Covid-19 and vaccines—known unknowns

On 25 February 2021, The BMJ hosted a webinar on covid-19 vaccines. An expert panel discussed the safety and efficacy of the vaccines, as well as dose schedules and obstacles to global distribution. Nikki Nabavi and Juliet Dobson report

The webinar was inspired by an editorial published in The BMJ entitled “Covid-19’s known unknowns.” The key message: “The more certain someone is about covid-19, the less you should trust them.” Register for future events here

Before diving into discussions on the covid unknowns relating to vaccines, Richard Malley, from Boston Children’s Hospital, summarised what we know: “We have three vaccines currently authorised in Europe and in the US, and several others in selected settings and in other countries. We have seen, and are still seeing, more evidence of efficacy in clinical trials, with [some] limitations. We’re also getting some preliminary effectiveness data in real life, in different settings in Israel, in the UK. And, very importantly, there are more vaccines on the way.” 

He described some of the unknowns as “the impact on transmission or asymptomatic infection, the duration of protection, and the possible need for boosters; who should be prioritised; the issue of delaying the second dose, the full versus the half dose, the coverage of variants, and whether boosters will be required that will cover these variants. And then also, what to do with an individual who has recovered from covid—when should they be vaccinated and with how many doses.”

Safety and efficacy 

Susanne Hodgson, Oxford University, started by clarifying the difference between efficacy (how well your vaccine works in a clinical trial setting) and effectiveness (how well your vaccine works in a real life scenario). She explained the importance of understanding the possible endpoints and how well each vaccine works against each endpoint. “If a person becomes exposed to a virus, the vaccine could prevent infection. If you are infected, a vaccine could reduce the likelihood of you developing symptoms. And if you do become symptomatic, a vaccine could modify the severity of the disease, and indeed whether you die from that infection or not. The vaccine could also modify the likelihood that you transmit infection onto other people.” She shared the efficacy of some of the current vaccines, and their efficacy particularly in older age groups, noting that the Oxford AstraZeneca vaccine has not yet reported efficacy in this age group, as the researchers had only five cases in patients older than 55.

Hodgson said: “We really needed a disease modifying vaccine because the mortality and morbidity associated with covid are what is causing such chaos—but ideally, we also want a vaccine that is capable of preventing infection, so that we can really have an impact on transmission and control this pandemic.”  She explained the importance of understanding both symptomatic and asymptomatic infections in randomised controlled trials. 

“Full transparency is a prerequisite for a science based vaccination programme,” said Peter Doshi, University of Maryland and The BMJ. What do we mean when we ask if the vaccine “works” and  agreed with Hodgson that there are two desired possible outcomes—reducing hospital admissions (ICU admissions and deaths), as well as leading towards some form of herd immunity—but reminded us that  the randomised controlled trials “were not designed to study either.” At best, he suggested, we might get some hints from randomised control trials, but definitive answers “remain elusive, and non-randomised, observational studies will fill this [information] void.”

Doshi outlined the fact that the vaccine trials all say that they will be making their data published and available, although many of these will not be released for a very long time. “Most are tying the data release to the end of the trial.”

Ben Goldacre spoke about OpenSAFELY, “a new, fully open source, highly secure data analytics platform that we’ve built during the covid-19 pandemic, which contains an unprecedented scale of data.” He spoke of the privacy challenges around so much data, such as patient identification. 

Dose schedule 

Malley explored first why so many options exist for the dose schedule. He described the limitations that the vaccines have, such as stability issues. “For example, for Pfizer it is the maturing of vaccines, although they’re trying to address many of these,” or the price of the vaccine, which is “a big deal for the developing world.” He then mentioned some “immunological unknowns.”

Malley and Adam Finn believe that giving more of the vulnerable people their first dose of the vaccine for now confers better protection for the entire population. Malley shared some assumptions relating to the dose schedule. “One that I believe is incorrect is that clinical trials determine the optimal schedule,” . . . “the reality is that these numbers are almost picked out of a hat.” Malley said that “the second fallacy is that the schedule shouldn’t be messed with,” but explained that this differs “greatly across different countries, and we are not seeing that any one country is seeing significantly better results of vaccination than others.”

“A third fallacy is that the protection that was seen after the first dose will disappear if you don’t give a second dose on schedule,” he added.

Equity and obstacles to vaccine distribution and reach 

Carlos Correa focused on the intellectual property of covid-19 vaccines. He explained that the UN and other international agencies decided that “all products related to covid-19 should be considered global public goods. So the idea was that there should be no appropriation of vaccines or protective equipment or diagnostic kits, for example. The world should work in the direction of allowing everybody to use whatever technologies and allow inventions related to covid-19 products.”

But “intellectual property is crucial for innovation. If there is no threat to property, we are not going to produce new vaccines or improve them, and therefore we actually need to ensure that intellectual property is granted.”

Priti Patnaik spoke about equity and obstacles to access to covid-19 vaccines. She asked what some of the global mechanisms are  that have been put in place to tackle the question of equitable access and the potential obstacles to equity we face today—and is equity even possible? She highlighted the COVAX initiative, and how fewer doses have become available than expected. 

David Himmelstein spoke about the situation in the US, where the private healthcare system has, “resulted in a really disastrous health system in the face of this pandemic.”

“We really have no organisation of care. And despite the Affordable Care Act, we still had 30 million uninsured at the outset of the pandemic. So we had an utter lack of primary care infrastructure that we could rely on to distribute the vaccine. And overlaid on that was a historic distrust of the health care system by poor people and particularly by people of colour,” he said. 

He discussed the impact of “very low vaccine coverage rates among those who should be vaccinated, among black people, Hispanic people, and Native Americans. And even in the healthcare workforce, we see that pattern . . . refusing to be vaccinated even when it’s offered at no cost, because of their longstanding suspicion and exclusion from the healthcare system.”

At the other end of the spectrum “concierge medical practices” are “administering the vaccine in large numbers to people at lowest risk because they are the most privileged members of those practices and executives in that firm getting to the front of the queue.”

Finally, he said that the US “had a year of complete neglect under the Trump administration, which made absolutely no preparation for the distribution of the vaccine in any parts of our healthcare system.”

Possible futures

Chadi Saad-Roy and Caroline Wagner discussed modelling of immune responses to SARSars-CoV-2 and what that shows in terms of longer term epidemiological predictions. Models show that “the burden and timing of covid-19 infections, a potential for viral adaptation, will depend on what adaptive immune responses look like and how long they last . . . . Overall, the models seem to reflect that a one dose strategy is good in the short term in terms of mitigating the number of cases because you immunise more people. But in the longer term, outcomes will depend on the relative robustness of what the first dose can do. So our models show that if you can switch to a two dose strategy as vaccines become more available, that might help to mitigate those outcomes. This obviously clearly illustrates that we have to rapidly and equitably distribute vaccines because the potential for evolution in one place will undermine efforts made anywhere else.”

This webinar was part of The BMJ‘s series of covid-19 known, unknowns webinars. Find out more and register for future events here

Nikki Nabavi, editorial scholar, The BMJ

Juliet Dobson, editor, The BMJ

Competing interests: none declared.

Safety and efficacy 

Chair: Rajeka Lazarus, University Hospital of Bristol and Weston

Susanne Hodgson (Oxford University)

Peter Doshi (University of Maryland/The BMJ) 

Ben Goldacre (Nuffield Department of Primary Care and Health Sciences

Dose schedule 

Chair: Adam Finn, JCVI, University of Bristol

Richard Malley (Boston Children’s Hospital)

Equity and obstacles to vaccine distribution and reach 

Chair: Allyson Pollock

Carlos Correa (Intergovernmental South Centre)

Priti Patnaik (Geneva Health Files) 

David Himmelstein (City University, New York)

Possible futures:

Chair: George Davey Smith

Chadi Saad-Roy and Caroline Wagner (Princeton University)

Sheila Bird (University of Edinburgh)


Chair: Phil Hammond