Skip to content
Report Health data and COVID-19 tech

Checkpoints for vaccine passports: Science and public health

Prior to deploying a vaccine passport system, it is essential to have scientific confidence in its impact on public health

10 May 2021

Reading time: 21 minutes

Black outline of scientific microscope in red circle

The foundation of any COVID status certificate or ‘vaccine passport’ is that it allows stratification of people by COVID-19 risk and therefore allows a more fine-grained approach to preserving public health, keeping the community safer with fewer restrictions. Vaccine passports allow only those who pose an acceptably lower risk to others to take part in activities that would normally present a risk of transmission, e.g. working in care homes, travelling abroad, or entering venues and events such as pubs, restaurants, music festivals or sporting fixtures.

Therefore, the first question to ask of a COVID vaccine passport system is whether an individual’s status, for example that they have been vaccinated, conveys meaningful information about the risk they pose to others? Does the scientific evidence base we have on COVID-19 vaccines, antibodies and viral testing, support making that link, and if so, how certain should we be about an individual’s risk based on those proxies?

The development and deployment of a significant number of viable vaccines in just over a year is a remarkable scientific achievement. Tests have also rapidly improved in quality and quantity, and scientific understanding of COVID-19 infection and transmission has improved greatly since the beginning of the pandemic. In spite of these inventions and innovations, unfortunately the novelty of the disease means the answers to significant questions are still uncertain.

Checkpoints for vaccine passports

This is one of six requirements for a socially beneficial vaccine passport system, as outlined in a report based on an extensive review of the key debates, evidence and common questions around digital vaccine passports

Vaccination and immunity

Our knowledge of COVID-19 vaccine efficacy against different its strands and immunity following an infection continues to evolve. Key questions about vaccines include:

  • What are the effect of vaccines on those vaccinated?
  • What are the effect of vaccines on spreading the disease to others?
  • What is the efficacy of vaccines against different emerging variants?
  • What is the efficacy of vaccines over time?

Our expert deliberative panel expressed concern about developing any system of COVID vaccine passport based on proof of vaccination while so much is still unknown – as systems could be built on particular assumptions that would then change. Any system that was developed would have to be flexible enough to deal with emerging evidence.

One certainty is that no vaccine is currently entirely effective for all people. Although evidence is encouraging that the current COVID-19 vaccines offer strong protection against serious illness, vaccination status does not offer conclusive proof that someone vaccinated cannot become ill.

The evidence is even more emergent on the effect of vaccines on the transmission of COVID-19 from one person to another. Any public health argument in favour of introducing vaccine passports relies on evidence that someone being vaccinated would protect others, but this remains unclear.1

A vaccine can provide different types of immunity:

  • Non-sterilising immunity, where an infected individual is protected from the effects of the disease but can still transmit it (and may instead have an asymptomatic case where previously they would have displayed symptoms).2
  • Sterilising immunity, where a vaccinated person does not get ill themselves and cannot transmit the disease.

Experts in our deliberation identified a ‘false dilemma’ in discussions about the efficacy of these different types of immunity: even a population vaccinated with ‘non-sterilising’ immunity should still prevent the disease spreading, as infected individuals will have weaker forms of it and fewer ‘virions’ (infectious virus particles) to spread. Emerging evidence suggests that ‘viral load’ is lower in vaccinated individuals, which may have some effect on transmission, and one study (in Scotland) found the risk of infection was reduced by 30% for household members living with a vaccinated individual, but much remains unknown.3

An issue raised in the deliberation was that focusing on individual proof of vaccination might underemphasise the collective nature of the challenge. Vaccination programmes aim at (and work through) a population effect: that when enough people have some level of protection, whether through vaccination or recovery from infection, the whole population is protected through reaching herd immunity. Even following vaccination, the UK Government’s Scientific Advisory Group for Emergencies offers caution: ‘Even when a significant proportion of the population has been vaccinated lifting NPIs [non-pharmaceutical interventions, like social distancing] will increase infections and there is a likelihood of epidemic resurgence (third wave) if restrictions are relaxed such that R is allowed to increase to above 1 (high confidence)’. This pattern of vaccination and infection may be occurring in Chile, where high vaccination rates have been followed by a surge in cases.4

Different vaccines have different levels of efficacy when it comes to protecting both the person receiving the vaccination and anyone they come into contact with. This is partly due to vaccines having different levels of effectiveness, based on differently underlying technologies.

As of May 2021, 12 different vaccines are approved or in use around the world, utilising messenger ribonucleic acid (mRNA), viral vectors, inactive coronavirus, and virus-like proteins:5

Vaccines approved for use, May 2021

Different levels of efficacy will also be partly due to different individuals responding differently to the same vaccine – the same vaccine may be effective in protecting one recipient and less so in protecting another.

The efficacy of the vaccines may change with different variants of the disease. There are concerns that some vaccines, for example the current Oxford-AstraZeneca vaccine, may be less effective against the so-called South African variant.6 There will continue to be mutations in COVID-19, such as the E484K mutation which has been found in the Brazilian, South African and Kent strains of the disease (this is an ‘escape mutation’ which can make it easier for a virus to slip through the body’s defences) and the E484Q and L425R mutations present in many cases in India.7 Such mutations make understanding of vaccination effects on individual transmission a moving target, as vaccines must be assessed against a changing background of dominant strains within the population.

Booster vaccinations against variants may help manage the issue of strains. It is possible these may be necessary, as the efficacy of vaccines against any strain may change over time; the WHO has said, it is ‘too early to know the duration of protection of COVID-19 vaccines’.8 With the disease only just over a year old and the vaccines having deployed only in the last few months, it will be some time before conclusive evidence is available on this.

Any vaccine passport system would need to be dynamic – taking into account the differing efficacy of different vaccines, known differences in efficacy against certain variants and the change in efficacy over time – as well as representing the effect of the vaccine on the individual carrying a vaccine passport.

There are also questions about any lasting immunity acquired by those recovering from COVID-19. The WHO has noted that while ‘most people’ who recover from COVID-19 develop some ‘period of protection’, ‘we’re still learning how strong this protection is, and how long it lasts’.9

Inclusion of testing

A number of COVID vaccine passport schemes in development (and the UK Government’s review into what it calls COVID status certification) may allow a combination of three characteristics to be recorded and used in addition to vaccination: recovery from COVID-19, testing negative for COVID-19, or testing positive for protective antibodies against

We can group these characteristics into statuses based on medical process, and those based on medical observation.

Status based on medical process includes vaccination status and proof of recovery from COVID-19. In both cases, a particular event – recovering from an infection or having a vaccination – that might have some impact on an individual’s immunity is taken as a proxy for them posing less risk. As described above, the potential efficacy of this must be understood in the context of what remains unknown about an individual’s ability to spread the disease, their own immunity and the change in their immunity over time.

Status based on medical observation – or direct observation of results correlating to risk – includes two forms of testing: a negative test result for the virus, or a positive test result for antibodies that can offer protection against COVID-19.10 Incorporating robust tests might provide a better, though very time-limited, measure of risk (the biggest challenges to this would be practical and operational). Status based on test results would also avoid the need for building a larger technical infrastructure, particularly one involving digital identity records. But current testing mechanisms do have drawbacks.

There are two main kinds of diagnostic tests that could be used for negative virus test certification:

  1. Molecular testing, which includes the widely used polymerase chain reaction (PCR) tests, detect the virus’s genetic material. They are generally highly accurate at detecting negative results (usually higher than 90%), but their exact predictive value depends on the background rate of COVID-19 infection,11 and depends on the point in the infection that the test is taken.12 These tests often detect the presence of coronavirus for more than a week after an individual stops being infectious. They also need to be processed in a lab – during which time an individual may have become infected and infectious.
  2. Antigen testing, which includes the rapid lateral flow tests used in the UK Government’s mass-testing programmes, detect specific proteins from the virus. If someone tests positive, the result is generally accurate – but as these types of test only detect high viral loads, positive cases can be missed (a ‘false negative’) particularly when self-administered. Certificates based on antigen tests are likely to have a high degree of inaccuracy – tests might be useful in screening and denying (a ‘red light’), rather than allowing (a ‘green light’ test), entry to individuals at a specific point in time. They are unlikely to be useful for any kind of durable negative certification.

Antibody tests, meanwhile, confirm that an individual has previously had the virus. There are two sources of variability from these tests. First, people may have variable antibody response when they are infected with COVID-19 – while most people infected with SARS-CoV-2 display an antibody response between 10 and 21 days after being infected, detection in mild cases can take longer, and in a small number of cases antibodies are not detected at all.13 Second, the tests themselves are not completely accurate, and the accuracy of different tests varies.14

It also remains unclear how an individual antibody test result should be interpreted. The European Centre for Disease Prevention and Control advises that it is currently unknown, as of February 2021, whether an antibody response in a given infected person confers protective immunity, what level of antibodies is needed for this to occur, how this might vary from person to person or the impact of new variants on the protection existing antibodies confer.15 The longevity of the antibody response is also still uncertain, but it is known that antibodies to other coronaviruses wane over time.16

Questions remain as to how viable rapid and highly accurate testing is, particularly those that can be completed outside a lab setting. Although a testing regime allowing entry to venues could avoid a number of the challenges associated with using vaccination status (extensive technical infrastructure and access to health data, possible discrimination against certain groups) it also provides practical and logistical challenges – from administering such tests for access to a sporting event or hospitality venue, to the feasibility of regularly testing children – as well as there being uncertainty around the accuracy of tests.

Risk and uncertainty

At a time when uncertainty – about vaccine efficacy, when life will return to ‘normal’ and much else besides – is endemic, it is natural that politicians, policymakers and the public alike are grasping for certainty. There may be a danger in seeing COVID vaccine passports as a silver bullet returning us quickly to normality, with passports suggesting false binaries (yes/no, safe/unsafe, can access/cannot access) and false certainty, at a time when governments need to be communicating uncertainty with humility and encouraging the public to consider evidence-based risk. Our expert panel raised concerns that the UK Government saying it was ‘led by the science’ brought disadvantages, encouraging a simplistic view of it being infallible and squeezing out space for nuance and debate.

Conveying a proper sense of uncertainty and risk will be important as individuals make decisions about their own health that may also have an impact on collective public health. For example, if I have been vaccinated, but know there is a chance it may not be fully effective, how does that change how I assess the risk to me in engaging in certain behaviours?
What information will I need to also assess my risk of spreading the disease to others? Is it useful for a venue that admits me to understand that a passport may provide a false sense of certainty that I do not have or cannot easily spread the disease?

Any reliance on proof that the process of vaccination has been completed will also require careful consideration about the actual change in risk as a result of that system: experts raised the risk that use of passports could increase the spread of the disease, as individuals who believe themselves to be completely protected engage in riskier behaviour. A review of the limited evidence so far suggests vaccine passports could reduce other protective behaviours.17

While vaccine passports could make people more confident in some areas, for example by providing reassurance to vulnerable people who have been isolating, it could also slow down the return to normality by suggesting to some that their fellow citizens are a permanent threat.
Creating categories of ‘safe’ and ‘unsafe’ that could continue to keep risk salient in people’s minds even once the risk is reduced (for example a risk closer to that of flu: dangerous but not overwhelmingly so) could be counterproductive to reopening and restarting society and the economy.

Recommendations and key concerns

If a government wants to roll out its own COVID vaccine passport system, or permit others to do so, there are some significant risks it needs to consider and mitigate from the perspective of public health.

The first is that vaccine passport schemes could undermine public health by treating a collective problem as an individual one. Vaccine passport apps could potentially undermine other public health interventions and suggest a binary certainty (passport holders are
safe; those without are risky) that does not adequately reflect a more nuanced and collective understanding of risk posed and faced during the pandemic. It may be counterproductive or harmful to encourage risk scoring at an individual level when risk is more contextual and collective – it will be national and international herd immunity that will offer ultimate
protection. Passporting might foster a false sense of security in either the passported person or others, and increase rather than decrease risky behaviours.9

The second is the opportunity cost of focusing on COVID vaccine passport schemes at the expense of other interventions. Particularly for those countries with rapid vaccination regimes, there may be a comparatively narrow window where there is scientific confidence about the impact of vaccines on transmission and enough of a vaccinated population that it is worth segregating rights and freedoms. Once there is population-level herd immunity or COVID-19 becomes endemic with comparable risks to flu, it will not make sense to differentiate and a vaccine passport scheme would be unnecessary.

COVID vaccine passport schemes bring political, financial and human capital costs that must be weighed against any benefits. They might crowd out more important policies to reopen society more quickly for everyone, such as vaccine roll-out, test, trace and isolate schemes, and other public health measures. Focusing on vaccine passports may give the public a false sense of certainty that other measures are not required, and lead governments to ignore other interventions that may be crucial.

If a government does want to move forward, it should:


Set scientific preconditions. To move forward, governments should have a better understanding of vaccine efficacy and transmission, durability and generalisability, and evidence that use of vaccine passports would lead to:

  • reduced transmission risk by vaccinated people – this is likely to involve
    issues of risk appetite, as the risk of transmission may be reduced but will
    probably not be nil
  • low ‘side effects’ – that passporting won’t foster a false sense of security in either the passported person or others, which might lead to an increase of risky behaviours (not following required public health measures), with a net harmful effect. This should be tested, where possible, against the benefits of other public health measures.


Communicate clearly what certification means. Whether governments choose to issue some kind of COVID status certification, sanction private companies to do so or ban discrimination on the basis of certification altogether, individuals will make judgements based on the health information underlying potential schemes in informal settings such as gathering with friends or dating.


Governments must clearly communicate the differences between different types of certification, the probabilistic rather than binary implications of each, and the relative risks individuals face as a result.


To support effective communication, governments, regardless of whether they themselves intend to roll-out any certification scheme, should undertake further quantitative and qualitative research of different framings and phrasing on public understanding of risk, to determine how best to communicate efficacy of each kind of certification.

Requirement two: Purpose

Read about the second of six requirements that governments and developers will need to deliver to ensure any vaccine passport system deliver societal benefit


  1. Scientific Advisory Group for Emergencies (2021) ‘SAGE 79 minutes: Coronavirus (COVID-19) response, 4 February 2021’, GOV.UK. 22 February 2021, Available at: (Accessed: 6 April 2021).
  2. Ada Lovelace Institute (2021) The epidemiological and economic impact of vaccine passports and COVID status apps. Available at: (Accessed: 7 April 2021)
  3. European Centre for Disease Prevention and Control (2021) Risk of SARS-CoV-2 transmission from newly-infected individuals with documented previous infection or vaccination. Available at: (Accessed: 13 April 2021). Science News (2021) Moderna and Pfizer COVID-19 vaccines may block infection as well as disease. Available at: (Accessed: 13 April 2021)
  4. Bonnefoy, P. and Londoño, E. (2021) ‘Despite Chile’s Speedy COVID-19 Vaccination Drive, Cases Soar’, The New York Times, 30 March 2021. Available at: (Accessed: 6 April 2021)
  5. Zimmer, C., Corum, J. and Wee, S.-L. (no date) ‘Coronavirus Vaccine Tracker’, The New York Times. Available at: (Accessed: 21 April 2021). Parker et al. (2021) An interactive website tracking COVID-19 vaccine development. Available at: (Accessed: 21 April 2021)
  6. BBC News (2021) ‘COVID: Oxford jab offers less S Africa variant protection’, BBC News. 7 February 2021. Available at: (Accessed: 6 April 2021).
  7. Wise, J. (2021) ‘COVID-19: The E484K mutation and the risks it poses’, The BMJ, p. n359. doi: 10.1136/bmj.n359. Sample, I. (2021) ‘What do we know about the Indian coronavirus variant?’, The Guardian, 19 April 2021. Available at: (Accessed: 22 April)
  8. World Health Organisation (2021) Coronavirus disease (COVID-19): Vaccines. Available at: (Accessed: 6 April 2021)
  9. ibid.
  10. The Royal Society provides a different categorisation, between measures demonstrating the subject is not infectious (PCR and Lateral Flow tests) and those suggesting the subject is immune and so will not become infectious (antibody tests and vaccination). Edgar Whitley, a member of our expert deliberative panel, distinguishes between ‘red light’ measures which say a person is potentially infectious and should self isolate, and ‘green light’ ones, which say a person tests negative and is not infectious.
  11. Asai, T. (2020) ‘COVID-19: accurate interpretation of diagnostic tests—a statistical point of view’, Journal of Anesthesia. doi: 10.1007/s00540-020-02875-8.
  12. Kucirka, L. M. et al. (2020) ‘Variation in False-Negative Rate of Reverse Transcriptase Polymerase Chain Reaction–Based SARS CoV-2 Tests by Time Since Exposure’, Annals of Internal Medicine. doi: 10.7326/M2
  13. European Centre for Disease Prevention and Control (2021) Immune responses and immunity to SARS-CoV-2, European Centre for Disease Prevention and Control. Available at: (Accessed: 10 February 2020).
  14. Ainsworth, M. et al. (2020) ‘Performance characteristics of five immunoassays for SARS-CoV-2: a head-to-head benchmark comparison’, The Lancet Infectious Diseases, 20(12), pp. 1390–1400. doi: 10.1016/S1473-3099(20)30634-4.
  15. European Centre for Disease Prevention and Control (2021) Immune responses and immunity to SARS-CoV-2, European Centre for Disease Prevention and Control. Available at: (Accessed: 10 February 2020).
  16. Kellam, P. and Barclay, W. 2020 (no date) ‘The dynamics of humoral immune responses following SARS-CoV-2 infection and the potential for reinfection’, Journal of General Virology, 101(8), pp. 791–797. doi: 10.1099/jgv.0.001439.
  17. Drury. J., et al. (2021) Behavioural responses to Covid-19 health certification: A rapid review. 9 April 2021. Available at (Accessed: 13 April 2021)
  18. ibid.

Related content