By Karina Toledo | Agência FAPESP – Two studies recently posted to the preprint platform medRxiv and not yet peer-reviewed show how the slow pace of COVID-19 vaccination in Brazil can jeopardize its effectiveness in terms of reducing the number of deaths from the disease during this intense stage of the pandemic.
One of the articles, whose lead author is Eduardo Massad, a professor at the University of São Paulo’s Medical School (FM-USP), estimates that about 127,000 Brazilian lives would have been saved by end-2021 if mass vaccination had begun on January 21, with some 2 million doses being injected per day. The current average is 200,000 per day, which is only 10% of the rate considered ideal by the researchers in light of the capacity demonstrated by Brazil’s national health service (Sistema Único de Saúde, SUS) in many past vaccination campaigns.
If mass vaccination had started a month later, on February 21, the number of deaths averted in the rest of the year would have fallen to 86,400. The projection dwindles as time passes, falling to 54,500, 30,300 and 16,400 with mass vaccination starting on March 21, April 21 and May 21 respectively.
“All the signs are that mass vaccination in Brazil will begin for real in August, provided Butantan Institute and FIOCRUZ fulfill their promise to deliver 150 million doses by July,” Massad told Agência FAPESP. “I’m not counting vaccines from other sources, such as Pfizer, Moderna and Janssen, because only tiny amounts of these will reach Brazil in the first half, if any. For a different scenario to be possible, deals should have been struck to buy them in 2020.”
Based on mathematical modeling and the trends observed in December 2020 (transmission rate and daily new cases, not considering the novel Brazil variant P.1), the group led by Massad calculated that the number of deaths from COVID-19 in Brazil would reach 352,900 in 2021 in the absence of vaccination.
“That’s an underestimate because the epidemic curve became much steeper with the advent of the new strains of SARS-CoV-2,” Massad said. “In the current context, the number of deaths would probably exceed 400,000 [by December 31 in a no-vaccination scenario]. So all the numbers presented in the article should be considered minima or floors.”
According to Massad, Brazil has experienced a third wave of COVID-19 since the number of daily cases jumped in January. “The second wave wasn’t slowing much when the novel variant [P.1] emerged and transmission accelerated again,” he said. “In the past four weeks, we’ve seen new highs almost every day.”
Considering current trends, he added, 100,000 more Brazilians will die by year-end owing to the slow pace of vaccination. If the start of mass vaccination could be moved up to May, the number could be halved. “We’ll probably have 30 million to 40 million doses by mid-year, and that’s not enough to vaccinate even half the high-risk group, which comprises 77 million. With 40 million doses, we can only vaccinate 20 million people in the period, or less than a third of what’s needed,” he said.
For Massad, therefore, the impact of vaccination on the epidemic curve will be practically nil in first-half 2021. “If daily cases start falling, it will undoubtedly be because of the natural course of the disease or due to social distancing and mobility restrictions, which are increasingly hard to implement,” he said.
All the estimates presented in the study assumed a vaccine with 90% efficacy, higher than those of the vaccines now available in Brazil (Covidshield supplied by AstraZeneca and FIOCRUZ, and CoronaVac supplied by Sinovac Biotech and Butantan Institute). However, Massad said all the vaccines approved for emergency use or awarded a definitive license have proved equally effective in preventing deaths from COVID-19, the key parameter used in the study.
Speed makes a difference
The second study on the same topic was led by Thomas Vilches, a post-doctoral fellow at the University of Campinas’s Institute of Mathematics, Statistics and Scientific Computing (IMECC-UNICAMP), collaborating with researchers at São Paulo State University (UNESP) in Botucatu.
This study is also based on mathematical modeling. It projects 18 scenarios relating to the impact of vaccination on COVID-19 hospitalizations and deaths in São Paulo. Among the factors that vary in the different scenarios are the vaccine (CoronaVac or Covidshield), the speed of distribution (630,000 or 1.2 million doses per day nationwide), the protection afforded by each vaccine against severe symptoms (ranging from 0% to 100%), protection against infection (also 0%-100%), and compliance with social distancing restrictions due to each vaccinated individual’s perception of risk.
In designing the model, the researchers assumed that about 20% of São Paulo’s population had already been infected and had antibodies against SARS-CoV-2. They also assumed a reproduction rate of 1.04 (meaning 100 infected people transmit the virus to 104 people). This is the number reported for São Paulo in January by the COVID-19 BR Observatory.
“The baseline scenario assumes daily vaccination of 0.3% of the population, which corresponds to some 630,000 doses per day,” Vilches said. “This rate can be achieved only once production by Butantan Institute has reached ‘cruising speed’ [1 million doses per day]. Under the scenario we call ‘doubled vaccination rate’, 0.6% of the population are vaccinated per day, which can be done if you add up the doses produced by Butantan Institute and FIOCRUZ.”
Even the baseline scenario is faster than the current pace of vaccination. “Brazil has a significant capacity to distribute vaccines thanks to the structure of the SUS. Our problem is production,” Vilches said.
Under the best-case scenario – 100% relative protection (50.3% efficacy and hence protection against infection) and 100% protection against severe symptoms – vaccination with CoronaVac at the baseline speed would reduce COVID-19 deaths by 45.3% in the current peak period. The reduction would be 57% with Covidshield.
If vaccination speed is doubled, the percentages jump to 65.7% fewer deaths with CoronaVac and 74% with Covidshield.
“At double the speed, even under the worst-case scenario [0% protection against severe symptoms and 0% against infection], CoronaVac could reduce the number of deaths by 30% and Covidshield by 46.8%,” Vilches said.
The main message of the study, he added, is that if vaccination is to reduce deaths from the disease significantly, every possible effort should be made to speed it up throughout Brazil. That includes buying vaccines from all available sources. “The more the merrier,” he said. “There’s no justification for not doing so. Vaccinating fast is fundamental to avert deaths and prevent the emergence of even more aggressive novel variants.”
The preprint “Vaccination efforts in Brazil: scenarios and perspectives under a mathematical modeling approach” by Thomas Nogueira Vilches, Felipe Alves Rubio, Rafael Augusto Forti Perroni, Gabriel Berg de Almeida, Carlos Magno Castelo Branco Fortaleza and Cláudia Pio Ferreira is at: www.medrxiv.org/content/10.1101/2021.02.22.21252208v1.full.pdf.
The preprint “Modelling the impact of delaying vaccination against SARS-CoV-2 assuming unlimited vaccines supply” by Marcos Amaku, Dimas Tadeu Covas, Francisco Antonio Bezerra Coutinho, Raymundo Soares Azevedo and Eduardo Massad is at: www.medrxiv.org/content/10.1101/2021.02.22.21252189v1.full.pdf.