Discovery data reveals insights into South Africa’s experience of COVID-19
We have spent much time investigating the pandemic’s national and regional peaks, and also considering which interventions have saved lives and helped the country to navigate COVID-19.
We recently published our unique new insights on the progress of the COVID-19 pandemic in South Africa and some of our learnings from the global experience.
Advanced disease risk modelling was applied to data from 19 medical schemes administered by Discovery Health, representing about 7% of the South African population. This includes 3.5 million covered lives, with 79 000 confirmed cases of COVID-19 (12% of the national reported experience of the pandemic), 14 300 hospitalisations and tragically, more than 1 500 deaths.
Encouraging signs that South Africa has effectively navigated the peak of the pandemic
While the number of global COVID-19 cases continues to rise, there are definite signs that South Africa has effectively navigated the peak of the local epidemic.
Despite the younger average age of the South African population being a favourable risk mitigant, with the extent of crowded living conditions and complex comorbidities put us at risk of a severe experience of COVID-19. During the early phase of the global pandemic, there was much anxiety about how this would play out in our society. We can now look back, proud of the fact that despite our relatively large epidemic, our healthcare system coped well, with satisfactory health outcomes and low mortality rates. These dynamics delayed the timing of the national peak, created staggered regional outbreaks and contained regional attack rates (number of new COVID-19 cases per 100,000 lives) with longer, flatter peaks.
In countries like the United Kingdom and the United States of America, almost two months passed between the first confirmed case of COVID-19 and full lockdown. South Africa’s early lockdown came only 18 days after our first confirmed case of COVID-19 (on 6 March 2020).
The clinical and epidemiological benefits are demonstrable and have translated into lower fatality rates than many countries. While we remain concerned about the dire economic impact of the lockdown, there are strong epidemiological reasons to thank our effective national response.
Through Discovery Health’s very detailed actuarial modelling on mortality rate trends, we estimate that by the end of 2020, up to 16 000 South African lives will have been saved as a result of our healthcare system being well prepared, our ability to deploy improved medical treatment protocols and our lockdown related non-pharmaceutical interventions which changed the pattern of COVID-19 spread through South Africa. This is an outstanding achievement.
Non-pharmaceutical interventions such as limiting people’s movement definitively contributed to flattening the curve, effectively delaying the peak of the pandemic and consequently giving adequate time for the healthcare system and infrastructure to be fully prepared. As an example, this delayed peak allowed our country to import sufficient personal protective equipment (PPE) to better protect healthcare professionals at the front-line of care, and also to build and repurpose hospital beds to avail additional capacity.
We also see compelling evidence that enhanced hygiene practices, wearing of masks, proper hand washing and social distancing have substantially limited the spread of other seasonal respiratory tract infections reducing the burden on the healthcare system. Our data demonstrates an innocuous influenza season this year, with more than a 50% reduction in influenza demonstrated by pathology tests, associated out of hospital visits and almost no influenza related pneumonia admissions.
We also benefitted from global learnings around evidence-based treatment modalities that were documented by other countries who were late to lock down. The first peak in Europe and the United States occurred in April / May, whilst the South African epidemic’s peak was delayed to July as a result of lockdown.
These treatment insights centre predominantly around the use of corticosteroids (cortisone), high flow nasal oxygen and anti-coagulants in severe COVID-19. We also had time to learn from others’ hardship that early ventilation of COVID-19 pneumonia is associated with poor outcomes, and consequently adopted treatment modalities and delayed ventilation wherever possible. Discovery Health administered scheme members who benefited from these treatments experienced a 41% reduction in the average length of hospital admissions and a 44% reduction in ICU admission or ventilation. The data also shows a 25% improvement in mortality for those COVID patients admitted to ICU, with the change occurring around mid-June.
We must reverse the apparent decline in general health-seeking behaviour
Our data shows that a serious, unexpected consequence of the pandemic is a general reduction in consumers seeking healthcare services, for important underlying conditions. The data demonstrates a clear reluctance by our members - especially among those in high-risk groups - to consult their doctors. Whilst we acknowledge the higher risks of contracting COVID-19 and other hospital acquired infections in all healthcare facilities, there is be greater harm in by avoidance of healthcare engagement by those living with chronic illness, or those experiencing a new serious healthcare concerns.
Compared to the same period last year, there have been 30% fewer GP consultations among medical scheme members who live with chronic illness. We also note a 51% decline in members registering for treatment for depression and a 42% decrease in wellness tests and healthcare screening activities. This means fewer diagnoses of underlying chronic illness. Most concerning of all is that we see a 51% reduction in breast-screening mammograms performed with a consequent 44% reduction in breast cancer diagnoses. Unfortunately, it seems clear that these cancers are still out there, but are simply undiagnosed at this time. These patients will present in a later, more advanced stage of disease, requiring more complex treatment and possibly facing poorer outcomes.
As such, we are proactively working with doctors so they reach out to high-risk members, to ensure appropriate ongoing care. We made digital healthcare platforms available to members prior to the pandemic and have been encouraged by the rapid increase in the adoption of virtual consultations (telemedicine) across specialties.
It’s possible that one quarter of South Africans have been infected with COVID-19
Records show that by 20 September, South Africa had conducted over 4 million tests, and experienced almost 660 000 confirmed COVID-19 infections, with almost 16 000 tragic deaths. Our country has a case fatality rate of 2.4% from COVID-19.
COVID-19 is still a new disease, only having been described for the first time in January this year. We are continually learning more it. What is confounding experts is the true penetration of COVID-19, considering how highly contagious it is. Global evidence shows that an extraordinary number of cases remain undiagnosed with the estimated rate of asymptomatic cases varying from 40% to 75% in various studies.
The South African Medical Research Council (SAMRC) regularly publishes information on deaths in the country. Their reliable longitudinal data shows an inordinate increase in unexpected natural deaths over the period of the South African epidemic, compared to historical data. These so-called “excess deaths” are explicitly linked to natural causes (i.e. are not trauma-related) but not definitively identified on pathology testing as COVID-19 related. It is however almost certain that a high proportion of these excess deaths are attributable to COVID-19.
These excess deaths are outside of the confidence intervals of what the usual natural deaths in the period could possibly be, and the timing of the excess deaths concomitant exactly with the COVID-19 regional outbreaks makes it unequivocal. The fact that weekly excess deaths increased sharply between early June and mid-July, and have been decreasing significantly every week since, confirms that the vast majority of these excess deaths arose as a result of COVID-19. The effects of poverty, or delayed access to treatment for non-COVID related conditions, would not result in such a pattern in excess deaths. This situation is not unusual and certainly not particular to South Africa. Many countries across the world, including where there are highly developed healthcare systems like in the United Kingdom, have reported excess natural deaths during the global pandemic - deaths originally not directly attributed to COVID-19, but retrospectively recognised with high confidence as COVID-19 related.
The extent of excess deaths, as reported by the SAMRC, gives us a sense of the real impact of COVID-19 in South Africa, and allows us also to work backwards to understand the true extent of the infections.
Based on SAMRC data there had been just over 44 467 excess natural deaths as of 8 September. For purposes of this analysis, we have conservatively attributed 90% of these excess deaths to COVID-19 implying that 40 000 deaths could be the real mortality impact of COVID-19 in South Africa - up to 2.7 times higher than that reported from definitive COVID-19 deaths demonstrated by the pathology testing.
Drawing on data from the Imperial College, the Lancet Journal publication and Discovery Health’s member experience, we apply an infection mortality rate of 0,29 to these 40 000 deaths to work backwards to calculate the total infected population. It is important to note that the infection mortality rate in South Africa changed, and favourably reduced by 25% in early June 2020, following the introduction of new treatment modalities as described earlier in this article. On this basis, we estimate that approximately 13.8 million South Africans have been infected with COVID-19. This represents an infection rate of about 23% for the population as a whole, and implies that COVID-19 has infected approximately one third of all South African adults.
We are given additional comfort around the veracity of this analysis and the data-extrapolation by comparison with antibody screening data recently published in South Africa. UCT academics together with the City of Cape Town Metro recently published their research demonstrating a 37% - 42% positive sero-prevalence of antibodies in patients screened at their ante-natal and HIV clinics in the Cape Metro region. Simply, this implies that around 40% of people screened at these clinics have antibodies to COVID-19, indicating a previous COVID-19 infection. The high antibody prevalence in this cohort of economically active, younger people, provides a convergent and useful data point that supports our data extrapolation of about one third of adult South Africans having been infected. Our analysis also aligns with recent sero-prevalence studies and subsequent statements by Minister Zweli Mkhize and Prof Shabhir Madhi (on the Ministerial Advisory Committee), that the National Department of Health believes about 12 million South Africans have been infected with COVID-19.
Data reveals pattern of COVID-19 transmission rates
Discovery’s data shows infection rates among members aged 20 to 40 years old – economically active people – are 1.7 times higher than for people aged 60 and over. Within this group, the highest infection rates are among those working in the healthcare and servicing sector, where face-to-face interaction and direct exposure is highest. As expected, older members, those with low Vitality engagement (high BMI and low activity levels,) and those living with chronic illness face the highest risk of severe COVID-19 illness. And, in children, COVID-19 is often undiagnosed as its presentation is almost always asymptomatic or very mild.
Discovery’s advanced disease risk modelling also finds that:
- People over age 60 have between a 22% and 34% elevated risk of being admitted to hospital on contracting COVID-19, relative to similar, 40-year-old people. These findings align to international evidence.
- The most prevalent chronic conditions amongst Discovery Health Medical Scheme members are diabetes, hypertension and hypercholesterolemia. These conditions increase the risk of COVID-19 related hospitalisation by 27%, 11% and 7% respectively.
- The more chronic conditions a person lives with, the higher their COVID-19 hospitalisation risk with members experiencing an elevated risk of admission of between 20% (for one condition) and 86% (for six or more chronic conditions).
However, we can offset the elevated risks of severe COVID-19 illness associated with aging and lifestyle-related chronic conditions through healthy living. In May this year, Discovery released the results of its Vitality Resilience Index showing that increased physical activity (assessed through engagement in Vitality) could lower the relative risk of being hospitalised to treat COVID-19.
Discovery’s updated analysis adds to this picture as we see that engaged Vitality members experience, on average, 37% lower mortality risk than non-engaged members.
Engagement in Vitality can, in fact:
- Offset the elevated COVID-19 risk associated with aging. A 65-year-old male, with no chronic conditions, who exercises for half an hour at least four times a week, has the same mortality risk from COVID-19 as a 45-year old who exercises once a week.
- Offset the elevated COVID-19 risk associated with lifestyle-related chronic conditions. A 45-year-old male living with hypertension, but highly engaged in healthy behaviours, has the same risk as an unengaged 45-year-old male without chronic conditions.
Identifying high risk members and enabling remote monitoring improves outcomes
At Discovery Health we are doing all we can to protect high-risk members from severe outcomes when they contract COVID-19. To date we have funded care at home with remote monitoring for 7564 proactively identified high-risk members who tested positive for COVID-19. This included the deployment of pulse oximeters. These devices allow for regular checking of oxygen saturation levels to ensure the early detection of silent hypoxia, an unnoticed reduction in oxygen levels over time and a very dangerous aspect of COVID-19. Early recognition of deteriorating oxygen saturation allows proactive intervention and more aggressive treatment of COVID-19 with corticosteroids (cortisone), anti-coagulants and home oxygen.
Discovery Health Medical Scheme’s unique enablement of this care at home has lowered the risk of hospital admission in these individuals by 27%, decreasing case severity by 10%, and improving mortality outcomes by an astounding 38%. We are really proud of these results as this preventive care has directly saved lives.
Five key actions to define the months ahead
Our positive reflection on the epidemiological and clinical aspects of the progress of the pandemic following our country’s early lockdown must not detract from the very challenging economic impacts as evidenced in the most recent GDP data. We must take the necessary steps to reduce our risk of experiencing a second wave of infection in our country and accelerate the path to re-opening our country fully, and rebuilding our economy.
Our research demonstrates that there are several proactive steps we should all take to curb a second wave of infection, as we move into COVID-19 alert level 1. We can do this through:
- Continued adherence to evidence-based, non-pharmaceutical interventions such as social distancing, mask-wearing and general hygiene protocols.
- Embracing technologies designed to optimise our health in the time of COVID-19 - from virtual doctors’ consultations to use of the National Department of Health’s recently launched Bluetooth contact-tracing app, COVID Alert SA. I recently published a LinkedIn post explaining the way in which this app allows us to understand our risk of exposure and to keep others safe. Thorough contact tracing is critical to our safe return to work and reopening of the economy. Discovery has played a pro bono role in the development of this app in partnership with our National Department of Health.
- Partnerships with family physicians to enact remote monitoring protocols to protect high-risk people who contract COVID-19, and significantly improve their healthcare outcomes.
- Engaging in physical activity to live a healthy lifestyle and to reduce our risk profile should we contract COVID-19.
- Accessing healthcare responsibly when we need it and ensuring adequate routine care for chronic illness. We must also complete our regular cancer screening tests.
REFERENCES
Prevalence of Asymptomatic SARS-CoV-2 Infection, Daniel P. Oran, AM, Eric J. Topol, MD (2020)
https://www.samrc.ac.za/reports/report-weekly-deaths-south-africa
https://www.thelancet.com/journals/laninf/article/PIIS1473-3099(20)30243-7/fulltext
CEO & Founder @ PricklyPairStudios - AI Agents | Transforming Businesses with AI | AITinkerers Cape Town
1 年Cu all too h h. H hi v
Senior Cutover Manager & Programme & Portfolio Champion
4 年Time to be Pro active !
Strategy Manager @ Corporate Wellness Consulting | Healthcare Service Delivery
4 年Excellent insights Ryan, and well articulated. Thank you for sharing.
Talent management | Leadership development | Executive coaching | Culture | Engagement & retention | Business results
4 年Ryan Always doing what you are #GREAT at - pulling all the information together. Wellness mindset in its totality is going to be key to health in the future. Tiaan - I know how much you enjoy seeing information integrated - would love your insight!
Drumming ??4 Peace & Healing | President Barack Obama Lifetime Achievement Award | Grammy Award Artist | A Multifacet Maestro, Harmonious Force in Business | Music | Education & Promoting Positive Behavior Workplace
4 年Good on you guys Ryan