SARS-CoV-2 infection following booster vaccination: Illness and symptom profile in a prospective, observational community-based case-control study

Background: Booster COVID-19 vaccines have shown efficacy in clinical trials and effectiveness in real-world data against symptomatic and severe illness. However, some people still become infected with SARS-CoV-2 following a third (booster) vaccination. This study describes the characteristics of SARS-CoV-2 illness following a third vaccination and assesses the risk of progression to symptomatic disease in SARS-CoV-2 infected individuals with time since vaccination

s u m m a r y Background: Booster COVID-19 vaccines have shown efficacy in clinical trials and effectiveness in real-world data against symptomatic and severe illness.However, some people still become infected with SARS-CoV-2 following a third (booster) vaccination.This study describes the characteristics of SARS-CoV-2 illness following a third vaccination and assesses the risk of progression to symptomatic disease in SARS-CoV-2 infected individuals with time since vaccination.Methods: This prospective, community-based, case-control study used data from UK-based, adult (≥18 years) users of the COVID Symptom Study mobile application, self-reporting a first positive COVID-19 test between June 1, 2021 and April 1, 2022.To describe the characteristics of SARS-CoV-2 illness following a third vaccination, we selected cases and controls who had received a third and second dose of monovalent vaccination against COVID-19, respectively, and reported a first positive SARS-CoV-2 test at least 7 days after most recent vaccination.Cases and controls were matched (1:1) based on age, sex, BMI, time between first vaccination and infection, and week of testing.We used logistic regression models (adjusted for age, sex, BMI, level of social deprivation and frailty) to analyse associations of disease severity, overall disease duration, and individual symptoms with booster vaccination status.To assess for potential waning of vaccine effectiveness, we compared disease severity, duration, and symptom profiles of individuals testing positive within 3 months of most recent vaccination (reference group) to profiles of individuals infected between 3 and 4, 4-5, and 5-6 months, for both third and second dose.All analyses were stratified by time period, based on the predominant SARS-CoV-2 variant at time of infection (Delta: June 1, 2021-27 Nov, 2021; Omicron: 20 Dec, 2021-Apr 1, 2022).Findings: During the study period, 50,162 (Delta period) and 162,041 (Omicron) participants reported a positive SARS-CoV-2 test.During the Delta period, infection following three vaccination doses was associated with lower odds of long COVID (symptoms≥ 4 weeks) (OR=0.83,CI[0.50-1.36],p < 0.0001), hospitalisation (OR=0.55,CI[0.39-0.75],p < 0.0001) and severe symptoms (OR=0.36,CI[0.27-0.49],p < 0.0001), and higher odds of asymptomatic infection (OR=3.45,CI [2.86-4.16],p < 0.0001), compared to infection following only two vaccination doses.During the Omicron period, infection following three vaccination

Introduction
Booster vaccination remains a key strategy in response to the ongoing challenges posed by COVID-19.National UK government guidance at the time of writing suggests that a seasonal booster should be offered to certain persons, including older adults and those in defined clinical risk groups . 1 The aim of the national vaccination programme remains to reduce severe disease including hospitalisation and mortality and to prevent overburden of the National Health Service.
3][4][5][6] These in vitro findings are supported by real-world effectiveness studies, which have shown a lower risk of both asymptomatic and symptomatic infection with Omicron variants following booster vaccination. 7,8Lower infection rates may translate to lower healthcare utilisation, and a study across nine US states demonstrated that bivalent vaccines administered after at least 2 monovalent vaccine doses reduced COVID-19 related emergency attendances during the Omicron period. 9onetheless, some people still become infected with SARS-CoV-2 following three or more vaccine doses.A study of over one million older adults from US Veteran Health Administration facilities conducted across Delta and Omicron-predominant periods reported an incidence of breakthrough COVID-19 of 125.0 per 10, 000 persons following a booster vaccine, although notably the incidence of hospitalisation or death was reported as low (8.9 per 10, 000 persons). 10o our knowledge, no studies have analysed the symptom profile of individuals infected with SARS-CoV-2 following three or more vaccine doses and there is little evidence on whether a third vaccine dose reduces the risk of prolonged symptoms ("long COVID").In a previous study using data from over one million COVID Symptom Study participants, we described the characteristics of illness following the primary COVID-19 vaccination course, compared to illness in unvaccinated individuals. 11Vaccination was associated with a reduction in severe illness, and reduction in long-duration symptoms following the second vaccine dose.Almost all individual symptoms were less frequently reported in individuals infected post-vaccination, who were also more likely to be asymptomatic.The exception was sneezing, which was more common in infected individuals after the first vaccine dose.It remains unclear whether a third vaccine confers further protective effects against severe, symptomatic, and long duration disease.
Furthermore, accumulating evidence suggests a waning of vaccine effectiveness against symptomatic and severe illness with time since vaccination.Real-world population studies have suggested that efficacy of the primary vaccination course against severe illness (including hospitalisation and death) may wane from around six months following second vaccination. 12,13][16][17][18] In a retrospective case control study of > 30,000 participants aged over 65 years, Patalon et al. observed a significant waning in the relative protection of the BNT162b2 booster vaccine against the Omicron variant, from 53.4% one month after vaccination to 16.5% three months after vaccination. 19The authors suggested that a further booster dose may be needed to restore immunity in older adults.However, little is known about whether there is a waning of vaccine effectiveness against long duration illness (long COVID).This is important both for individual prognostication and to forecast longer-term healthcare utilisation.
Elucidating disease severity, duration and symptom profiles in individuals infected with SARS-CoV-2 after booster vaccination has clinical importance, facilitating identification of groups to target for ongoing vaccination promotion efforts and intervention, and to forecast medical resource requirements.
This study aimed to: 1. Evaluate markers of illness severity and duration and assess symptom profile in individuals reporting SARS-CoV-2 infection after their third ("booster") vaccination dose, compared to those reporting infection following two vaccine doses; 2. Investigate any change in the effectiveness of vaccination over time in reducing severity, symptoms and duration of illness related to SARS-CoV-2 infection following both booster and second vaccine doses.

Study design and participants
This prospective, community-based, case-control study used data from UK-based, adult (≥18 years) participants of the COVID Symptom Study logged through a free smartphone app developed by Zoe (London, UK) and King's College London (London, UK).The app was launched in the UK on March 24, 2020. 20At registration, each participant reported baseline demographic information (e.g., age, sex, ethnicity, whether healthcare worker), geographic location, and information on health risk factors including comorbidities, lifestyle, frailty, and visits to hospital.Participants were encouraged to selfreport any pre-specified symptoms daily, providing prospective, longitudinal information on incident symptoms.All users were prompted to record any COVID-19 testing results (whether tests were provided via the app or from other sources), and any COVID-19 vaccine(s) and subsequent symptoms.
Inclusion criteria for the study population were: 1) age ≥18 years; 2) living in the UK; 3) second or third vaccine dose received after June 1, 2021; 4) a positive reverse transcription polymerase chain reaction (RT-PCR) or lateral flow antigen test (LFAT) reported at least seven days after vaccination (in the case of multiple positive test results, only the first was selected).We excluded frontline healthcare workers (3.2% 7493/234,951 users) due to differences in vaccination schedule and propensity to test as a result of occupational exposure to SARS-CoV-2.
We considered only positive tests reported after June 1, 2021, when the Delta variant was predominant in the UK (prevalence > 70% of reported infections) and up to April 1, 2022, when the UK government stopped providing coronavirus tests free-of-charge to the general public.Each included participant was at least seven days post-vaccination and had at least 12 weeks of symptom reporting following the positive test.The data census for symptom reporting was June 24, 2022, to allow each user at least 12 weeks of reporting.
To analyse the effect of booster vaccination on post-vaccination infection we compared disease severity, duration, and individual symptoms in users reporting a positive SARS-CoV-2 test at least seven days after a third vaccine dose (cases) to those reporting a positive test at least seven days after a second dose (controls).Controls were matched 1:1 with cases by a Euclidean-based algorithm based on age, sex, Body Mass Index (BMI), time elapsed between the first vaccination and the infection, and test week, a method used in our previously published study. 11,21he following were assessed as indicators of disease severity: (i) self-reported hospitalisation; (ii) reported acute, functionally limiting ("severe") symptoms (two or more of fever, shortness of breath, and severe fatigue), and (iii) asymptomatic infection.For assessment of disease duration, we analysed associations of booster vaccination with reported symptoms lasting both ≥4 weeks, and ≥12 weeks.In analyses of symptom duration, we considered only users who logged using the app at least once per week for 4 weeks or longer after reporting a positive COVID-19 test.
For symptom profile analysis, symptoms reported between 3 days before and up to 14 days after the positive test date were considered, reflecting the acute phase of the disease.This window was used because it might have taken up to 3 days to request a RT-PCR test and receive a result following symptom onset, and symptoms can occur up to 14 days following SARS-CoV-2 exposure. 11,22he full list of symptoms that can be reported using the app is shown in Supplementary Table 1.Symptoms were only included if they had a reported prevalence of more than 5% in the study population.
To assess for any waning of VE, we considered as a reference group participants infected within 3 months, and compared their disease severity, duration, and symptom profiles with participants infected between 3 and 4, 4 and 5, and 5 and 6 months following vaccination, for both third and second vaccine doses.We chose as the reference group individuals with Time Since Vaccination (TSV) < 3 months as antibody levels have been demonstrated to significantly reduce by 3 months after 3rd vaccination. 23,24Cases and controls were matched through a Euclidean-based algorithm based on age, sex, BMI, and test week. 11,21e stratified all analyses into two time periods according to when Delta and Omicron variants of COVID-19 were predominant (> 70%): 1st June 2021-27th November 2021 (Delta) and 20th December 2021-1st April 2022 (Omicron).

Disease severity, duration and symptom profile following booster versus second vaccine dose
To assess the effect of booster vaccination on disease severity, duration and symptom profile, we used univariate logistic regression models adjusted for age, sex, BMI, Index of Multiple Deprivation (IMD), and frailty.IMD is an area-based measure of relative deprivation.Frailty was assessed by the PRISMA7 questionnaire embedded in the app and classified as a binary variable (PRISMA7 ≥3 = frail; PRISMA7 < 3 = not frail). 26,27For disease severity, we considered as outcomes the odds of self-reported: 1) hospitalisation; 2) reporting acute, functionally limiting symptoms, defined as two or more of: fever, shortness of breath, and fatigue during the first 14 days, and 3) reporting no symptoms (asymptomatic).For duration, we considered as outcomes the presence of symptoms reported for ≥4 weeks and ≥12 weeks (post-COVID-19 syndrome).For symptom profile analysis, the outcome was the presence of each individual symptom.

Waning of vaccine effect
To assess for potential waning of VE against symptomatic or severe infection, we used the univariate logistic regression models as described above, comparing disease severity, duration and symptom profiles of individuals testing positive within 3 months of the most recent vaccination (reference group) to profiles of individuals infected between 3 and 4, 4-5, and 5-6 months.

Ethical approval
All app users provided informed consent for data usage for COVID-19-related research.In the UK, the app and study were approved by King's College London's (KCL) ethics committee (REMAS no.18210, reference LRS-19/20-18210).

Role of the funding source
Funders had no role in the design, analysis, or interpretation of the data.Zoe Global, funded in part by the Department of Health and Social Care, made the app available for data collection as a not-forprofit endeavour.

Results
During the Delta and Omicron periods, 50,162 and 162,041 users reported a first positive test for SARS-CoV-2, respectively.Of these, 1910 (Delta period) and 154,057 (Omicron period) participants reported a positive test at least 7 days after the third vaccination dose, and 48,252 (Delta period) and 7984 (Omicron period) reported a first positive test at least 7 days after the second vaccination.These differences are explained by the timing of Omicron relative to the booster vaccination programme.Table 1 shows the demographic characteristics of cases and controls after 1:1 matching.During both periods, the majority of users were female (around 60%) and lived in areas with a greater level of social deprivation.For the Omicron period, controls were significantly younger (p < 0.0001), and there was a significant difference in the number of people with frailty (higher amongst controls).There was no significant difference in the prevalence of individual comorbidities or at least one comorbidity between the two groups.

Disease severity, duration and symptom profile following booster versus second vaccine dose
Before matching, during both Delta and Omicron periods, there was a lower proportion of individuals hospitalised, reporting severe symptoms (two or more of shortness of breath, fever and severe fatigue), and with duration of symptoms ≥12 weeks amongst individuals infected after 3 doses versus after 2 doses.During the Delta period, there was a substantially higher proportion of asymptomatic infections and a lower proportion of individuals with long COVID amongst cases versus controls.During the Omicron period, the proportion of individuals with asymptomatic infections and long COVID was similar amongst cases and controls, although a significantly higher proportion reported severe symptoms (Fig. 1).
During both periods, the most prevalent symptoms (reported by more than 50% of infected participants) were runny nose, headache, fatigue, and sneezing.During the Delta period, the proportion of infected individuals reporting almost all individual symptoms following a third vaccine was lower than following the second dose.During the Omicron period, the proportion of individuals reporting most symptoms was lower following a third vaccine, except for runny nose, sneezing, sore throat, and hoarse voice, which were more frequently reported (Fig. 3).In the univariate matched case-control analysis of symptoms, adjusted for age, BMI, sex, frailty, and IMD, during the Delta period there was a significantly lower likelihood of all individual self-reported symptoms, with ORs ranging from 0.36 and 0.75 (Fig. 4a and Supplementary Table 4).In contrast, during the Omicron period, 18 out of 28 symptoms were significantly less frequently self-reported in individuals vaccinated with 3rd dose.Upper respiratory symptoms were significantly more likely to be self-reported during the Omicron period in individuals vaccinated with 3rd dose (runny nose: OR=1.26,CI [1.18-1.69],p < 0.0001; sore throat: OR=1.17CI[1.10-1.25,p < 0.0001; sneezing OR=1.40,CI[1.32-1.50],p < 0.0001; hoarse voice: OR=1.13,CI= [1.06-1.21],p < 0.000) (Fig. 4b and Supplementary Table 5).Results were consistent when stratified by age group (see Supplementary Fig. 2).

Waning of vaccine effect
During the Delta period, there was a significantly higher odds of symptoms lasting 4 or more weeks (from 1.17 to 1.40) and a decrease in the odds of asymptomatic infection (from 0.83 to 0.65) with increasing TSV.There was no difference in symptom severity with TSV > 4 months (Fig. 5a, Supplementary Table 6).For the Omicron period, there was a clear waning of VE against severe symptoms beyond 6 months (increase in OR of severe symptoms from 1.29 to 1.61), while for asymptomatic infection and illness duration ≥4 weeks there was only a significant waning of VE up to 5 months (decrease in OR from 0.71 to 0.56 for asymptomatic infection and increase from 1.28 and 1.35 for symptoms ≥ 4 weeks).There were no significant associations with other indicators of disease severity (Fig. 5b, Supplementary Table 7).
Regarding symptom profile, during both Delta and Omicron periods, there was a clear waning of VE for almost all individual symptoms (Fig. 6, Supplementary Tables 8 and 9).This was more evident during the Omicron period, although during this period no further waning of VE was observed when compared to 4 < =TSV < =5 months and 5 < =TSV < =6 months.
For both analyses, the same results were obtained when stratified by age group (see Supplementary Figs. 3 and 4).

Discussion
We present data on 155,967 and 56,236 community-based adults in the UK with reported test-confirmed SARS-CoV-2 infection after a third (booster) or second vaccination, respectively.During the period when Delta was the predominant COVID-19 variant in the UK, a third vaccination conferred reductions in disease severity and duration, and a higher proportion of infected individuals reporting asymptomatic illness.The observed protective effect of booster vaccination against long COVID is supported by findings from our previous study of over one million COVID Symptom Study participants, where long duration illness was reduced following a second vaccination dose, at a time when the Delta variant was predominant in the UK. 11In the current study, protective effects of vaccination were also apparent during the Omicron period, with a significant reduction in acute, functionally limiting symptoms, although not in hospitalisation rate.The latter finding contrasts with a finding from a recent US-based study, which reported that booster vaccines were effective in reducing COVID-19-associated Emergency Department and Urgent Care encounters and hospitalisations amongst immunocompetent adults, during a period when Omicron sublineages accounted for the majority of sequenced viral genomes in the US. 9 Notably, the latter study looked at effects of bivalent mRNA vaccines, which were not approved as booster vaccines in the UK until August 2022.
This study adds to existing evidence on the symptom profile of infection following three or more vaccinations.During the period when Delta was the most prevalent variant, following a third vaccination dose, infected individuals were less likely to report almost all individual symptoms, compared with those infected following only two vaccine doses.This is supported by findings from our previous study using COVID Symptom Study app data, which demonstrated lower rates of individual symptom reporting in vaccinated versus unvaccinated individuals. 11Also in keeping with our findings, a study of 1199 US essential and frontline workers demonstrated that recent vaccination with three mRNA vaccine doses was associated with attenuated symptoms and duration of illness. 28In the current study, during the Omicron period there was a reduction in symptoms commonly reported during the earlier stages of pandemic, such as fever and cough, as well as fatigue and headache and a reduction in fever and chills following third dose vaccination was also observed in the study of US frontline workers. 28This is important since these symptoms have previously been shown to be predictive of more severe disease. 29uring the Omicron period, there was increased reporting of certain "upper respiratory" symptoms in infected individuals following a third dose, including sneezing, runny nose, and hoarse voice.This is consistent with findings from our previous study, which found a higher reporting of sneezing following a second vaccine dose. 11This observation is also supported by a recent nationwide study, which found higher reporting of certain symptoms including runny nose and sore throat in fully vaccinated versus unvaccinated individuals, 30 as well as findings from a pre-print Japanese registry-based observational study which found that upper respiratory tract symptom burden was increased following vaccination. 31In that study, vaccination mitigated overall systemic symptoms and risk of severe disease, in keeping with our findings.One potential explanation for this could be "immune priming".Vaccines work by priming the immune system, meaning that should the pathogen be encountered naturally, the immune system is able to react more quickly and effectively.Although upper respiratory symptoms have seldom been described in any studies looking at immune priming following vaccination, they are well documented in inflammatory conditions such as allergic rhinitis. 32uring both Delta and Omicron periods, there was a trend towards waning of VE, with increased symptom reporting with time since vaccination following both second and third vaccine doses.This observation is supported by in vitro studies measuring antibody levels and immunogenicity against COVID-19 following two or three doses of vaccine and by meta-analyses showing reduced effectiveness of vaccines against SARS-CoV-2 infection and symptomatic COVID-19 over time, although protection against severe disease remained high. 5,23,33,34This study adds to existing evidence by suggesting waning of VE against symptomatic illness, and also against long duration illness (symptoms lasting ≥4 weeks).Of note, the proportion of infected individuals reporting ongoing symptoms ≥12 weeks was much lower than symptoms ≥4 weeks, and there were differences in trends in waning of vaccination protection against symptomatic illness with time since vaccination during Omicron between these groups.This supports the suggestion that ongoing symptomatic COVID-19 (4-12 weeks) and post-COVID-19 syndrome (≥12 weeks) may capture different disease states and highlights the need for ongoing standardisation of definitions used for long COVID in clinical practice and research. 35ngoing surveillance of predominant COVID-19 variants and effectiveness of current vaccines against emerging variants remain vital components of the coordinated international response to the ongoing challenges posed by COVID-19.Future research should consider effects of further booster vaccinations (such as the seasonal autumn booster programme) using bivalent vaccines approved in the UK and other comparator settings, on risk of infection, indicators of disease severity, and symptom profile.

Strengths and limitations
This study has several strengths.The mobile application data collection method facilitates collection of daily prospective information on a comprehensive set of symptoms, permitting analysis of both individual symptoms and overall illness duration (although note that necessary data censoring could have underestimated symptom duration for both cases and controls, as some individuals only had 2 weeks of logging after their positive test result).The matching of cases and controls on time since vaccination and timing of the post-vaccination test reduced the potential for bias, although small differences between the groups remained on some matched variables.We acknowledge potential differences in logging by vaccinated individuals or those undertaking regular COVID-19 testing during the study period (for example, if required for work).Access to testing is a potential source of bias, although this was mitigated by only including reported infections up until April 1st, 2022, after which free tests were no longer universally available in the UK.
Our study has some limitations.We were unable to include an unvaccinated control group for comparison, as there were very few unvaccinated individuals enroled at the time of the study.Our findings might not apply at all timepoints post-vaccination, to countries with different recommended vaccination types and schedules, or settings with different proportions of SARS-CoV-2 variants.Additionally, the data were self-reported; recording of comorbidities, test results, and vaccination status might not have been mass index; SD=Standard deviation; IMD=Index of Multiple Deprivation; IMD[1-3] indicates high social deprivation, IMD[4-7] intermediate, IMD[8-10] low; age is in years; comorbidity status=at least one comorbidity; for age and BMI the mean and standard deviation are provided, and for categorical variables the absolute value and percentages (%).*/=Indicates statistically significant/nostatistically significant difference when compared to the control population (Fisher's p < 0.05).

Fig. 1 .
Fig. 1.Proportion (%) of positive individuals in each time period with each severity outcome, stratified by dose and predominant variant.

Fig. 2 .Fig. 3 .
Fig. 2.Odds ratio of asymptomatic infection, duration of symptoms ≥4 weeks, duration of symptoms ≥12 weeks, severe condition (two out of three among severe shortness of breath, fatigue, and fever), and hospitalisation in app participants following booster vaccination vs 2nd dose, adjusted by age, BMI, sex, frailty, and IMD, stratified by variant.

Fig. 4 .
Fig. 4. Odds Ratio of individual symptoms in individuals vaccinated with the third dose versus individuals vaccinated with the second dose adjusted by age, BMI, sex, frailty, and IMD for the Delta period and (b) Omicron period.

Fig. 5 .
Fig. 5. Odds ratio of asymptomatic infection, duration of symptoms ≥4 weeks, severe condition (two out of three among severe shortness of breath, fatigue, and fever), and hospitalisation in individuals vaccinated with second dose during Delta period (a) and third dose during the Omicron period (b) and infected within the first 3 months from vaccination versus individuals vaccinated with the second dose and infected within 3-4, 4-5, and 5-6 months from vaccination adjusted by age, BMI, sex, frailty, and IMD.

Table 1
Demographics and comorbidities of cases and controls for disease severity, duration and symptom profile analysis.