Professor Ghani’s work with Imperial College Covid-19 Response Team’s “Unlocking” Roadmap Scenarios for England” is key to shaping the country’s four stage “roadmap” out of lockdown planned to end by June 21st.
However the use of modelling has come under fierce criticism throughout the pandemic with some scientists claiming that the government has over-relied on models which do not account for the harms of lockdown and may not include the most up to date information on key factors that could change the course of the pandemic.
The Imperial model which is being used to help “Unlock” Britain has itself been criticised for failing to account for the effect of the seasons on virus transmission – studies have shown UV light reduces the infectivity of Covid-19 – nor adjusted for the changing nature of the population at risk as a result of the vaccine rollout.
However Prof Ghani said: “We did look at seasonality, however we found seasonality has little impact on a large susceptible population early on in a pandemic.”
She added that the models did adjust to changing data: “We revise internally every time we think there are new things that need incorporating – alongside making necessary changes to the model code which take longer than revising the numbers that are put in – but requests from the UK government come on a schedule that is driven by them. That has varied from 1-2 times per week to every 2-3 weeks in recent months.
“Chris Witty has reasonably said we need to measure the impact of unlocking and have time to check what is happening as we have calculated the opening. The data takes a while to feed in from all data sources coming through and it is hard to see the signal.”
She added: “We also need to worry about variants in Covid and by reducing transmission as far as possible keeping cases low we will help prevent these.”
She said it was not the job of modellers to measure lockdown harms: “We are acutely aware of the effect of all the interventions that are happening on people and we are not involved in that decision-making. There could be a separate body to get harms evaluated but that is not my remit as a scientist.”
And she acknowledged the limitations of models: “Models should not be used as predictions. We think we have the best understanding of an evolving situation which is moving so rapidly and we are constantly updating everything forward.
“But the projections have huge uncertainties and it is impossible to make predictions because it depends on the behaviours of the population and the actions that everyone takes including the government.”
Professor Ghani and her response team, together with London startup AquAffirm, have recently developed an interactive modelling tool – Covidism.org – to help government’s in other countries inform key decisions out of lockdown. The platform – which is currently being used by the World Health Organisation as well as Africa, Latin America and Asia – is designed to help them monitor the spread of Covid-19 and has been used by scores of other countries.
Professor Ghani said: “The UK has a large coordinated group of modellers. It is very influential and its research strength runs throughout the UK. It has informed global policy and we’ve worked in particular with Africa, Latin America Asia and other countries.”