THE VACCINE ARRIVED LATELY AND IN INSUFFICIENT QUANTITY IN BRAZIL
Although we will not stop celebrating, we have to be realistic: the delay in vaccination projects very pessimistic scenarios from February/March. Add to that the fact that, on this date (01/21/2021), we have 6 million doses available, an insufficient number (20%) compared to the estimate projected for Phase 1 of the National Vaccination Plan of the Ministry of Health, estimated at 31 million doses. Worse, we still do not have a date for when the inputs for the production of new doses will arrive.
In collaboration with The COVID 19 Localisation Modelling Group, our research group made some inferences from the results of simulations of the "Covid 19 Westminster" model, developed using Systems Dynamics. Such inferences are consistent with that of many Brazilian researchers and specialists, and reinforces the need to redouble care regarding health protocols.
The analyses were based on the summary table of graphical results of the article If Covid 19 in London is 70% more infectious it’s a triple whammy. It is good to remember that the UK Government has declared an emergency level 4 (TIER 4) for London. The table presents six charts with different hypotheses. Our attention first fell on the chart "3 - Total Deaths with Mutation", plus "Delay in Vaccination", and the hypothesis of "only vaccination", as it is clear that the health protocols have not been obeyed by a large part of the population. We also believe that the mutation already identified in Manaus, will tend to spread to the rest of the country.
This leads us to the orange curve, which considers a reference number of COVID-19 deaths in Westminster, with "y1=200" and "x1= December/2020", which implies an exponential increase in deaths, reaching "y2=500" and x2="December/2021". The total number of deaths in the period is the integral of the curved area x1:y1/x2:y2, being estimated at more than 3,300 deaths. With vaccine alone, about 55 lives would be spared, with vaccine, respect for protocols and surveillance, 275 lives would be spared.
In the simulation of The COVID 19 Localisation Modelling Group the estimated probability of deaths is 7.4 times higher, double the forecast for February 2021 (3.4 times). And they explain:
Getting infected does not change your likelihood of a severe infection or death unless you can’t get hospital care. What it does instead is change the timing. That is critical because many people will be infected earlier —so more people get infected before they can be vaccinated. The same proportion of those extra infections (just under 1% is the track record so far in London) are severe. Our model produces and estimate from the data that survival rates have about doubled treating people with a severe infection since March and April but still about 0.2% of infections in London lead to death.
We are currently identifying some locations and preparing the model and data to do the first simulations for Brazil, which should happen next week, with the results published later this month.
Our group is also developing its own pandemic simulation and projection model to assist in the creation of public health policies based on time series, using Artificial Intelligence, Machine Learning, Deep Learning and System Dynamics, in close contact with The COVID 19 Localisation Modelling Group.
In conclusion, we renew our warning that the care of the pandemic should be increased due to the mutation of the virus and the delay in vaccines, with respect to sanitary protocols for the use of masks and social distance. This applies to all individuals, vaccinated or not, for an indefinite period.
COVID-19 Pandemic Research Group – Brazil:
- Guttenberg Ferreira Passos (Project leader);
- Ilan Chamovitz
- Ludmila Deute Ribeiro
- Niraldo Nascimento
- Paulo Ely Nakamura
- Suellen Teixeira Zavadzki de Pauli
Partner organizations:
- PRODEMGE – Company Government of the State of Minas Gerais - Brazil
- Amazon Web Services (AWS)
- Secretaria de Estado de Saúde de Minas Gerais (SES-MG) - Minas Gerais State Health Secretariat
Acknowledgments: