Intended for healthcare professionals


Outsourcing covid-19 vaccination to the private sector will increase health inequalities

BMJ 2023; 382 doi: (Published 04 September 2023) Cite this as: BMJ 2023;382:p2012
  1. Daniel M Altmann, professor of immunology1,
  2. Rosemary J Boyton, professor of immunology and respiratory medicine, honorary consultant physician1
  1. 1Imperial College London, UK

It is challenging, confusing, and divisive to even try to describe the current state of play as we enter another autumn of living with covid-19. Reliable, national level disease data are harder to collect, and this data vacuum has left an information gap. Where do we actually stand as we enter the autumn and the forthcoming return to school? There is an uptick of cases of covid-19 in the UK as assessed by hospital admissions (up by 20% in the week before last) or the ZOE app.1 We are technically in another covid-19 wave, although starting from a low baseline and the uptick in cases and hospital admissions is nowhere near as large as previous years.1 Many millions of people around the world are infected with covid-19 at any given time, some chronically so. New mutations continually arise. Some of these become the next emerging variants of concern and some are more transmissible and/or more immune evasive than the wave they displace. In England, it has just been announced that the covid booster campaign will start earlier than planned following the emergence of a new variant, BA.2.86.2

In various parts of the world variants of concern compete for dominance. In the UK, this comprises a period dominated by XBB.1.16 and the possible emergence of EG.5.1 (Eris) and now, BA.2.86.2 Cases of long covid remain massively high, with around 700 000 of the current UK caseload incurred during the “mild Omicron” triple-boosted period since 2022.3 As far as one can foresee, this winter promises more of the same: frequent, recurrent, breakthrough infections—some will be mild, some not, leading to workplace absences, further cases of long covid, and a country still on schedule for somewhere around 20 000 excess covid-19 deaths per annum, adding to the current cumulative toll of 228 000.4

Faced with this prospect, the Joint Committee on Vaccination and Immunisation (JCVI) have grappled with unenviable choices about where to head next across the uncharted territory of the national covid-19 booster campaign. Hard fought herd immunity (in fact, generally hybrid immunity from a mix of variant infections and vaccine doses) is holding up to the limited extent that hospital admissions and deaths from covid-19 remain low as a proportion of total infections compared to 2020-2021. The initial covid-19 booster programme was widely targeted to all aged over 50 as well as the clinically vulnerable, while the 2023 spring booster campaign was for those aged 75 years and above.

This autumn, the eligible population is narrower than for the initial booster campaign. As the JCVI state, the aim now is “to focus the offer of vaccination on those at greatest risk of serious disease and who are therefore most likely to benefit from vaccination. For autumn 2023, JCVI has begun to include cost effectiveness considerations in the development of its advice.”5 Those eligible are the over 65s, carers, frontline healthcare workers, and the clinically vulnerable. The major effect of narrowing the eligible group is that all over-50s who were previously eligible will now be 1-2 years out from their last booster.

One might consider what the criteria for assessing cost effectiveness of a booster programme beyond the rather narrow scope described above. A cost/benefit analysis of doses administered against excess deaths and the autumn/winter pressure on the NHS is obviously paramount. However, there is a case for considering working days lost and lost school attendance, and new cases of long covid, with all the associated costs in human misery across the age ranges, losses to the workforce, the economy, as well as pressure on GPs and the health service. There is also a long term, big picture obligation to continue to control the evolution of ever more evasive variants.

However, one should offset unbridled enthusiasm for a universal booster campaign with knowledge that the public appetite to come forward for the vaccine boosters may be dwindling. Even among the clinically vulnerable who were eligible for the spring 2023 boosters, uptake in London was only 40%.6 Also, as all have come to appreciate, these aren’t all or nothing arguments of protection. The vaccines are imperfect and many people are experiencing repeat breakthrough infections. Yet at a population level, we can observe breakthroughs of covid-19 rising as population immunity wanes with time out from the last booster.

The latest suggestion offered for those excluded by the national booster campaign is that it will be possible to buy covid-19 booster doses privately. No information has yet been released about the specific vaccines or their pricing, but figures around £100 per booster have been cited. If so, the cost may be prohibitive to many. If uptake (and messaging) was poor for the free booster in spring, what might uptake look like if it costs individuals £100 a time? All metrics, whether deaths, hospital admissions or cases of long covid show this to be an infection strongly skewed to socioeconomic deprivation. The risk of long covid roughly doubles across the socioeconomic spectrum.7 This is a complex mix of occupational susceptibility (who can work from home, who can afford to take a lateral flow test and stay at home until they are covid negative) and the disproportionate impact of covid-19 on socio-economic groups and ethnic minorities.

Discussion of the forthcoming options for the design of the covid-19 booster doses is critical, yet the debate has barely been aired. There’s a substantial disconnect here between immunological knowledge and public health discourse. At the start of the omicron waves, the immunological evidence was debated to inform the divergent policy choices between bivalent boosters of BA.1/Wuhan Hu-1 in the UK and BA.5/Wuhan Hu-1 in the USA. Since then, the covid-19 sequences have diverged a considerable distance from those early omicron viruses, as indeed has “immune imprinting” of our immune repertoires by our continuing exposures.89 This emphasises the complexities of how to best focus the immune response on neutralisation of the current variants. Whatever the options are which are now being weighed up, this would assume huge immunological background insights into the patterns of cross-protection. The simple answer is that even for a highly imprinted immune repertoire, there’s potential to refocus the selected response onto neutralisation of a new omicron subvariant, depending on the vaccine given. The vaccinology and pre-clinical trials have continued in the background, with many additional iterations of variant spike sequence-based vaccines or multi-antigen vaccines in-waiting. The high-tech details do matter here—this isn’t the equivalent of councils shopping for the cheapest gritting suppliers before the snow.

The JCVI’s decision to limit the booster rollout is well intentioned and carefully weighed, yet in light of the above, we do still need further debate. Can it really be right that, in a national health service, those with the cash to spare can opt to pay privately for an additional top-up, while those less able to pay and potentially at greater risk, will have to take their chances on their next reinfection being a mild one and not one that puts them out of work? For a disease that has already taken an enormously unequal toll across the socioeconomic spectrum, we now risk widening that gap further.


  • Competing interests: DMA has received payment for consultancy work with AstraZeneca, Novavax and Pfizer

  • Provenance and peer review: commissioned, not peer reviewed.