potentially fall in line with those of the other provinces, the Eastern Cape and Free State.

In addition, KwaZulu-Natal and Limpopo, two age-vulnerable states, have experienced dramatic peaks in excess deaths during the second wave, even though they didn’t face the greatest number of excess deaths in the first wave (Fig. 2). Although it may be too late to prevent the second wave in these provinces, we can use these insights to guide an effective policy-response that will flatten the curve once again.

Using Our Insights to Target the Response

With the second wave of COVID-19 in full force, South Africa in particular is suffering immensely. The new coronavirus variant has resulted in an overwhelming surge of cases, scaring countries globally and pressuring policy-makers who are afraid of the negative economic impacts of another stringent lockdown. In addition, the race for the acquisition of COVID-19 vaccines could result in Africa not receiving an adequate amount of doses soon enough to meet this rapid spread. As the intensity of the second wave inevitably extends beyond South Africa, and with governments hesitant to institute strict policies, the consequences could be detrimental.

To mitigate the impact of the pandemic, governments must tailor their response at a sub-national level, based on vulnerability. Our Africa CCVI dashboard (Fig. 5) allows policy-makers to see where the most at-risk regions exist within their respective nations.

Figure 5: Visualization of overall vulnerability scores covering 751 regions across 48 countries, accessible on the Africa CCVI website.

While the WHO has deployed technical experts in Eastern Cape, KwaZulu Natal, North West, Free State, and Gauteng to address the pandemic in South Africa, local policy-makers must consider implementing additional measures specific to their particular vulnerabilities. They could also focus on provinces that the WHO hasn’t targeted recently but are now suffering according to our analysis, such as Limpopo with its vulnerable older population.

Knowing the most at-risk provinces in South Africa that are vulnerable specifically due to their age structure and extent of refugee sites and food insecurity, government officials should consider implementing measures that won’t compound the secondary effects of these pain-points.

Gauteng, for example, has increased their hot-spot tracking, contact tracing, testing, and isolation efforts in addition to upscaling their health-care facilities, in an attempt to avoid the challenging effects of instating a stringent lockdown. These strategic efforts may have helped reduce Gauteng’s peak of excess deaths driven by the second wave. Areas that are fragile but also have a large elderly population must balance the pros and cons of lockdowns, as they can leave individuals without vital resources, such as healthcare and meals. On the other hand, it might also jeopardize those who are older and more at-risk if not implemented, especially when those older individuals often serve as heads of the household and primary caregivers.

If officials decide to implement another lockdown, which would most likely be the fastest route to flattening the newest curve, they must be ready to provide income, healthcare, and meal support, among others, to those who will suffer from the inability to receive such resources under normal conditions. Although doing so is not an easy task, it might help strike a balance between preventing rapid spread and preventing humanitarian crises and economic downfall, allowing nations to navigate the next few months of the pandemic until the vaccine ushers in much-needed relief.

Surgo Ventures contributors to this article include: Daniela Garcia, Bettina Hammer, Rahul Joseph, Anubhuti Mishra, and Peter Smittenaar.

We use all the tools available from behavioral science, data science, and artificial intelligence to unlock solutions that will save and improve people’s lives.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store