Vaccine uptake and SARS-CoV-2 antibody prevalence among 207,337 adults during May 2021 in England: REACT-2 study

Introduction

England has experienced a high and uneven burden of COVID-19 across the population, with high mortality and morbidity in care homes, in minority ethnic groups, and in deprived communities.(1). Prior to the launch of the COVID-19 vaccination programme the prevalence of IgG antibody to the SARS-CoV-2 spike protein provided an estimate of the cumulative burden of infection across the population.(2) The COVID-19 vaccination programme started in England in December 2020 with a phased roll-out based on recommendations from the UK Joint Committee on Vaccination and Immunization (JCVI), which built on the WHO Roadmap.(3,4). Once vaccination was introduced, antibody prevalence captured both prior infection and occurrence of antibodies following one or two vaccine doses.

Implementation of the vaccination programme in England has been rapid, with initial deployment of first doses speeded up by delaying the second dose to 12 weeks; by 15 May 2021 69% of adults in England had received their first and 39% their second dose of vaccine, and by 1 July 2021, these proportions were 85% and 63% respectively.(4) There is growing evidence that the programme, initially targeting older adults, health and care workers, care home residents and clinically extremely vulnerable people, is reducing transmission, hospitalisations and mortality.(5)

The REal-time Assessment of Community Transmission-2 (REACT-2) study is a community survey to measure the prevalence of antibodies to the SARS-CoV-2 virus among random samples of adults in England. We previously reported the extent and variation in antibody prevalence after the first wave of COVID-19 in England including the unequal risk by region, occupation and ethnicity.(2) Subsequent rounds of REACT-2 showed a decline in prevalence over the summer of 2020 as antibodies waned in the population,(6) followed by an increase following the second wave and the first few weeks of the vaccination programme.(7) Here we report the results of REACT-2 round 6 carried out between 12 and 25 May 2021 to measure antibody prevalence across the adult population in England.

Methods

REACT-2 methods have been published elsewhere(2,8). Briefly, in each round of the study we invited a non-overlapping random community sample of adults aged 18 years or over in England, based on the National Health Service general practitioner registrations list. The sample was designed to provide approximately equal numbers of people in each of the 315 lower tier local authority (LTLA) areas in England. Those who registered were sent a Fortress lateral flow immunoassay (LFIA) test kit for SARS-CoV-2 antibody self-testing and asked to perform the test at home and complete a questionnaire, including reporting of their test result and uploading a photo of the completed test. Participants were also asked to report on COVID-19 vaccination, with dates. Survey instruments are available on the study website¹. Those who had not received a vaccine dose were asked whether they had been invited to take part in the vaccination programme and their response. For those who had not yet been offered a vaccine we asked about their intention to accept. People who reported being unsure or who would decline vaccination were asked to select from a list of possible reasons for hesitancy(9) with the option also of providing free-text responses.

Round 6 adopted the design of previous rounds with the addition of a boosted sample of older adults to increase the power to detect whether the risk of becoming a case, being hospitalised or dying from COVID-19 differed between those testing positive and those testing negative on the LFIA following vaccination. Round 6 aimed for a total sample size of 240,000, with inclusion of 70,000 additional people in age groups 55-64 and 65-67 years. We estimated that this boosted sample of older age groups would give high probability (p=0.013) of detecting clinically important differences in outcomes (relative risk 0.5 for hospitalisation).

The LFIA used in REACT-2 detects immune responses to the S1 subunit protein targeted by available vaccines. It was selected following evaluation of performance characteristics (sensitivity and specificity) against pre-defined criteria for detection of IgG,(10) with extensive public involvement and user testing.(11) We estimated clinical sensitivity of the LFIA on finger-prick blood (self-read) for IgG antibodies following natural infection at 84.4% (70.5, 93.5) among RT-PCR confirmed cases in healthcare workers, and specificity 98.6% (97.1, 99.4) in pre-pandemic sera.(10,12)

Here we report data on vaccine coverage and results of antibody testing based on complete, validated data from round 6 received on 28 May 2021.

Results

Vaccine coverage and confidence

There were high rates of vaccine coverage (weighted to be representative of the adult population in England) with 72.9% (95% confidence interval 72.7-73.0) reporting having had at least one dose by the time of the survey, varying by age from 25.1% (24.5-25.6) of those aged 18 to 24 years, to 99.2% (99.1-99.3) of those aged 75 years and older (Supplementary Table S2).

Figure 1 shows variation in odds of vaccination; coverage increased steeply with age in line with the vaccine roll-out, and odds adjusted for age were lower in men (odds ratio [OR] 0.89 [0.85-0.94]) than women. After adjustment, vaccine coverage was lower in people of Black (0.41 [0.34-0.49]), and higher in people of Asian (1.13 [1.01-1.27]) ethnicity, than white ethnicity. There was a gradient of coverage in relation to deprivation and household income, with higher coverage in the least deprived and highest income households (Supplementary Table S3). There was variation by region, with odds being lower in London (OR 0.86 [0.79-0.94]) and higher in the North West (OR 1.23 [1.13-1.35]) than the South East. At LTLA level, the weighted vaccine coverage varied between 43.4% and 87.5% (Figure 2, Supplementary Data File). People who reported a history of COVID-19 were less likely to be vaccinated (OR 0.61 [0.55-0.67]). There was high coverage among health and care workers with odds of 9.84 [8.79-11.02] and 4.17 [3.20-5.43] respectively compared to non-key workers, and odds were higher in education workers than other workers at 1.63 [1.50-1.76]. However, coverage was low in hospitality and retail workers compared to other workers (OR 0.73 [0.64-0.82] and 0.77 [0.70-0.85] respectively) (Supplementary Table S3).

• Download figure
• Open in new tab

Figure 1 Logistic regression for vaccine coverage (one or more doses received, y/n) by key covariates

Legend: The plot shows odds ratios for logistic regression models with vaccination status as the binary outcome variable (one or more doses received, y/n) and a range of covariates as explanatory variables, crude and adjusted for age and sex. Data for this chart are in Supplementary Table S3

• Download figure
• Open in new tab

Figure 2 Maps of England showing variation at lower-tier local authority (LTLA) level by age for (a) antibody prevalence (adjusted and weighted), (b) proportion (weighted) with at least one vaccine dose, (c) proportion (weighted) double vaccinated

Overall vaccine confidence was high, with 97.6%, [97.5-97.6] reporting having been vaccinated or intending to accept a vaccine when offered, with an increase in all age groups from round 5 (26 January – 8 February 2021) (Supplementary Table S4). Vaccine hesitancy was uncommon with 1726 (0.8%, [0.7-0.8]) people reporting having refused or planning to refuse a vaccine when offered. Refusal was higher in those aged 25-34 (OR 1.25 [1.07-1.47]) compared to 35-44 years, and lower in the older age groups; it was lower in men ((0.83[0.75-0.91]) than women, and higher in those with lower income and educational level, and in Black, Mixed and Other ethnicities compared to white and Asian ethnicity. Vaccine hesitancy was also more likely in people who had a history of COVID-19, and in smokers. (Supplementary Table S5, Supplementary Figure S1) The most cited reasons for refusal (selected from a list) were concerns about the long-term health effects, side effects, or waiting to see whether it works, although the last reason was less frequently cited in round 6 than round 5 (Supplementary Figure S2).

Antibody positivity following vaccination

The proportion of people testing positive following vaccination dose varied by number of doses, time since vaccination, age, sex and prior COVID-19 infection. Figure 3 shows the proportion of respondents who tested positive for antibodies (unadjusted and unweighted) by number of weeks since their first or second vaccine dose in people who had their second vaccination between 10 and 12 weeks after the first, or who had one vaccine less than 12 weeks earlier. It shows a rise in positivity after the first dose up to 4-5 weeks, and then a decline up to 11 weeks followed by a steep rise following a second dose. At each time point positivity was lower in men than women, in older than younger people, and higher in those with a history of COVID-19, and in those receiving Pfizer-BioNTech than AstraZeneca. (Figure 3 and Supplementary Figure S4)

• Download figure
• Open in new tab

Figure 3: Percentage of respondents testing positive by weeks since dose (for people who had only one dose less than 12 weeks ago, or had second dose between 10 and 12 weeks after the first) (a) for all vaccine types, and (b) separately for Pfizer-BioNTech and AstraZeneca

Legend: The plots show the proportion of respondents who tested positive for antibodies according to time since their first and second vaccine dose. To ensure that we are comparing respondents who had their second vaccination at a similar time interval after the first, the plots show subsets of people who either 1) had their second vaccinations between 10 and 12 weeks after their first, or 2) have had one vaccine < 12 weeks ago. This captures 80% (168,877) of the 212,177 respondents who had received one or more vaccines. In (b) proportions are shown for AstraZeneca and Pfizer-BioNTech recipients; there were insufficient numbers to plot Moderna. Proportions are not adjusted for test performance.

Table 2 shows summary figures for antibody positivity 21 or more days following first, and 14 or more days following second doses separately for Pfizer-BioNTech, AstraZeneca and Moderna. After adjusting for test performance, the positivity after two doses was 100% (100-100) for Pfizer at all ages except 80 years and older when it was slightly lower at 97.8% (95.9-99.6) (Table 2a). For AstraZeneca adjusted positivity was over 90% up to age 69, and then 89.2% (88.5-89.9) in 70-79 year olds and 83.6% (78.5-88.3) in those aged 80 years and over (Table 2b). There was more variation following a single dose; after a single Pfizer-BioNTech dose, adjusted positivity ranged from 100.0% (91.1-100.0) in those aged 18-29 to 32.2% (18.2-51.1) in those aged 70 – 79 years (Table 2a). For AstraZeneca this was 72.2% (68.5-75.9) in the youngest and 46.2% (40.0-52.7) in the oldest age group (Table 2b). There were small numbers reporting Moderna vaccination with most being positive after a single dose (Table 2c).

View this table:

• View inline
• View popup

Table 2: Antibody positivity (adjusted for test performance, not weighted) following vaccination by vaccine type, dose, time since dose and age

Discussion

Our results confirm the rapid and widespread uptake of vaccination among adults across England and indicate a high antibody response among those who have had two doses of vaccine. Nonetheless, some groups still had relatively low vaccine coverage, in part explained by the ongoing prioritisation by age, with lowest vaccination rates at ages 18-24 years. However, lower coverage was also seen in men, people of Black, mixed or other ethnicity compared to Asian and white ethnicities, as well as those living in London, those in more deprived areas and those with lowest household income. Coverage was lower in occupational groups such as hospitality, retail, personal care, and other public facing roles, even after multiple adjustment including age and other factors, potentially placing these groups at increased risk of infection and ongoing transmission. Vaccination rates were also lower in those reporting suspected or confirmed prior COVID-19. In part this may be because in those with known infection it is recommended to wait four weeks before vaccination. However, it is also possible that those with prior infection may be less motivated to be vaccinated in the belief that they are protected.

Our overall estimate of antibody prevalence in mid-May 2021 was 61%, at a point when 69% of adults had received at least one dose of vaccine, with 39% having had two doses.(4) While lower than some other national estimates,(16,17) this is a large increase from the 14% we estimated at the end of January 2021 and shows the substantial impact of the vaccine programme. Specifically, our data show a clear increase in antibody response after second vaccination doses, providing strong support in favour of obtaining a second dose when offered. However, although viral neutralisation titres show a strong association with protection,(7)(18), the relationship between IgG positivity and risk of infection is still not fully understood and emerging variants, including delta, may require higher protective titres to achieve neutralisation.(19)

We confirmed in an independent population sample declines in antibody positivity following first vaccine dose, which we first reported in REACT 2 round 5 (26 January to 8 February 2021);(2) antibody positivity appeared to peak 3-4 weeks following first doses and then declined until after second doses were given. This may have implications for the duration between first and second doses particularly during a period of rapid epidemic growth.

Across regions there was a slightly lower prevalence in the South West where there were lower levels of prior infection, with the highest antibody prevalence found in regions with high rates of prior infection, London and the North West;(20) this is despite the lower vaccine coverage in London. Higher prior infection rates also resulted in higher antibody prevalence in non-white ethnic groups; however, there was a lower level of antibody prevalence in people living in deprived compared to more affluent areas, reflecting differential uptake of vaccines.

Our study has limitations. While we approached a random sample of the population, it is possible that differential non-response may have resulted in biased estimates of prevalence and vaccine uptake. Nonetheless, we achieved a large sample and relatively high participation (over 25%) amongst those invited to take part and weighted our findings to be representative of the population of England as a whole. Antibodies detected in this study reflected anti-S IgG in response to prior infection and vaccination, but this is only one part of the protective immune response, and therefore we cannot equate presence or absence of antibody with immunity. In addition, although offering distinct advantages in terms of costs, logistics and scale, the LFIA does have limitations. Due to limited sensitivity, it does not detect low levels of antibody which may result in lower antibody prevalence estimates than some other studies (21) and potentially be of clinical significance, for example, as a marker of protection against severe disease and/or hospitalisation; as people read their own test results there will be some classification errors including the missing of faint positive lines in people with low visual acuity. On the other hand, within-study comparisons, for example, by ethnicity or deprivation, should be relatively unaffected by test performance. Nonetheless we adjust our prevalence estimates for the performance of the test, based on previous validation work,(10) as well as presenting the observed proportions of test positives.

In summary our findings confirm the successful roll out of the vaccination programme in England with a concomitant high proportion of individuals having detectable antibodies. Those who have had two vaccines at the time of this study, including those over 65 years, had a very high proportion of antibody positivity which is likely to be associated with high levels of protection from severe disease, and possibly infection. Nonetheless, there remain some key groups with a lower prevalence of antibody, notably unvaccinated younger people, certain minority ethnic groups, those living in deprived areas and workers in some public-facing employment. Obtaining improved rates of vaccination in these groups is essential to achieving high levels of protection against the virus through population immunity.

Contributors

HW, GC and PE designed the study and drafted the manuscript. MW and ST conducted the analyses. MW and ST verified the underlying data. HW, GC, CA, AD, CAD, PJD, SR, DA, WB, PE provided study oversight. AD and PE obtained funding. All authors have reviewed and approved the final manuscript

Declaration of interests

No conflicts declared

Data availability

Supporting data are available here.

List of Tables and Figures