Exponential growth, high prevalence of SARS-CoV-2, and vaccine effectiveness associated with the Delta variant

Prevalence and growth

Prevalence of infection with SARS-CoV-2 increased substantially in England between rounds 12 and 13 (Fig. 1) as the third wave took hold, linked to the rapid replacement of Alpha by the Delta variant. In round 13, between 24 June and 12 July 2021, we found 527 positives from 98,233 swabs, giving a weighted prevalence of 0.63% [95% credible interval (CrI) 0.57%, 0.69%], and, on average, a factor of >4 rise relative to the weighted prevalence in round 12 of 0.15% (CrI 0.12%, 0.18%) (table S1). The prevalence in round 13 was similar to that observed in early October 2020 and late January 2021 during, respectively, the rise and fall of the second wave (Fig. 1).

The Delta variant completely replaced Alpha during the period of our study, consistent with genomic data from outbreak investigation and routine surveillance (11). Of the 254 lineages determined for round 13, 100% were the Delta variant, compared with round 12 during which 36 of 46 (78.3%) were Delta and the remaining 10 were the Alpha variant. The growth of Delta against Alpha from round 10 (11 to 30 March 2021) to round 13 corresponded to a daily growth rate advantage of 0.14 (CrI 0.10, 0.20) for Delta, which, in turn, implied an additive R advantage of 0.86 (CrI 0.63, 1.23) (Fig. 1). This is consistent with estimates based on trends in the proportion of positive PCR assays where the S gene was not detected [presumed to be Alpha (12)] and on differences in household attack rate for households where Delta was identified rather than Alpha (13). Within the Delta variant, we did not detect the K417N mutation associated with the AY.1 and AY.2 lineages. Under the assumption that REACT-1 participants provide an unbiased sample of infections, we can exclude, with 95% confidence, a population prevalence of non-Delta lineages greater than 0.004%, corresponding to 2350 infections in England on average during round 13.

Nationally, we observed an exponential trend in prevalence with sustained growth for rounds 12 and 13 (between 20 May and 12 July 2021) (Fig. 1 and table S2) despite England having one of the highest adult vaccination rates internationally (5). Averaging over the period of each of rounds 12 and 13 separately, we estimated the reproduction number R at 1.44 (CrI 1.20, 1.73) (round 12) and 1.19 (CrI 1.06, 1.32) (round 13), corresponding to doubling times of 11 days (CrI 7, 23 days) and 25 days (15, >50 days) respectively. Across rounds 12 and 13, R was 1.28 (CrI 1.24, 1.31) with a doubling time of 17 days (CrI 15, 19 days). Patterns of growth for the period of the study were robust when considering alternative definitions of positivity, such as only nonsymptomatic individuals or positive samples with lower cycle threshold (Ct) values, corresponding to higher viral load (table S2).

Age

Alongside the rapid rise of the Delta variant, recent growth in England appears to have been driven by younger age groups (table S3 and fig. S1). For example, in 13- to 17-year-olds, weighted prevalence in round 13 [1.56% (CrI 1.25%, 1.95%)] was higher than in round 12 [0.16% (CrI 0.08%, 0.31%)] by a factor of 9. Similar patterns were observed in England for the same period in a longitudinal household study (14). In contrast, at ages 65 to 74 years, weighted prevalence increased from round 12 [0.07% (CrI 0.04%, 0.12%)] to round 13 [0.25% (CrI 0.19%, 0.34%)] by a factor of 3 to 4. More generally, participants aged between 5 and 24 years were overrepresented among infected people in our study, contributing 50% of infections (weighted age-standardized) while only representing 25% of the population of England aged 5 years or above (15). Therefore, whether because of mixing patterns, infectiousness or susceptibility, this group was driving transmission and, during a period of exponential growth, any vaccination targeted at the younger ages would have a disproportionate impact in slowing the epidemic (16).

Prevalence among vaccinated and unvaccinated

Participants who reported having received two doses of vaccine were at substantially reduced risk of testing positive relative to those who reported not being vaccinated. For round 13, the prevalence of swab positivity among unvaccinated participants [1.21% (CrI 1.03%, 1.41%)] was greater for all ages than among those who had received two doses of vaccine [0.40% (CrI 0.34%, 0.48%)] by a factor of 3 (table S3). The prevalence in unvaccinated relative to double-vaccinated individuals was similar for round 12, with a prevalence of 0.24% (CrI 0.18%, 0.33%) in those unvaccinated versus 0.07% (CrI 0.05%, 0.10%) in those reporting two doses (table S3).

However, these estimates conflate the effect of vaccination with other correlated variables such as age, which is strongly associated with likelihood of having been vaccinated and also acts as a proxy for differences in behavior across the age groups. Specifically, in England, few children and young people under the age of 18 years have been vaccinated twice, whereas few over the age of 65 years remain unvaccinated (Table 1 and Fig. 1). We therefore restricted the analyses to those aged 18 to 64 years (n = 64,415 in round 12, n = 57,457 in round 13), which permitted direct contrast of infection rates between double-vaccinated and unvaccinated groups (Table 1).

At these ages, we compared swab-negatives with (i) all swab-positives and (ii) the subset of swab-positives who were symptomatic [i.e., reporting one or more common COVID-19 symptoms in the month prior to testing (fever, loss or change of sense of smell or taste, new persistent cough)]. After adjusting for age, sex, region, ethnicity, and index of multiple deprivation (IMD) (17), for all swab-positives, we estimated vaccine effectiveness (VE) of 64% [95% confidence interval (CI) 11%, 85%] in round 12 and 49% (CI 22%, 67%) in round 13 among people who had received two doses of vaccine of any type. For those with symptoms, we estimated VE of 83% (CI 19%, 97%) in round 12 and 59% (CI 23%, 78%) in round 13 (Table 2).

Independent data on vaccination status was provided for 57,338 (89%) participants aged 18 to 64 in round 12 consenting to data linkage, and 49,923 (87%) in round 13 (materials and methods). Using these linked data, we estimated adjusted VE at 75% (CI 35%, 90%) in round 12 and 62% (CI 38%, 77%) in round 13. The apparently higher VE for the linked participants reflected differences in odds of infection among the linked and unlinked groups (table S4), suggesting possible bias introduced by consent to linkage, but also some misclassification of vaccine status in the self-reported data (table S5). Because reported dates of vaccination were more reliable in the linked data, we used those data to examine the effect of including a lag period of 14 days after the second vaccination and observed similar odds ratios for zero lag and 14-day lag following the second dose (Table 1). In addition, we observed a similar unweighted prevalence of swab positivity among double-vaccinated individuals who did and did not report prior infection more than 28 days before their swab (table S5), which suggests in our study that prior infection did not materially affect the estimate of VE. Moreover, the strong correlation among age, vaccine type, and time since vaccination in England, together with limited numbers, prevented us from being able to reliably assess the impact of vaccine type or time since infection independently of age.

Although vaccination was associated with lower prevalence of swab positivity, there remained potential for large numbers of people who had received two doses of vaccine to become infected. During the period of round 12, we extrapolated from our data that 29% of infections in England occurred in double-vaccinated people, rising to 44% during the period of round 13. These increases in prevalence in vaccinated individuals in round 13 could be driven by increased social mixing or by a higher proportion of infections being the Delta variant, or attributable to waning of protection from infection. Also, although lower than for unvaccinated individuals, nearly one in 25 double-vaccinated individuals [3.84% (CI 2.81%, 5.21%)] tested swab-positive if they reported contact with a known COVID-19 case (table S6).

Cycle threshold values

We analyzed Ct values associated with positive results among vaccinated and unvaccinated individuals as a measure of viral load. For all positives in round 13, at ages 18 to 64 years, median Ct value for vaccinated participants [27.6 (CI for median, 25.5, 29.7)] was higher than for unvaccinated ones [23.1 (CI 20.3, 25.8) (positive defined as N gene Ct below 37 or both N gene– and E gene–detected; see materials and methods) (Fig. 2 and table S7). The higher Ct values among vaccinated people may suggest lower infectiousness (18), consistent with transmission studies conducted when the Alpha variant was dominant, in which vaccinated individuals were at substantially lower risk of passing on infection (19). As a secondary analysis, we reduced the Ct threshold for positivity to capture strong positives, which resulted in a smaller difference in median Ct values between vaccinated and unvaccinated individuals (Fig. 2, C and D). At the same time, our estimate of VE for those who reported having received two doses of vaccine increased to 54% (CI 29%, 71%) for a Ct threshold of 35, plateauing between 57% (CI 32%, 72%) and 58% (CI 33%, 73%) for a Ct threshold of 33 and 27, respectively.

Time series of infections, hospital admissions, and deaths

We next investigated how swab positivity measured in REACT-1 related to daily hospital admissions and deaths in publicly available data (6), finding a best-fitting lag between swab positivity and hospitalizations of 20 days and between swab positivity and deaths of 26 days (Fig. 3). At these lags, from early February 2021, there was a clear divergence between swab positivity and deaths, coinciding with the rollout of England’s mass vaccination campaign, with a smaller divergence between swab positivity and hospitalizations. However, as the Delta variant became dominant in mid-April 2021, the associations between infections and hospitalizations and deaths began to reconverge, both for people below and above 65 years (fig. S2).

Geographical variation

At the regional level, estimates of R were consistent with the overall trend within round 13. Prevalence in round 13 was highest in London at 0.94% (CrI 0.76%, 1.16%), up from 0.13% (CrI 0.08%, 0.20%) in round 12 (table S3). There was a suggestion of a possible slowing of the rise in London in the most recent data, although with wide confidence intervals (table S8).

At the subregional level, there was a suggestion of prevalence of infection decreasing in some areas and increasing in others (fig. S3). For example, in the North West of England, high prevalence in a large urban area covering Greater Manchester and Lancashire during the first half of round 13 was less evident in the second half, whereas prevalence increased between the first and second halves in nearby south Yorkshire, part of the Yorkshire and The Humber region. These data are indicative of rapidly changing local spread of the virus within the context of the national exponential rise in infections.

Ethnicity, household size, and neighborhood deprivation

Ethnicity, household size, and area levels of deprivation jointly contributed to the risk of higher prevalence of swab positivity, in addition to age. Unadjusted prevalence (table S3) showed highest prevalence in people of Black ethnicity at 1.21% (CrI 0.75%, 1.93%) compared with 0.59% (CrI 0.53%, 0,65%) in people of white ethnicity; highest prevalence in those in the largest households of six or more people at 1.35% (CrI 0.90%, 2.01%) compared with 0.44% (CrI 0.32%, 0.61%) and 0.44% (CrI 0.36%, 0.53%) in single- and two-person households, respectively; and highest prevalence in participants living in the most deprived neighborhoods at 0.82% (CrI 0.65%, 1.04%) compared with the least deprived at 0.48% (CrI 0.39%, 0.59%). Prior rounds of REACT-1 have shown different ethnicities at increased prevalence at different times, consistently higher prevalence of infection in larger households, and usually increased prevalence in more deprived neighborhoods (20–25). In models including each of the above variables, similar patterns were observed in the odds of testing positive, although odds were reduced when all three of the above variables were considered jointly, together with age, sex, region, and key worker status (table S9). Age remained an important predictor of swab positivity in these mutually adjusted models. Also, in these analyses, women had lower odds of infection than men at 0.80 (CI 0.67, 0.96) in round 13, although not in round 12 at 1.34 (CI 0.93, 1.92) (table S9); this difference may be related to increased social mixing associated with England’s progression in the Euro 2020 football competition during June and July 2021, as was seen previously in Scottish data, reflecting their earlier exit from the competition (26).

Discussion

We report a rapidly rising prevalence of infection in England during 20 May to 12 July 2021 associated with the replacement of Alpha by the Delta variant in a highly vaccinated population. Our central estimate of VE against all SARS-CoV-2 infections for two doses of vaccine (self-report) was 49% in the most recent data, increasing to 58% when we defined effectiveness only for strong positives, and 62% in the linked data. These estimates are lower than some others (19, 27, 28) but consistent with more recent data from Israel (29).

Estimates of VE are not absolute but will vary depending on a variety of factors. Our estimates were higher when we restricted our analyses to people reporting symptoms of COVID-19 in the previous month and to those who consented to linkage of health records, although still lower than those from routine testing of symptomatic people presenting for RT-PCR in England (27). Unlike routine testing, our data are based on a random sample of the population and include asymptomatic people, as well as symptomatic individuals who may not present for testing; our results may therefore give a less biased representation of infection risk. Also, our estimated effectiveness was lower than that from a longitudinal household survey that included asymptomatic individuals but was conducted before the emergence of Delta, where vaccine status was based on a mix of self-reported and linked data (19). More generally, estimates of VE may depend on vaccine type, interval between doses, possible waning over time, and the extent of past natural infection among the comparator (unvaccinated) group.

We show that the third wave of infections in England was being driven primarily by the Delta variant in younger, unvaccinated people. This focus of infection offers considerable scope for interventions to reduce transmission among younger people, with knock-on benefits across the entire population. Also, given the rapid rise of the Delta variant that occurred in Europe, the US, South Asia and elsewhere, and its estimated increased transmissibility, patterns in England were informative of what was subsequently observed elsewhere. In our data, the highest prevalence of infection during June to July 2021 was among 13- to 24-year-olds. In the UK, the Joint Committee on Vaccinations and Immunizations recommended in August 2021 that vaccination should be offered to all 16- and 17-year-olds and then in September 2021 further extended the UK program to include children aged 12 to 15 years, as has been done in the US and some other countries. This expansion of the vaccination program to those at highest risk of infection had the potential to reduce transmission in the autumn and winter 2021 as levels of social mixing, including indoors, increased (30). Also, development of vaccines against Delta and other variants may be warranted in the light of evidence of antigenic change measured by neutralization (31) and the relationship between neutralization titer and protection from mild disease (32).

Estimates of VE against serious outcomes of greater than 90% have been reported for those who have received two doses of either BNT162b2 (33) or ChAdOx1-S (34) vaccines. This is in keeping with our observation of a weakening of the association between infections and hospitalizations and deaths from mid-February to early April 2021 when the Alpha variant was dominant. However, in our more recent data (since mid-April 2021), infections and hospitalizations began to reconverge, potentially reflecting the increased prevalence and severity of Delta compared with Alpha (35), a changing age mix of hospitalized cases to younger ages, and possible waning of protection (29, 36).

Our study has limitations. One estimate of effectiveness was based on self-reported vaccine status, because we could only obtain linked vaccination data for the subset of participants who gave consent, with individuals who did and did not consent to linkage appearing to have different patterns of swab positivity across the vaccinated and unvaccinated groups. Because age, date of vaccination, and vaccine type are so strongly correlated in England, and with limitations in numbers, we were wary of introducing a time variable into the analyses to investigate the waning of VE explicitly. However, the design of the study—based on estimation of infection prevalence from independent samples within (as well as across) separate rounds, conducted monthly—itself provides strong control for any time effects.

Over the course of the study since round 1 in May 2020, toward the end of the first lockdown in England, we observed a gradual reduction in response rates, from 30.5% in round 1 to 11.7% in round 13. These rates are conservative estimates because they are based on numbers of swabs with a valid RT-PCR result compared to the total number of letters of invitation sent out, some of which may have been returned, sent to the wrong address, or left unopened by the recipient. Nonetheless, the drop in response rates means that our sample may be becoming less representative, particularly in some groups such as young people (18 to 24 years) and those living in the most deprived areas where response rates by round 13 had fallen to 4.2% and 5.1%, respectively. Note, however, that these response rates have been achieved without the use of financial or other incentives.

Our method of sampling was designed initially to achieve sufficient numbers in each lower-tier local authority (LTLA) in England so that we could analyze subregional trends and also, by weighting the sample, provide estimates of prevalence that were representative of the population of England. Whereas previously we had aimed to achieve approximately equal numbers of people in our sample by LTLA, in rounds 12 and 13 we switched to sampling in proportion to population in order to capture greater resolution in inner-city areas, which were relatively underrepresented in our previous sampling regimen. In either case, as we reweight the sample according to the national population profile, weighted prevalence should be comparable across rounds, albeit with lower precision in later rounds because of the lower response rates.

Our data show that rapid exponential growth of SARS-CoV-2 prevalence occurred during the third wave in England at a time when the Delta variant became dominant. The rapid rollout of the vaccination program in England has so far limited the number of infections and serious cases relative to the unvaccinated population. Level or declining prevalence was observed during summer 2021 in the Northern Hemisphere, reflecting school vacations, greater time spent outdoors, and reduced social interactions. But without additional interventions, increased mixing (including indoors) in the presence of the Delta variant likely explains renewed growth that occurred in autumn 2021, even in populations with high levels of vaccination. Continued surveillance to monitor the spread of the epidemic is therefore required.

Materials and methods

The REACT-1 study methods have been described elsewhere (9). Briefly, at each round, we sent an invitation by post to named individuals from the list of patients registered with a National Health Service (NHS) general practitioner in England, obtained from NHS Digital, covering almost the entire population. We included all 317 LTLAs in England, and by combining the Isles of Scilly with Cornwall and the City of London with Westminster, we report results across 315 LTLAs overall.

For round 1 to round 11, we aimed to obtain approximately equal numbers of participants in each LTLA to be powered to provide local estimates of prevalence. From round 12 onward, we adjusted the sampling procedure to select the sample randomly in proportion to population at the LTLA level, thus obtaining more samples in LTLAs with higher population density in inner urban areas. However, we ensured that data were comparable across rounds as we reweighted the data at each round to be representative of England as a whole (see below).

For those registering to participate, we obtained age, sex, address, and residential postcode from the NHS register and collected additional information on demographics, health, and lifestyle via online or telephone questionnaire. This included information on ethnicity, smoking, household size, key worker status, contact with a known or suspected COVID-19 case, and whether, at time of survey, participants had experienced one or more of 29 symptoms in the past week or past month (participants not reporting symptoms may have developed symptoms later, but these were not captured). Participants were also asked for consent to longer-term follow-up through linkage to their NHS records including data from the national immunization program. The questionnaires are available on the study website (37).

Response rates have varied by age and over time and place, and are available for each round [“For Researchers: REACT-1 Study Materials” (37)]. Overall response rate was defined as the percentage of invitees from whom we received a valid swab result; this was 20.4% across all rounds, and 13.4% and 11.7% for rounds 12 and 13, respectively. In round 13, response rate varied by age from 4.2% at ages 18 to 24 years to 24% at ages 65 to 74 years and by IMD decile from 5.1% in the most deprived areas to 20.8% in the least deprived.

Participants were requested to provide a self-administered throat and nose swab (obtained by parent or guardian for children aged 5 to 12 years) following written and video instructions. Swabs were placed into a dry tube (no solution or preservative), refrigerated at home, picked up by courier, and then sent chilled to a single commercial laboratory for testing for SARS-CoV-2 by RT-PCR.

Supplementary Material

Supplementary Materials