Detailed risk prediction is needed to combat Covid-19
Actions around specifically identified groups at risk of contracting the disease need to be made if SA is to reopen successfully
After months of lockdown, SA faces the immense challenge of safely reopening for business and restoring other vital activities, including education and government services. We now know a lot about who is at risk of Covid-19 infection and which activities and behaviours, in which environmental settings, put individuals and others in danger.
While our knowledge is as yet incomplete, we can broadly predict who is more likely to become severely ill, require hospital care, or die despite accessing such care. We know most infections will occur in enclosed spaces, especially in the home, while shopping, dining, working and on public transport. We know it is unlikely that we will see infections from well-spaced outdoor activities, but we do need to be concerned about transmission from people who are asymptomatic or pre-symptomatic.
While this risk prediction is better at defining risk for groups but is not very good at the individual level, we can use it to focus our attention on those most in danger of severe illness or death, while allowing the majority of South Africans to resume, with significant modification, many of their usual activities.
Prolonged suppression of economic activity and the diversion of the health system’s efforts away from other essential services cannot continue: the resulting harm would greatly exceed the direct effects of Covid-19. The question is how we can return to work, tend to other pressing health needs, and resume some aspects of normal life without compromising the Covid-19 response.
The SA government has a “risk-adjusted path” for reopening. The concept of risk adjustment is tied to a strategy to protect those people most at risk from Covid-19. It needs to be based on the evidence of who is affected and what works, and requires the close support of the SA public.
The experience of high-income countries points to the co-morbidities (pre-existing health conditions) most strongly associated with severe Covid -19: heart disease, non-asthmatic lung disease, kidney and liver disease, diabetes and obesity. Males are more affected than females. The dominant risk factor for hospitalisation and death in studies from abroad is older age, especially in the population over 60. A large proportion of these deaths, more than 50% in some regions, have been in residential facilities for the elderly.
Covid-19 has also disproportionately claimed the lives of ethnic minorities, probably due to social determinants of health (poverty, overcrowding, hygiene conditions, access to care, nutrition, and so on) rather than genetics, though the latter contribution is yet to be fully explored.
While factors such as the rate of asymptomatic infection and its role in transmission are still a puzzle, existing evidence provides a basis for determining the needs and possible actions for SA
Multiple reports confirm that children generally experience much milder forms of the illness or have no symptoms at all, and compared to adults are far less likely to spread the coronavirus. This provides comfort that children can safely return to school, but there is concern for the health of teachers and families, and those in their community “chain of transmission”.
Resumption of education is a high priority, and a return to school will release working family members from childcare duties at home.
Improving our understanding of the risk of serious harm from Covid-19 and the resulting need for hospital care in our population can provide guidance for individuals, the community and the government. In data on Covid-19 patients admitted to Western Cape public hospitals, those aged 60-69 had a 13-fold higher risk of death compared to those under 40.
The majority of deaths included individuals with one of four co-morbidities: high blood pressure, diabetes, HIV, and kidney disease. The results suggest a two to two-and-a-half times increase in risk of death in patients with HIV and/or TB. A two-fold increase might mean a 1%-2% risk of death in patients with HIV who contract the coronavirus, far less, perhaps, than was originally feared.
In private-sector data from Discovery Health, 2.4% of all reported cases died, with age similarly the dominant factor: patients aged 70-79 were more than five times more likely to die than those aged 50 or less, with chronic heart, lung and kidney disease, cancer and obesity also conferring risk.
Transmission risk of the coronavirus is now understood to be related primarily to droplet spread from person to person, through prolonged (15 minutes or more) close contact in indoor spaces, with a lesser role for spread by touching of contaminated surfaces. This evidence has major implications for a risk-based strategy.
While factors such as the rate of asymptomatic infection and its role in transmission are still a puzzle, existing evidence provides a basis for determining the needs and possible actions for SA. Until a vaccine or effective treatment arrives or herd immunity emerges four actions are needed:
- Establishing risk categories: Develop risk stratification based on local data. We know the elderly and those with obesity, diabetes, heart, lung and kidney disease, TB and HIV are most at risk. Those under 55-60 with no co-morbidities would be at much lower risk. The population needs more information and help to understand and self-identify who is most likely to get infected by the virus and who gets seriously ill.
- Describing individual and community behaviours for risk categories: high-risk residents of facilities for the elderly and people with risk-associated conditions should be protected by the community and the state and strongly encouraged to continue to shelter in place. They should be able to say to the community — “please protect me”. This protection could take the form of special prevention and care programmes, strict visitation policies, support from family/friends, and “pods” (self-isolated groups of carers taking additional precautions themselves).
Low-risk people would be able to travel, work, and have social interactions, with strict precautions (masks, hand hygiene, social-distancing) in enclosed spaces. They should be encouraged to say: “I promise to protect you,” taking additional action when interacting with vulnerable people, and restricting the number of social contacts to contain any spread. They, too, should isolate themselves if they get symptoms associated with Covid, sheltering in place with mild symptoms and seeking help with more severe ones.
- Establishing supportive government policy: lightly restricted outdoor activity (with physical-distancing and limited number of contacts) could be allowed, with heavily enforced indoor or closed space (public transport) control and evidence-based preventive behaviours, such as “no mask (properly worn), no entry”, “no hand wash, no entry”. Financial support should be made available for high-risk people because of their work restrictions.
- Continuing to invest in healthcare service capacity: we need a reliable personal protective equipment (PPE) supply chain all the way to end-users, with oxygen, respiratory support devices, and financial and mental health support for a sufficient (expanded) professional workforce serving the entire population.
As we reopen the economy, interventions focused on vulnerable, at-risk groups are needed to avoid loss of life while allowing businesses and other key activities to resume. These major changes for the whole of society are possible with leadership and a coherent plan of action and a commitment to testing and learning what works.
• Dr Kantor is a consultant at Insight Actuaries & Consultants. This article is written with acknowledgment to Dr Pierre Barker of the Institute for Healthcare Improvement, Boston, US.