Article Text

Download PDFPDF

Just allocation of COVID-19 vaccines
  1. Anders Herlitz1,
  2. Zohar Lederman2,
  3. Jennifer Miller3,
  4. Marc Fleurbaey4,
  5. Sridhar Venkatapuram5,
  6. Caesar Atuire6,
  7. Lisa Eckenwiler7,
  8. Nicole Hassoun8,9
  1. 1Institute for Futures Studies, Stockholm, Sweden
  2. 2Rambam Health Care Campus, Haifa, Israel
  3. 3School of Medicine, Yale University, New Haven, Connecticut, USA
  4. 4Princeton University, Princeton, New Jersey, USA
  5. 5King's College London, London, UK
  6. 6University of Ghana, Legon, Greater Accra, Ghana
  7. 7George Mason University, Fairfax, Virginia, USA
  8. 8Binghamton University, Binghamton, New York, USA
  9. 9Cornell University, Ithaca, NY, USA
  1. Correspondence to Dr Zohar Lederman; zoharlederman{at}

Statistics from

Authorized COVID-19 vaccines must be distributed fairly. Several proposals have emerged offering guidelines for how to do this. However, insofar as the aim is to have the greatest health impact, these proposals fall short. We offer three suggestions to strengthen them

The most advanced attempt at coordinating vaccine distribution is the COVID-19 Vaccines Global Access (COVAX) facility, a collaboration that brings together governments, companies, international organisations and others to accelerate the development and manufacture of COVID-19 vaccines.1 A total of 182 countries have joined the facility so far, which has secured about US$2 billion for its advance market commitment (AMC). The AMC will allow 92 low-income and middle-income countries to obtain vaccine doses as they are approved or authorised. Currently, COVAX is set up so that in a first phase poor countries can vaccinate 3% of their populations, while rich countries can vaccinate up to 50%. Though the facility hopes to allow all members to vaccinate at least 20% of their populations by the end of 2021.2 3 Other proposals by the WHO suggest prioritising healthcare workers, the elderly and, those with comorbidities that put them at greater risk of severe illness if infected with COVID-19, people from certain high-risk sociodemographic groups and some teachers.

Another sophisticated proposal, the ‘Fair Priority’ Model, suggests countries with vaccines contribute to global distribution once their COVID-19 transmission rates drop to R<1. In phase I, they argue that vaccines should primarily be distributed to minimise standard expected years of lives lost. Other considerations, such as gross national income and poverty rates, can and should become prominent in later stages of distribution. Finally, researchers at Vanderbilt University suggest distributing vaccines to countries based on their ability to distribute vaccines, their capacities to provide care and whether they have helped test and develop new interventions.4

These proposals, while helpful, have three critical limitations. First, they fail to recognise that fairness should primarily concern mediation of claims and interests of different persons, not countries. Proposals for fair distribution must, in other words, address health problems for individuals. Moreover, since most individuals have little choice as to their country of origin or residence, we should not discriminate against them based on location. A fair proposal cannot allow rich countries to hoard vaccines or prioritise their own populations first, nor can it give individuals less priority simply because they live in a country with less infrastructure, capacity or willingness to distribute vaccines. Finally, equity between individuals in achieving the greatest health impact is probably not compatible with equal distribution on the country level. We expand on some of these points below.

Second, allocation principles must explicitly focus on both direct and indirect health effects of COVID-19. Direct health effects include death and disability caused (in full or in part) by the virus. Indirect health effects include death and disability caused (in full or in part) by the social response to the virus. One major concern, for instance, is how COVID-19 indirectly will have devastating consequences in India and other low- and middle- income countries since the response to the pandemic undermines existing infrastructure to manage other infectious diseases such as malaria, tuberculosis and HIV. A vaccine allocation policy based on a desire to mitigate the negative health impact of the pandemic must explicitly aim to reduce all mortality and morbidity, not just morbidity and mortality directly attributable to COVID-19.

Third, having the greatest global health impact requires assisting countries with their vaccine distribution, production and consumption. A fair allocation system must consider how vaccine distribution will determine the success of whatever strategy is adopted. Vaccines may differ significantly in their effectiveness and the resources they require for successful and wide distribution and consumption. Some vaccines, such as Pfizer’s mRNA vaccine, need to be stored at extremely low temperatures (−70°C). This makes this vaccine highly impractical in many countries where adequate transportation networks, consistent energy supplies and sufficient cold chain storage are lacking (in only 28% of sub-Saharan Africa do healthcare facilities have reliable energy).5 Many vaccines also require multiple injections over time, which make them difficult to distribute, especially in countries that lack the staff to administer the vaccines or where the nearest clinic is a multiple days’ walk away. This means streamlining distribution chains from vaccine producer to vaccine consumer and producing and distributing auxiliary equipment (eg, bottles for vaccine, syringes, distribution clinics).

Efforts must also be geared towards maximising vaccine production. Currently, at least 80% of manufacturing capacity is in the generics sector.6 This sector is an extremely valuable resource that should be included in a responsible response to the pandemic. This may be done in different ways, for example, collaborations between vaccine developers and actors in the generics sector.7

Additionally, policy-makers should put in place measures to promote impactful uptake of vaccines rather than their mere distribution. Ensuring transparent, accurate and trustworthy information about vaccines can help advance uptake.8–11 Manufacturers should consider full clinical transparency, that is, sharing trial protocols and timely results dissemination and sharing of individual patient-level data to bolster public confidence in an approved vaccine. What is more, working to understand people’s particular and often legitimate reasons for distrust that go beyond incomplete or distorted information, like long-term economic hardship or past exploitation of a given community by scientific and medical institutions, and working to create the social, economic and political conditions that would enable more trusting relationships is crucial. Indeed, we must recognise trust and trustworthiness as global health goods that must be in place prior to the arrival of biomedical interventions.12

SARS-CoV-2 has been catastrophic in rich countries as well as in poor ones, but a fair vaccine allocation must help us combat the pandemic’s direct and indirect health effects for individuals irrespective of country of origin or residence. We cannot allow rich countries to prioritise their own populations. Furthermore, we should focus well beyond COVID-19’s direct health effects and enhance vaccine production, distribution and uptake. Endorsing the WHO’s Solidarity Call to Action for equitable global access to COVID-19 health technologies can help everyone secure safe and effective vaccines for COVID-19 as soon as possible (16). In a global pandemic, an outbreak anywhere threatens people everywhere.


The authors would like to thank Larry Temkin, Johnathan Wolff, Govind Persad, and Ezekiel Emanuel for discussion and the Global Health Impact team for research assistance (



  • AH, ZL and NH are joint first authors.

  • Twitter @millerbioethics, @sridhartweet, @Caesar Atuire

  • Contributors ZL, AH and NH are the main contributors. All other authors participated in the editing of the manuscript.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Patient consent for publication Not required.

  • Data availability statement All data relevant to the study are included in the article.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.