Will South Africa experience a third wave of COVID-19 infection? If so, when and how severely?

South Africa experienced the first wave of COVID-19 starting in April 2020 with the Western Cape affected first, and abating by September 2020, with Gauteng being the last province to settle. A devastating second wave followed three months later, linked to the migratory patterns of the holiday period, and as the evidence demonstrates - spurred on by super spreader events in late November and early December. Now we face concern that a third wave could hit us sometime after Easter - coinciding with Autumn and potentially compounding the morbidity of the annual flu season.

In recent weeks, COVID-19 cases have spiked across Europe to above the risk threshold of 10 new infections per 100 000 people per day. Several European countries - including the Netherlands, Italy, Germany, Greece, Poland, and France - have returned to lockdown measures as third waves of infection have set in, fuelled by the highly transmissible B117 variant (originally detected in Britain). Newly imposed lockdown restrictions in Italy, imposed from 15 March, require residents to stay strictly at home over the Easter holiday.

This global experience is concerning to us in South Africa, and increases our concern of a third wave of COVID-19. Its scale and timing are unclear, and depend on our behaviour. We are in control of the outcome of this infection trajectory. It is up to us and, as per the unequivocal evidence that is now published and universally supported, will be determined by:

• Continued adherence to preventive measures including mask-wearing, hand hygiene and social distancing.
• Avoidance of gatherings, particularly over the religious holidays. We must ensure no super-spreader events are allowed to happen - not only very large gatherings of many tens of people, but also smaller gatherings which become spreader events.
• Accelerating vaccination, especially to the most at-risk

Reinfections are a significant risk, one supported by recent data insights

Over the course of the first and second waves of COVID-19 infection, over 1.5 million people in South Africa have been diagnosed with COVID-19 and over 52 000 COVID-19 related deaths have been recorded. However, given over 80% of infections are either asymptomatic or mildly symptomatic, these figures significantly understate the full extent of the outbreak.

The South African Medical Research Council has recorded over 150 000 excess natural deaths since March 2020, suggesting that the COVID-19 death toll is significantly higher than what has been recorded. Actuarial analysis, based on this data which is further corroborated by recent seroprevalence studies, suggests that more than 55% of the population has already been infected with COVID-19.

While this suggests that a large proportion of the population will now have some degree of COVID-19 immunity, latest scientific evidence shows that lasting immunity against reinfection is not assured, in simple terms. Despite having been infected previously, there is a strong possibility an individual could be reinfected for a second time. This could be aggravated by the continued emergence of COVID-19 mutation i.e. variants that are capable of immune escape as we have seen happening in South Africa and around the world.

Our Discovery Health data from wave 1 and 2 demonstrates that during wave 2, combined with the probability of contact to infection, one in three Discovery clients with an earlier infection, had a chance of reinfection. Some of these reinfected clients required hospitalisation during the second infection.

What do we currently know?

At Discovery Health, big data and sophisticated analytics help us quantify risk, and steer much of our decision-making to further protect the lives of our customers.

Emerging from our analysis of 3.7 million medical scheme beneficiaries, being 7% of the country’s population, we have derived in-depth understanding of the pattern of COVID-19 infections. The second wave of COVID-19 occurring over the December and January holiday period, was quite different to the first wave, in many respects:

• A much higher rate of spread and number of infections - 86% higher daily new COVID-19 infections and 70% higher daily COVID-19 hospital admissions at the peak of the second wave
• More severe disease, with a higher proportion of infected people requiring hospital admission - a 10% higher morbidity rate
• The combination of the above two factors - more infections with worse morbidity - led to 70% higher hospital admissions in absolute terms, with over 520 Discovery Health members admitted to hospital every day at the peak of the second wave.

There is however some good news in this regard. Treatment advances following the first wave meant that, with enhanced care, medical scheme members spent 37% less time in hospital per admission, on average. Additionally, while more people tragically died as a consequence of more infections, overall mortality rates did not increase from first to second wave. For clarity, the case fatality rate - or the proportion of deaths from reported positive cases - did not change during the second wave.

Three third wave scenarios: Why the Easter break could be a tipping point

Our actuarial analysis forecasts the probability and impact of a third wave in South Africa in three scenarios:

• Scenario 1: “Second wave run-off” resulting in 9 000 new COVID-19 related deaths until the end of the year. As a result of prudent and diligent behaviour, the avoidance of super-spreader events, good personal hygiene and mask adherence, we maintain the current dynamic of lowered infection rates. In effect, we prevent a third wave.
• Scenario 2: “Easter super-spreader events, in the absence of vaccines” resulting in an additional 76 000 COVID-19 related deaths. This wave starts in April and peaks in June or July. Easter super-spreader events drive high numbers of infections (including reinfections), leading to a third wave infection peak potentially larger than our first wave peak, but not quite as high as our second wave peak.
• Scenario 3: “Easter super-spreader events, accelerated vaccine rollout” resulting in an additional 35 000 COVID-19 related deaths. This wave also starts in April and peaks in June or July. However, acceleration of our vaccine rollout process from April onwards allows us to vaccinate high-risk groups by mid-Winter, resulting in higher levels of population immunity and lower consequent morbidity and mortality. Should the vaccine rollout only commence in June, and the Easter period turn out to feature super-spreader events, potential excess deaths increase to 58 000.

What determines which scenario wins out? Five paths for South Africa.

We cannot let our guard down

Infections have, in the case of the past two waves, steadily increased at points where lockdown restrictions have eased. The recent easing of restrictions and move to alert level 1 – just weeks ahead of the Easter break – may encourage people to feel more relaxed about the threat of exposure to COVID-19 and decrease their adherence to the preventive measures that curb the spread of infection. Until our vaccine rollout programme brings us to population immunity we must never assume that it’s safe to let down our guard.

Mobility – especially at night – puts us at risk of infection

Google mobility trends data, correlating with our Discovery Vitality Drive tracking data, shows that after months of stay-at-home measures, there was much higher “social mobility”, particularly at night, in the three weeks leading up to the peak of infections of the second wave. In fact, across the country, December 2020 mobility returned to levels last seen pre-COVID, during early 2020.

Super-spreader events must be avoided

As mentioned, right now there is great concern that Easter related social events - from religious to family gatherings, travel to popular and busy holiday destinations - will fuel COVID-19 super-spreader events. This concern is justified. We now know that our second wave of infection was fuelled in large part by social and post-matric celebratory super-spreader events that took place all over the country, and particularly in KwaZulu-Natal, over the December break. The consequence of these gatherings was a demonstrable spike - a 27% increase - in infections in the younger populations of the Discovery Health member base, infected during the second wave (data to 1 Feb 2021).

Our data also show that young members who flew to these post matric holiday destinations faced a 16 times higher risk of contracting COVID-19 in the first week of December than those who did not. It is important to keep in mind that about 80% of COVID-19 cases are entirely asymptomatic, and an asymptomatic infection is most likely in young healthy people.

Vigilance over variants

In large part, the increased scale of the second wave compared to the first, was driven by the emergence of a new variant of the SARS-CoV-2 virus in South Africa in November 2020 – called the 501Y.V2 variant. This form of the virus appears to have been more transmissible than strains of the virus that circulated in the first wave of infection. We have also demonstrated a higher proportion of hospital admissions during the peak prevalence of this variant.

As with any virus, the SARS-COV-2 virus (which causes COVID-19) is constantly changing. Viruses can mutate every time they reproduce themselves in their host (the infected person). So, new variants may emerge at any time, quite unpredictably, and even challenge the efficacy of vaccines. Preventing viral spread between people limits opportunities for the SARS-CoV-2 virus to mutate and also protects us from infection with any new, highly transmissible variants that may not yet be documented.

We must therefore continue to remain vigilant and in so doing, limit the likelihood of future variants and, in particular, variants of concern with increased infectivity and lethality.

Accelerating access to vaccines

My view is the COVID-19 vaccine rollout is the single most important public health intervention of the century, both saving lives and accelerating economic recovery. Notwithstanding the complexity and scale of the national COVID-19 vaccine programme, we simply must move as quickly as possible to accelerate vaccination for all South Africans, especially for those at high risk. People over the age of 60 years and those living with multiple co-morbidities, must be prioritised. Our data demonstrate unequivocally that these at-risk populations experience much more severe manifestations of the disease.

Let’s do all we can to make Scenario 1 a reality

It is in our hands to save lives. Our personal behaviour, and that of the people around us, will be a large determinant of which Scenario our country experiences. The loss of life described in scenario 3 is tragic and devastating. Let’s work together, keep each other honest and remain prudent as we try our very best as a country to experience Scenario 1, and avoid a severe third wave.

ENDS

For media queries, contact Nthabiseng Chapeshemano on nthabisengc@discovery.co.za / 084 037 1951 or Karishma Jivan on karishmaj2@discovery.co.za / 071 048 9009.