In September, 2020, the WHO Prioritisation Roadmap for COVID-19 vaccines gave priority to prevention of severe disease and the highest risk groups. In July, 2021, the revised Roadmap noted that despite the progressive emergence of SARS-CoV-2 variants of concern, defined as mutations conferring increased infectivity, virulence, or relative capacity for immunological escape, vaccine effectiveness against severe disease had been retained. 1 At the end of 2021, global differences in the inter-related variables of population seropositivity2 and vaccine coverage3 have widened, and omicron has been declared the fifth variant of concern. 4 Omicron was detected in Africa, 4 where successive waves of SARS-CoV-2 have resulted in prevalence of past infection higher than 80% in some regions5 and despite greatly increased global vaccine supplies, 3 the average coverage that can be achieved there in 2021 is estimated to be only 17%. 6 In 2022, we argue that COVID-19 vaccine strategies must remain focused on severe disease, and that global equity in achieving high adult coverage (ie, for those aged 18 years and older) of at least one dose is key to minimising severe COVID-19. The first four SARS-CoV-2 variants of concern were discovered in settings with high infection pressure before vaccines were available. The alpha variant of concern was detected in the UK, beta in South Africa, gamma in Brazil, and delta in India during the second half of 2020, but delta was not designated a variant of concern until May, 2021. 7 The delta variant of concern has infectivity around three-fold greater than the other variants of concern, which were all more infectious than the Wuhan strain7 and by July, 2021, had attained global dominance. 3 By contrast, the omicron variant of concern was first brought to attention by an outbreak among adults (ie, those younger than 30 years) in the South African province of Gauteng, a setting of high infectionacquired immunity following a third delta wave but low vaccine coverage in this age group. 4, 8 Omicron was declared a variant of concern on the basis of an unprecedented number of mutations, almost three-fold greater than for delta, previously associated with increases in infectivity and capacity for immune escape. 4 Based on the amount of detection in sewage and rapid increases in prevalence in South Africa of omicron, community prevalence could be higher than recognised. 8 As for previous variants of concern, disease severity can be anticipated to be greatest in infection-naive, unvaccinated people, with substantial preservation of protection against severe disease in the vaccinated. 1, 4 According to in-vitro data from synthetic polymutant spike-protein pseudoviruses containing 20 or more mutations, vaccine boosting of infection-derived immunity might provide the strongest protection against strains like omicron. 4, 9

Indirect protection of at-risk older adults through vaccinating high-transmitting age groups (young adults and adolescents) against SARS-CoV-2 was influential in early modelling studies, 1, 10 based on precedents for influenza. 11 However, although the effectiveness of COVID-19 vaccines against asymptomatic infection (a proxy for transmission) was high for the ancestral virus and alpha variant in the short term, it was reduced by around 20% against delta. 12 Studies from the Netherlands13 and England14 during delta dominance found similar secondary attack rates among household contacts of a fully vaccinated index case, irrespective of their vaccination status. Although vaccine effectiveness against transmission remained statistically significant in the Netherlands (63%; 95% CI 46–75), 13 in England, 39 …