As in many countries, mathematical models are being used in Australia to inform our pandemic response and the roadmap out.
Using various inputs on demographics, disease transmission, mobility patterns and vaccine efficacy, models can compare and predict what impact different public health interventions, such as vaccination and school closures, will have on infection rates, hospitalisations and deaths from Covid-19 – the model outputs.
Models have been very helpful in informing recommendations for vaccination coverage to enable a stepwise easing of restrictions.
Models are built on a combination of real data (such as vaccine effectiveness against the Delta variant) and assumptions (such as contribution of children to transmission) which, in the absence of real data, are derived from estimates informed by expert opinion.
Models are a valuable tool, but they are only one component of the many factors that contribute to policy decision-making.
As paediatricians, we are interested in what the outputs of the models may mean for children and adolescents.
The Doherty modelling considers possible consequences of a range of control scenarios. At one end, the model anticipates that on average there could be more than 250 admissions per week of children in Australia with Covid-19, with 22 children admitted to intensive care and approximately eight deaths per week.
This is based on a scenario over the next six months with high infection seeding at 70% two-dose vaccination coverage among those aged over 15 years, along with baseline restrictions and partial testing, tracing, isolation and quarantine (TTIQ). At the other end of the modelling, with low seeding, 80% coverage, baseline restrictions and optimal TTIQ, this declines to 55 weekly admissions, four in ICU, and less than two deaths per week.
So, how do these model outputs for children and adolescents stack up with what is happening in the rest of the world?
Victoria’s recently announced roadmap appears to take a similar approach to Singapore (population 5.7 million), who are about three months ahead of us in their vaccine rollout. The experience in Singapore therefore may provide some insight as to what we might expect by Christmas.
Currently 82% of the population is fully vaccinated. By the end of July, very high vaccination coverage in teenagers aged 12-18 years was achieved. Singapore started re-opening schools on 28 June, and started easing other restrictions in June, with more easing in August. Currently, there is an outbreak of around 1,200 cases per day, with1,083 in hospital (nearly all cases are hospitalised regardless of medical need), of whom 145 require oxygen and 19 are in intensive care.
There are around 120 people aged less than 18 years in hospital, but none require supplementary oxygen or intensive care. There have been no deaths among children during the entire pandemic.
Primary schools have been closed as a precautionary measure to ensure Covid-19 does not interrupt school exams next week. Authorities are monitoring the number of patients with serious illness to determine if any further restrictions are required.
Scotland (population 5.4 million) has high vaccination coverage, with more than 85% of adults vaccinated with two doses. Over the past few weeks, Scotland has had up to 2,000 cases per day among those aged less than 18 years. Scotland is only just starting to vaccinate adolescents aged 12-15 years.
Each week, about 70 children are admitted with Covid-19. These admissions also include any child who tests positive but are admitted for non-Covid-19 reasons, such as a broken leg. Since Delta arrived, there have been no deaths among children.
In contrast, the situation in the US is highly variable and it therefore is a fraught exercise to apply their experience to the Australian context. In total, there have been more than 42m infections and over 680,000 deaths. For children, there have been more than 5.5m cases among 75 million children, with 480 deaths.
The American Academy of Pediatrics said recently “it appears that severe illness due to Covid-19 is uncommon among children”. The hospitalisation rate is 0.9% (range 0.1-2%) among children, lower than during 2020 when it was 2-3%, and the mortality rate of Covid-19 in children has decreased from 0.06% in 2020, to 0.01% in 2021.
Currently, the greatest number of cases are in the southern states where there is low vaccination coverage and highly variable compliance to restrictions, including wearing face masks.
Many paediatric intensive care units are at capacity in these states because of widespread community transmission. In contrast, states and cities with high vaccination coverage have not seen such a surge. San Francisco (population 4.6 million) has had on average five admissions for children with Covid-19 each day throughout the entire pandemic, and schools have been fully open since mid-August with no outbreaks of Covid-19.
Infectious diseases modelling helps us to think through possible future scenarios, but these are not predictions. Models have, and will continue to have, an important role in informing the pandemic response.
Benchmarking against countries that are ahead of us in the pandemic by looking at their real-world surveillance data is important to help validate the models and to inform what we may expect as we prepare programmatically and psychologically for living with Covid-19.
Models are not “set and forget”, and Australia is not exactly the same as any other country, and so incorporating Australian real-world data, as it becomes available, into policy decision-making will be important moving forward.
• Professor Fiona Russell is director of child and adolescent health PhD program, University of Melbourne, a paediatrician, infectious diseases epidemiologist and public health researcher, Murdoch Children’s Research Institute
• Professor Andrew Steer is director, Infection and Immunity Theme, Murdoch Children’s Research Institute