medRxiv preprint doi: https://doi.org/10.1101/2022.08.31.22279264; this version posted September 6, 2022. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 1 Title: Impact of COVID-19 and Effects of Vaccination with BNT162b2 on Patient-Reported 2 Health-Related Quality of Life, Symptoms, and Work Productivity Among US Adult Outpatients 3 with SARS-CoV-2 4 5 Authors: Manuela Di Fusco1*, Xiaowu Sun2, Mary M. Moran3, Henriette Coetzer2, Joann M. 6 Zamparo3, Laura Puzniak3, Mary B. Alvarez4, Ying P. Tabak2, Joseph C. Cappelleri5 7 8 Affiliations: 9 1 Pfizer Inc., Health Economics and Outcomes Research, New York, NY, USA 10 2 CVS Health, Woonsocket, RI, USA 11 3 Pfizer Inc., MDSCA Vaccines, Collegeville, PA, USA 12 4 Pfizer Inc., Field Medical Outcomes and Analytics, New York, NY, USA 13 5 Pfizer Inc., Statistical Research and Data Science Center, Groton, CT, USA 15 * Corresponding author: 16 Manuela Di Fusco 17 Health Economics and Outcomes Research 18 Pfizer, Inc., New York, NY, USA 19 Manuela.difusco@pfizer.com 14 20 21 22 23 1 NOTE: This preprint reports new research that has not been certified by peer review and should not be used to guide clinical practice. medRxiv preprint doi: https://doi.org/10.1101/2022.08.31.22279264; this version posted September 6, 2022. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 24 ABSTRACT 25 Background: Although there is extensive literature on the clinical benefits of COVID-19 26 vaccination, data on humanistic effects are limited. This study evaluated the impact of SARS- 27 CoV-2 infection on symptoms, Health Related Quality of Life (HRQoL) and Work Productivity 28 and Impairment (WPAI) prior to and one month following infection, and compared results 29 between individuals vaccinated with BNT162b2 and those unvaccinated. 30 Methods: Subjects with ≥1 self-reported symptom and positive RT-PCR for SARS-CoV-2 at 31 CVS Health US test sites were recruited between 01/31/2022-04/30/2022. Socio-demographics, 32 clinical characteristics and vaccination status were evaluated. Self-reported symptoms, HRQoL, 33 and WPAI outcomes were assessed using questionnaires and validated instruments (EQ-5D-5L, 34 WPAI-GH) across acute COVID time points from pre-COVID to Week 4, and between 35 vaccination groups. Mixed models for repeated measures were conducted for multivariable 36 analyses, adjusting for several covariates. Effect size (ES) of Cohen's d was calculated to 37 quantify the magnitude of outcome changes within and between vaccination groups. 38 Results: The study population included 430 subjects: 197 unvaccinated and 233 vaccinated with 39 BNT162b2. Mean (SD) age was 42.4 years (14.3), 76.0% were female, 38.8% reported prior 40 infection and 24.2% at least one comorbidity. Statistically significant differences in outcomes 41 were observed compared with baseline and between groups. The EQ-Visual analogue scale 42 scores and Utility Index dropped in both cohorts at Day 3 and increased by Week 4, but did not 43 return to pre-COVID levels. The mean changes were statistically lower in the BNT162b2 cohort 44 at Day 3 and Week 4. The BNT162b2 cohort reported lower prevalence and fewer symptoms at 45 index date and Week 4. At Week 1, COVID-19 had a large impact on all WPAI-GH domains: 46 the work productivity time loss among unvaccinated and vaccinated was 65.0% and 53.8%, and 2 medRxiv preprint doi: https://doi.org/10.1101/2022.08.31.22279264; this version posted September 6, 2022. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 47 the mean activity impairment was 50.2% and 43.9%, respectively. With the exception of 48 absenteeism at Week 4, the BNT162b2 cohort was associated with statistically significant less 49 worsening in all WPAI-GH scores at both Week 1 and 4. 50 Conclusions: COVID-19 negatively impacted HRQoL and work productivity among mildly 51 symptomatic outpatients. Compared with unvaccinated, those vaccinated with BNT162b2 were 52 less impacted by COVID-19 infection and recovered faster. 53 54 Keywords: COVID-19, SARS-CoV-2, HRQoL, WPAI, Quality of Life, COVID-19 symptoms, 55 BNT162b2, humanistic 3 medRxiv preprint doi: https://doi.org/10.1101/2022.08.31.22279264; this version posted September 6, 2022. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 56 BACKGROUND 57 The impact of the COVID-19 pandemic on the sustainability of quality of life of patients has 58 been reported globally [1-4]. The prolonged multisystem symptoms associated to SARS-CoV-2 59 infection can negatively affect daily activities, ability to work, and social interactions, leading to 60 poor health-related quality of life (HRQoL) [1-4]. 61 The introduction of COVID-19 vaccination has significantly impacted the COVID-19 response, 62 and evidence regarding the efficacy, safety and effectiveness of vaccination is extensive [5]. 63 However, there is limited research on the potential benefits of vaccination on physical, mental, 64 social, emotional functioning and economic well-being. Most of the studies assessing humanistic 65 outcomes of COVID-19 infection have been limited to inpatients [1, 2, 6] were conducted 66 outside of the US or focused on specific disease states and organ-specific functions [7-9]. 67 Leveraging a US national retail pharmacy SARS-CoV-2 test database and using validated 68 patient-reported outcome measures (PROMs), this study assessed COVID-19 symptoms, 69 HRQoL and WPAI prior to through one month following SARS-CoV-2 infection in outpatients, 70 and compared results between unvaccinated individuals and those vaccinated with BNT162b2. 71 72 METHODS 73 Study Design and Participants 74 The source population consisted of individuals testing for SARS-CoV-2 at one of ~5,000 CVS 75 Health test sites across the US. As part of the registration process for scheduling a SARS-CoV-2 76 test at CVS Health, individuals are required to complete a screening questionnaire including 77 demographics, symptoms, comorbidities, and vaccination status. The screening variables and 4 medRxiv preprint doi: https://doi.org/10.1101/2022.08.31.22279264; this version posted September 6, 2022. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 78 RT-PCR test results are loaded in an analytic dataset, where ~80-90% of test results are reported 79 within 2-3 days. Leveraging this analytic platform, this study was designed as a prospective 80 survey-based patient-reported outcomes study targeting adults ≥18 with a positive RT-PCR test 81 result and self-reporting at least one symptom. These individuals were emailed an invitation as 82 soon as the test results became available, no later than 4 days from testing. The email invitation 83 directed the potential participants to an e-consent website to learn about the study, survey 84 schedule and informed consent. Figure 1 summarizes the study design. Recruitment of 85 participants was carried out between 01/31/2022 and 04/30/2022 (Ct.gov NCT05160636). 86 87 Data Sources and Variables 88 Baseline characteristics and symptoms 89 Baseline characteristics of the participants were obtained via the CVS Health pre-test screening 90 questionnaire. These included self-reported demographics, comorbidities (including 91 immunocompromised status), COVID-19 vaccination history, social determinants of health 92 including the Social Vulnerability Index (SVI), work and/or residency in a high-risk or 93 healthcare setting, and symptoms. The list of baseline COVID-19 symptoms was based on the 94 CDC [10]. 95 96 Exposure groups 97 Immunocompetent participants were considered fully vaccinated with BNT162b2 if they self- 98 reported receipt of 2 doses of BNT162b2 ≥ 14 days of SARS-CoV-2 testing. They were 99 considered partially vaccinated if reporting receipt of a single dose and boosted if reporting 100 receipt of 3 doses. Participants self-reporting an immunocompromising condition and receipt of 5 medRxiv preprint doi: https://doi.org/10.1101/2022.08.31.22279264; this version posted September 6, 2022. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 101 3 doses were considered fully vaccinated (i.e., 3-dose primary series completion); if reporting 4 102 doses, they were considered boosted. Participants were considered unvaccinated if they did not 103 report any COVID-19 vaccine dose prior to testing. Heterologous schedules were excluded. 104 105 HRQoL 106 To assess HRQoL, we used the validated EQ-5D-5L questionnaire [11, 12]. On the day of 107 enrollment, consented participants completed the EQ-5D-5L questionnaire twice, using two 108 versions: a modified version where all the questions were past tense to retrospectively assess pre- 109 COVID-19 baseline QoL, and the standard version in present tense to assess current QoL. To 110 minimize responder bias, the order of administration of the two versions was random. 111 Subsequent completion was requested at one month (short-term study design in Figure 1). The 112 EQ-5D-5L results at each time point were converted into the Utility Index (UI) using the US- 113 based weights by Pickard et al [12, 13]. 114 115 Work Productivity and Activity Impairment 116 To measure impairments in both paid work and unpaid work, we used the Work Productivity and 117 Activity Impairment General Health V2.0 (WPAI:GH) measure [14, 15]. Participants were asked 118 to complete this questionnaire twice, seven days after their RT-PCR test: once referencing seven 119 days prior to COVID-19 symptom onset and an additional assessment referencing the past seven 120 days. Similar to the EQ-5D-5L, subsequent completion of the WPAI was requested at one month 121 (Figure 1). Four WPAI scores were computed at each time point: percent of worktime missed 122 (absenteeism), percent of impairment while working (presenteeism), percent of work 6 medRxiv preprint doi: https://doi.org/10.1101/2022.08.31.22279264; this version posted September 6, 2022. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 123 productivity loss (considering both absenteeism and presenteeism), and percent of activity 124 impairment. Only employed subjects were included for work productivity analyses. 125 126 Post-COVID 19 Symptoms and Vaccination Status Update 127 To supplement the pre-test screening questionnaire and enable the collection of on-going or new 128 symptoms after the acute phase, participants were sent an additional survey four weeks following 129 the test asking to complete a checklist of COVID-19 related symptoms based on the CDC list 130 [16], To confirm vaccination status, participants’ subsequent responses to vaccination date 131 questions were compared with their index responses; if responses did not match, the information 132 was queried and adjudicated, and the latest information was typically used. 133 134 Statistical Analysis 135 Descriptive statistics were used to analyze participant characteristics at baseline. Continuous 136 variables were described using means and standard deviations. Categorical variables were 137 reported using number and percentage distributions. For continuous variables, t-tests were used 138 to test difference in means and Wilcoxon tests were used to test difference in medians. For 139 categorical variables, chi-square tests were used to test differences between groups When cell 140 frequency was less than 5, Fisher’s exact tests were used for 2-by-2 tables and Freeman-Halton 141 tests for r-by-c tables [17, 18]. P values were all two-sided and not adjusted for multiplicity. 142 Mixed models for repeated measures (MMRM) [19] were used to estimate the magnitude of 143 COVID-19 impact on HRQoL and WPAI over time. Models included variables of time, self- 144 reported SARS-CoV-2 vaccination status, and interaction of time by vaccination status, as well 145 as covariates of participant pre-COVID-19 symptom onset score, sociodemographic 7 medRxiv preprint doi: https://doi.org/10.1101/2022.08.31.22279264; this version posted September 6, 2022. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 146 characteristics (age, sex, regions, social vulnerability, race/ethnicity, high risk occupations), 147 previously tested positive for COVID-19, severity of acute illness (number of symptoms reported 148 on index date), and immunocompromised status. Assessment time was fitted as a categorical 149 covariate and a repeated effect (repeated by subject). Least squares mean (LS mean) and 150 standard errors of PRO scores for each time point of assessment were calculated. Per guidelines, 151 no adjustment was made for missing data when scoring the EQ-5D-5L UI and WPAI [11, 15]. 152 Missing data at each time were not imputed. All available data were included in the analysis. 153 154 Cohen’s d, or a variation of it, was calculated to assess the magnitude of score change from 155 baseline within the BNT162b2 vaccinated cohort and, separately, the unvaccinated cohort, as 156 well as the difference between BNT162b2 and unvaccinated cohorts [20, 21]. Specifically, 157 within-cohort effect size (ES) was calculated as mean change from baseline to follow-up, divided 158 by the standard deviation of change scores from baseline to follow-up. Between-cohort ES was 159 calculated as the difference in mean changes from baseline between cohorts, divided by the 160 pooled standard deviation of change scores. When calculating model based ESs, the numerators 161 were the predicted mean change from the model for within-cohort ESs, and predicted differences 162 from the model for between-cohort ESs. Denominators were the corresponding observed 163 standard deviations. Values of 0.2, 0.5, and 0.8 standard deviation (SD) units represent small, 164 medium, and large ES, respectively. These cut-off estimates have been widely used to establish 165 important differences in HRQoL studies [22]. As such, we considered the magnitude of 166 (standardized) effect sizes of at least 0.20 SD units as important or meaningful differences in 167 gauging the magnitude of within-patient change and between-group differences. All data 168 obtained were collected and analyzed with SAS Version 9.4 (SAS Institute, Cary, NC). The 8 medRxiv preprint doi: https://doi.org/10.1101/2022.08.31.22279264; this version posted September 6, 2022. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 169 study followed the Strengthening the Reporting of Observational Studies in Epidemiology 170 (STROBE) reporting guideline [23]. 171 172 RESULTS 173 Baseline characteristics 174 A total of 39,889 eligible candidates were outreached. Of those, 676 consented and completed 175 the first survey, for a consent rate of 1.7%. Compared with individuals in the CVS Health 176 analytic dataset who did not participate in our study, the study sample was over-represented by 177 women and Caucasians, with slightly more individuals vaccinated and with comorbidities 178 (Supplemental Table 1). The final study population included 430 subjects (Figure 2). 100% 179 completed the EQ-5D-5L questionnaire at pre-COVID-19 baseline and at Day 3, and 77.0% 180 completed it at Week 4. The WPAI-GH questionnaire was completed by 88.1% of the 181 participants at pre-COVID-19 baseline, 88.1% at Week 1 and 76.9% at Week 4. 182 The sociodemographic characteristics of the baseline participants are shown in Table 1. Overall, 183 the mean (SD) age was 42.4 (14.3), 76% were female, 68.6% Caucasian, 58.7% from Southern 184 US. There were 24.2% participants who reported ≥1 comorbidities, including 4.4% with 185 immunocompromising conditions and 39% reported a previous COVID-19 infection. 186 About 46% (197) were unvaccinated and 54% (233) were vaccinated with BNT162b2; of those, 187 respectively 140 (60%) and 93 (40%) received 2 and 3 doses. Compared with unvaccinated, 188 BNT162b2 participants were comparable with respect to gender, working and living settings, and 189 comorbidities, slight older with mean age 43.7 vs. 40.9 (p=0.049); living in less vulnerable area 190 with lower mean social vulnerability index (0.40 vs. 0.49, P<<0.001); and slight differences in 9 medRxiv preprint doi: https://doi.org/10.1101/2022.08.31.22279264; this version posted September 6, 2022. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 191 race/ethnicity and region. In the vaccinated group, mean (SD) time since vaccination before 192 infection was 186 (105) days. 193 At index date, the most reported acute symptoms were respiratory and systemic. BNT162b2 194 vaccinated participants reported fewer overall acute COVID-19 symptoms on average than 195 unvaccinated participants, mean 5.1 vs. 5.6, P=0.034 (Table 1). Directionally, the proportions of 196 all systemic and GI-related symptoms were numerically lower in the BNT162b2 cohort. Relative 197 to unvaccinated, those vaccinated with BNT162b2 reported significantly fewer symptoms of 198 fever (30.5% vs. 47.2% P<0.001), chills (42.9% vs. 57.4%, P=0.003), muscle or body aches 199 (49.4% vs. 59.4%, P=0.038), and diarrhea (15.9% vs. 25.9%, P=0.010), but more congestion or 200 runny nose (80.7% vs. 68.0%, P=0.003). 201 202 Post-COVID-19 symptoms 203 At Week 4, the mean number of symptoms was statistically lower in the BNT162B2 cohort (2.5 204 vs. 3.7, p=0.002). The overall prevalence decreased over time too, especially fever, cough, 205 headache, fatigue, diarrhea, muscle pain; however, ~70% of participants still reported at least 1 206 post-COVID-19 symptom. Directionally, the proportions of all symptoms were numerically 207 lower in the BNT162b2 cohort. Symptoms of worsening after physical or mental activities 208 (10.3% vs. 20.6%), general pain/discomfort (11.4% vs. 19.4%), change in smell or taste (10.9% 209 vs. 20.6%), headache (16.0% vs. 25.2%), sleep problems (20.0% vs. 29.7%), mood changes 210 (7.4% vs. 14.8%), memory loss (6.3% vs. 17.4%) and diarrhea (3.4% vs. 11.0%) were 211 statistically significant (P < 0.05) (Table 2). 212 213 Health-Related Quality of life 10 medRxiv preprint doi: https://doi.org/10.1101/2022.08.31.22279264; this version posted September 6, 2022. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 214 Utility Index scores 215 Mean pre-COVID-19 baseline UIs did not differ between the BNT162b2 and unvaccinated 216 cohorts, respectively 0.924 and 0.918 (P=0.547). COVID-19 infection had a detrimental effect 217 on the HRQoL of participants, especially during the acute episode (Day 3). In both the 218 BNT162b2 and the unvaccinated cohorts, UIs were lower at Day 3 and Week 4 relative to pre- 219 COVID-19. While UIs improvement was observed over time, the UI did not return to pre- 220 COVID levels at Week 4 (Table 3). 221 The BNT162B2 cohort was less impacted than the unvaccinated cohort, at both Day 3 and Week 222 4. After controlling for pre-COVID baseline score and other covariates, the least-square estimate 223 UI scores at Day 3 were, respectively 0.77 and 0.84 in the unvaccinated and BNT162B2 cohorts 224 (Table 4). Moderate ESs of, respectively, 0.64 and 0.49 were observed from baseline. At Week 225 4, the least-square estimate UI scores were, respectively, 0.86 and 0.90. Small-to-moderate ESs 226 of, respectively, 0.38 and 0.13 were observed from baseline. The differences between the two 227 groups were statistically significant (P<0.05). (Table 4) Small-to-medium ESs between cohorts 228 were observed and were 0.36 and 0.32 for Day 3 and Week 4, respectively. (Table 4, 229 Supplemental Figure 1 and 2) 230 231 EQ-VAS 232 The pattern of EQ-VAS scores was similar to that observed for UIs. Mean pre-COVID-19 233 baseline EQ-VAS were similar for the BNT162b2 and unvaccinated cohorts, respectively 86.9 234 and 87.8 (P=0.414) (Table 3). Similar to the UIs, the pre-COVID EQ-VAS were rated relatively 235 high by the participants, indicating a generally healthy cohort. The least-square estimate EQ- 236 VAS scores for the BNT162b2 and unvaccinated cohorts were, respectively, 76.2 and 72.6 at 11 medRxiv preprint doi: https://doi.org/10.1101/2022.08.31.22279264; this version posted September 6, 2022. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 237 Day 3 and 85.0 and 81.6 at Week 4. After controlling for pre-COVID-19 baseline score and 238 other covariates, the least-square estimates of change from pre-COVID-19 baseline in EQ VAS 239 for the BNT162B2 and the unvaccinated cohort were -11.1 and -14.8, respectively on Day 3, and 240 -2.3 and -5.7, respectively at Week 4. COVID-19 had a large adverse impact on EQ-VAS with 241 an ES of -0.89 for BNT162B2 cohort and -0.86 for Unvaccinated cohort on Day 3, and small ES 242 (-0.22) for BNT162B2 cohort and approaching medium ES (-0.42) for Unvaccinated cohort at 243 Week 4. BNT162B2 cohort was associated with 3.6 (P=0.013) on Day 3 and 3.4 (P=0.016) at 244 Week 4 less drop in EQ VAS than the Unvaccinated cohort. The ESs between cohorts were 245 small yet relevant, being 0.25 and 0.28 for Day 3 and Week 4, respectively (Table 4, 246 Supplemental Figure 1 and 2). 247 248 EQ-5D-5L dimensions 249 The health status of the study participants according to the dimensions of EQ-5D-5L is reported 250 in Figure 3 and Supplemental Table 2. In both groups, at Day 3, over half of the cohort reported 251 problems in usual activities, pain/discomfort and anxiety/depression, while the vast majority 252 reported no or slight problems in mobility and self-care. At Week 4, the vast majority continued 253 to report no or slight problems with mobility, self-care, as well as for usual activities; most 254 reported no, slight or moderate problems with pain/discomfort and anxiety/depression. 255 BNT162b2 cohort had lower mean responses across all 5 domains at both Day 3 and Week 4 256 relative to unvaccinated. 257 258 Work Productivity and Activity Impairment 12 medRxiv preprint doi: https://doi.org/10.1101/2022.08.31.22279264; this version posted September 6, 2022. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 259 Approximately 65% of participants reported being currently employed at baseline (155 in the 260 BNT162b2 cohort and 129 unvaccinated), and were eligible to complete the absenteeism, 261 presenteeism and work-productivity loss questions. At Week 1, COVID-19 had a large impact on 262 all four WPAI-GH domains for both the unvaccinated and BNT162b2cohort. The mean time loss 263 due to absenteeism was, respectively, 65.0% and 45.6%; the mean time loss due to presenteeism 264 was, respectively, 46.8% and 38.4%; the mean time of work productivity loss was 65.0% and 265 53.8%, and the mean time of activity impairment was 50.2% and 43.9%. All within-cohort ESs 266 were > 0.8, which are considered large effects (Table 3). After controlling for pre-COVID-19 267 baseline score, and other covariates, the BNT162b2 cohort was associated with less worsening in 268 WPAI-GH scores. Small-to-medium ESs were observed for work-related scores (absenteeism - 269 0.50, presenteeism -0.26, and work productivity loss -0.32) between the BNT162b2 cohort and 270 the unvaccinated cohort (Table 4). At Week 4, the mean time loss dropped across all four 271 domains. The time loss due to absenteeism dropped substantially; the change from baseline in 272 absenteeism was not found to be statistically significant between the BNT162b2 cohort and the 273 unvaccinated cohort. Small-to-medium ESs were observed for presenteeism (-0.38) work 274 productivity loss (-0.29), and activity impairment (-0.34) between the BNT162b2 cohort and the 275 unvaccinated cohort (Table 4, Supplemental Figure 3 and 4). 276 277 DISCUSSION 278 The impacts of SARS-CoV-2 infection go beyond its clinical outcomes. 279 We found that mild acute infection can negatively impact the humanistic outcomes for up to four 280 weeks post infection. Shortly after infection, the UI and EQ-VAS HRQoL scores dropped from 281 pre-COVID, and over half of the study population reported problems in usual activities, 13 medRxiv preprint doi: https://doi.org/10.1101/2022.08.31.22279264; this version posted September 6, 2022. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 282 pain/discomfort and anxiety/depression. At Week 1, the work productivity and activity 283 impairment time loss were over 50%. At Week 4, both the HRQoL and WPAI scores improved, 284 although they did not return to pre-COVID levels. Individuals vaccinated with BNT162b2 were 285 less impacted and recovered faster than unvaccinated individuals. Multivariable analyses showed 286 that BNT162b2 was significantly associated with higher EQ-VAS and UI scores, less symptoms 287 and better WPAI scores, except for absenteeism at Week 4. 288 There is limited evidence measuring the health-related wellbeing of non-hospitalized individuals 289 affected by COVID-19 [1-4]. To our knowledge, this is the first report measuring the impact of 290 COVID-19 on the HRQoL and WPAI among a national sample of outpatients in the United 291 States. In contrast to our study, previous research that used EQ-5D scales to measure COVID-19 292 impact on the HRQoL reported mean UI scores ranging from 0.61 to 0.86 depending on the 293 hospitalization treatment and time since discharge [1, 2]. The EQ-VAS scores ranged from 50.7 294 to 70.3 [1, 2]. In our study, the HRQoL scores at Day 3 and Week 4 among outpatients with mild 295 disease are higher than those, likely due to the different study populations and periods. In a small 296 US study assessing the impact of COVID-19 on WPAI ~4 months post-infection among subjects 297 enrolled in clinical trials before the introduction of vaccines, 46% of the non-hospitalized 298 patients reported health-related impairment in daily activities [24]. Among the employed, 11.5% 299 missed work and 38.9% reported impairment at work due to health. In our study, all the WPAI 300 scores among unvaccinated at Week 1 are higher, and those at Week 4 similar or lower than 301 those, likely due to the different cut-off, study populations, periods and design. 302 HRQoL and WPAI studies are scarce in COVID-19 related vaccination research. To our 303 knowledge, this is the first report on the effects of BNT162b2 on these patient-centric outcomes. 14 medRxiv preprint doi: https://doi.org/10.1101/2022.08.31.22279264; this version posted September 6, 2022. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 304 These results indicate an additive benefit beyond vaccine effectiveness that should be explored 305 further. 306 Strengths of this study include the nationally representative real-world source population of 307 mildly symptomatic outpatients, the prospective collection of the primary outcomes via validated 308 instruments, and the representativeness of the employed population for work productivity 309 analyses (~65% of the cohort). The age distribution of study participants was comparable with 310 the non-enrolled tested population (p=0.076 for mean age, Supplemental Table 1) and with CDC 311 research in non-hospitalized adults: 91.6% were between 18 and 64 and 8.4% were 65 or older, 312 which was quite similar to Hernandez-Romieu [25], 92.7% and 7.3%, respectively. In the 313 vaccinated group, the time between vaccination and breakthrough infection (mean: ~6 months) 314 was consistent with literature on vaccine-induced duration of protection [5]. 315 316 317 318 319 320 321 322 323 The study findings are subject to limitations. All the data analyzed was self-reported and may be subject to error, missingness, recall bias, social desirability bias, and selection bias associated with survey drop-out. Out of 430 participants completing Day 3 survey, 12% (51/430) missed Week 1 and 23% (99/430) missed Week 4 survey. The drop in responses may partly be the result of responders’ fatigue, and/or recovered cases not returning to follow-up surveys. Female and older age (≥30 years) were found to be more likely to miss follow-up surveys. However, after taking into account several variables, the model predicting missingness indicated that missingness was not associated with vaccination status, nor with HRQoL in terms of EQ VAS on Day 3 and its change from pre-COVID-19 baseline ( 15 medRxiv preprint doi: https://doi.org/10.1101/2022.08.31.22279264; this version posted September 6, 2022. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 324 Supplemental Table 4). 325 The study population differed from non-enrolled tested outpatients in being predominantly 326 female, white, with a higher vaccination uptake, and slightly more comorbidities (Supplemental 327 Table 1). The female over-representation is in line with prior research indicating that women are 328 more likely to contribute to health research surveys [2]. In our study, ~24% of participants 329 reported at least one comorbidity, in contrast to 20% among the non-enrolled tested and ~35% in 330 Hernandez-Romieu (2021) [25]. Various models were fit to account for potential effects due to 331 sociodemographic factors and comorbidities. These adjusted effect sizes between BNT162b2 and 332 unvaccinated cohorts were similar and consistent with those calculated from observed data 333 (unadjusted effect sizes). 334 The pre-COVID baseline scores were slightly higher than US population norms. The mean EQ- 335 VAS and UI were 86.9 and 0.924 for the BNT162b2 cohort, and 87.8 and 0.918 for the 336 unvaccinated cohort, respectively. Cha et al (2019) reported a mean EQ-VAS of 84.6 for the U.S. 337 general population [12]. Jiang et al (2021) reported the US population norm of EQ-VAS as 80.4 338 and the mean EQ-5D-5L UI as 0.851 [26]. The healthy pre-infection status of the study 339 population and the potential for retrospective recall bias may partially explain the differences. A 340 modified EQ-5D-5L questionnaire was used to retrospectively measure pre-COVID-19 baseline 341 data; despite literature suggesting concordance between prospective and retrospective 342 measurements of EQ-5D-5L [27, 28], and a good correlation between assessment of baseline 343 before the index date and recall assessment of the baseline after index date, there is currently no 344 information regarding the recall application of the EQ-5D-5L for COVID-related studies. 345 The pre-COVID-19 values for abseenteism and presenteeism were 3.1% and 9.5% in the 346 BNT162b2 cohort, and were generally in line with Tundia et al (2015) [29], whom reported 4% 16 medRxiv preprint doi: https://doi.org/10.1101/2022.08.31.22279264; this version posted September 6, 2022. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 347 absenteeism and 10% presenteeism for the US general population. The reported values were 348 slightly higher among unvaccinated, respectively 11.7% and 9.4%. 349 350 There is currently no standard definition of minimal clinically important difference of PROs in 351 COVID-19 research. We used effect size (ES) of Cohen’s d to quantify the magnitude of score 352 change from baseline within the BNT162b2 vaccinated cohort and the unvaccinated cohort, as 353 well the difference between these two cohorts [20]. An effect size ES of 0 between groups 354 indicates that the average (typical) treated vaccinated person has a score that is no different from 355 the typical control person; equivalently, scores of the typical treated person are more favorable 356 than 50% of the individual scores in the control group, meaning no incremental benefit (a coin 357 toss as to which intervention is better). If the vaccinated cohort is presumed more effective than 358 the unvaccinated cohort, effect sizes thresholds of 0.2, 0.3, 0.5, and 0.8 (in absolute value) 359 indicate that, based on the standardized normal distribution, the score of the typical person in the 360 vaccinated cohort is more favorable than 58% (8% incremental benefit), 62% (12% incremental 361 benefit), 69% (19% incremental benefit), and 70% (29% increment benefit) of the scores from 362 individuals in the unvaccinated cohort. For example, from baseline to Week 1, the increase in the 363 absenteeism WPAI score of the typical person in the vaccinated cohort was less (more favorable) 364 than the corresponding change in 69% of individuals in the unvaccinated group (effect size = - 365 0.5). Depending on the type of outcome, the same type of effect size interpretation for between 366 cohorts can be given within cohort. For instance, from baseline to Week 4, the quality of life EQ- 367 VAS score of the typical person in the vaccine cohort at Week 4 was less (worse) than 59% of 368 the individual scores in the vaccine cohort at baseline (effect size = -0.22); the quality of life EQ- 17 medRxiv preprint doi: https://doi.org/10.1101/2022.08.31.22279264; this version posted September 6, 2022. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 369 VAS score of the typical person in the unvaccinated cohort at Week 4 was less (worse) than 66% 370 of the individual scores in the unvaccinated cohort at baseline (effect size = -0.42). 371 372 The study did not assess the impact on pediatrics, caregivers, long-term outcomes (e.g., “Long 373 COVID”), and the data was collected during Omicron predominance in the US. Therefore, these 374 findings may not be generalizable to prior or future variants, other countries, time periods and 375 populations that were excluded. COVID-19 sequalae can affect a substantial portion of patients, 376 with long-term consequences for their health, continuity of care and ability to work [1, 2]. 377 Persistent symptoms and work impairment were reported ~4 months after infection among non- 378 hospitalized US patients enrolled in clinical trials [24]. Continued follow-up studies covering 379 longer time periods may inform whether the protection provided by COVID-19 vaccination 380 extends beyond the acute phase. Only generic PROMs were used in this study; COVID-19 381 disease-specific instruments are under development [30, 31], warranting research on their 382 implementation. The PROMs omitted questions on vaccine adverse events. The mean 6-month 383 interval between vaccination and breakthrough infection and a medical review ruled out cases of 384 residual symptoms from vaccination. Research on the impact of vaccine adverse events on 385 HRQoL is warranted. 386 Lastly, the study adopted an observational design, which is limited in establishing causal 387 relationships. Future studies using different data collection methods could corroborate the study 388 findings, including those with more rigorous study design. 389 Consistent with literature, our study found that COVID-19 is detrimental to mildly symptomatic 390 COVID-19 patients. The findings provide a meaningful contribution suggesting that the ability 18 medRxiv preprint doi: https://doi.org/10.1101/2022.08.31.22279264; this version posted September 6, 2022. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 391 of BNT162b2 to reduce adverse outcomes from COVID-19 disease can translate to lessening the 392 broad impact of COVID-19 and improvements in quality of life, work productivity and activity. 393 394 CONCLUSION 395 This study found that mild COVID-19 infection at a time of Omicron predominance adversely 396 impacted the HRQoL, daily activity and work productivity of patients. This detrimental effect 397 improved over time, although it persisted for at least one month post infection. Compared with 398 unvaccinated, those vaccinated with BNT162b2 were less impacted and recovered faster. These 399 findings advance research on COVID-19 associated humanistic outcomes and the potential effect 400 of BNT162b2 in lessening the loss of HRQoL, daily activity and work productivity due to 401 COVID-19. The results can inform the estimation of quality-adjusted life years and indirect cost 402 savings in health economic studies. 403 404 List of Abbreviations 405 CDC: Centers for Disease Control and Prevention; CI: Confidence interval; EQ-5D-5L: 406 EuroQoL Group 5 dimension and 5 level questionnaire; ES: Effect size; GH: General Health; 407 HRQoL: Health Related Quality of Life; MMRM: Mixed models for repeated measures; SD: 408 standard deviation; SE: standard error; STROBE: the Strengthening the Reporting of 409 Observational Studies in Epidemiology; UI: utility index; VAS: Visual analogue scale; WPAI: 410 Work Productivity and Impairment. 411 412 DECLARATIONS: 413 Ethics Approval and consent to participate 19 medRxiv preprint doi: https://doi.org/10.1101/2022.08.31.22279264; this version posted September 6, 2022. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 414 This study was approved by the Sterling IRB, Protocol #C4591034. Participation in the study 415 was voluntary and anonymous. Consent was obtained electronically via the CVS Health E- 416 Consent platform. Participants were informed of their right to refuse or withdraw from the study 417 at any time. Participants were compensated for their time. 418 Consent for publication 419 All authors have given their approval for this manuscript version to be published. 420 Availability of data and material 421 Aggregated data that support the findings of this study are available upon reasonable request 422 from the corresponding author MDF, subject to review. These data are not publicly available due 423 to them containing information that could compromise research participant privacy/consent. 424 Competing interests 425 MDF, MMM, JMZ, LP, MBA and JCC are employees of Pfizer and may hold stock or stock 426 options of Pfizer. XS and HC are employees of CVS Health and may hold stock of CVS health. 427 YPT was employee of CVS Health when current study was conducted. 428 Funding 429 This study was sponsored by Pfizer Inc. 430 Authors’ contributions 431 All named authors meet the International Committee of Medical Journal Editors (ICMJE) criteria 432 for authorship for this article. All authors contributed to study conception and design, data 433 acquisition, analysis, and interpretation, drafting and revising of the manuscript. 434 Acknowledgements 435 The authors acknowledge Alejandro Cane, Deepa Malhotra and Nancy Gifford (Pfizer 436 employees), Joseph Ferenchick, Shiyu Lin and Shawn Edmonds (CVS Health employees) for 20 medRxiv preprint doi: https://doi.org/10.1101/2022.08.31.22279264; this version posted September 6, 2022. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 437 specific contributions to this research project. Editorial support was provided by Laura Anatale- 438 Tardiff and Leena Samuel at CVS Health and was funded by Pfizer. 439 440 REFERENCES 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. Poudel AN, Zhu S, Cooper N, Roderick P, Alwan N, Tarrant C, Ziauddeen N, Yao GL (2021) Impact of Covid-19 on health-related quality of life of patients: A structured review. PLoS One. 16(10):e0259164. Nandasena H, Pathirathna M, Atapattu A, Prasanga P (2022) Quality of life of COVID 19 patients after discharge: Systematic review. PloS one. 17(2):e0263941. Figueiredo EAB, Silva WT, Tsopanoglou SP, Vitorino DFdM, Oliveira LFLd, Silva KLS, Luz HDH, Ávila MR, Oliveira LFFd, Lacerda ACR (2022) The health-related quality of life in patients with post-COVID-19 after hospitalization: a systematic review. Revista da Sociedade Brasileira de Medicina Tropical. 55. Amdal CD, Pe M, Falk RS, Piccinin C, Bottomley A, Arraras JI, Darlington AS, Hofsø K, Holzner B, Jørgensen NMH (2021) Health-related quality of life issues, including symptoms, in patients with active COVID-19 or post COVID-19; a systematic literature review. Quality of Life Research. 30(12):3367-81. International Vaccine Access Center, Johns Hopkins Bloomberg School of Public Health, World Health Organization. Results of COVID-19 Vaccine Effectiveness & Impact Studies: An Ongoing Systematic Review. https://view-hub.org/sites/default/files/202204/COVID19_VE_and_Impact_Lit_Review_Methods.pdf Accessed August 15, 2022. Arab-Zozani M, Hashemi F, Safari H, Yousefi M, Ameri H (2020) Health-related quality of life and its associated factors in COVID-19 patients. Osong public health and research perspectives. 11(5):296. Petersen EL, Goßling A, Adam G, Aepfelbacher M, Behrendt C-A, Cavus E, Cheng B, Fischer N, Gallinat J, Kühn S (2022) Multi-organ assessment in mainly non-hospitalized individuals after SARS-CoV-2 infection: the Hamburg City Health Study COVID programme. European Heart Journal. 43(11):1124-37. McKay SC, Lembach H, Hann A, Okoth K, Anderton J, Nirantharakumar K, Magill L, Torlinska B, Armstrong M, Mascaro J (2021) Health‐related quality of life, uncertainty and coping strategies in solid organ transplant recipients during shielding for the COVID‐19 pandemic. Transplant International. 34(11):2122-37. Talman S, Boonman-de Winter L, De Mol M, Hoefman E, Van Etten R, De Backer I (2021) Pulmonary function and health-related quality of life after COVID-19 pneumonia. Respiratory Medicine. 176:106272. Centers for Disease Control and Prevention. Updates - Symptoms of COVID-19. https://www.cdc.gov/coronavirus/2019-ncov/symptoms-testing/symptoms.html Accessed August 1 2021. EuroQol Research Foundation. (2019) EQ-5D-5L User Guide, Version 3.0. https://euroqol.org/publications/user-guides Accessed August 5, 2021. 21 medRxiv preprint doi: https://doi.org/10.1101/2022.08.31.22279264; this version posted September 6, 2022. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. Cha AS, Law EH, Shaw JW, Pickard AS (2019) A comparison of self-rated health using EQ-5D VAS in the United States in 2002 and 2017. Quality of Life Research. 28(11):3065-9. Pickard AS, Law EH, Jiang R, Pullenayegum E, Shaw JW, Xie F, Oppe M, Boye KS, Chapman RH, Gong CL (2019) United States valuation of EQ-5D-5L health states using an international protocol. Value in Health. 22(8):931-41. Reilly MC, Zbrozek AS, Dukes EM (1993) The validity and reproducibility of a work productivity and activity impairment instrument. Pharmacoeconomics. 4(5):353-65. Reilly Associates. (2002) WPAI Scoring. http://www.reillyassociates.net/WPAI_Scoring.html Accessed August 5, 2021. Centers for Disease Control and Prevention. Long COVID or Post-COVID Conditions. https://www.cdc.gov/coronavirus/2019-ncov/long-term-effects/index.html Accessed August 15, 2022. Rosner B (2015) Fundamentals of biostatistics. Eighth ed. Cengage learning, Boston, MA Freeman G, Halton JH (1951) Note on an exact treatment of contingency, goodness of fit and other problems of significance. Biometrika. 38(1/2):141-9. Fitzmaurice GM, Laird NM, Ware JH (2012) Applied longitudinal analysis. John Wiley & Sons, Hoboken, NJ Cohen J (1988) Statistical power analysis for the behavioral sciences. 2nd ed. Lawrence Erlbaum Assoc, Hillsdale, NJ. McLeod LD, Cappelleri JC, Hays RD (2016) Best (but oft-forgotten) practices: expressing and interpreting associations and effect sizes in clinical outcome assessments. The American journal of clinical nutrition. 103(3):685-93. Erratum: https://doi.org/10.3945/ajcn.116.148593. Mouelhi Y, Jouve E, Castelli C, Gentile S (2020) How is the minimal clinically important difference established in health-related quality of life instruments? Review of anchors and methods. Health and Quality of Life Outcomes. 18(1):1-17. STROBE Statement—Checklist of items that should be included in reports of cohort studies. https://www.strobe-statement.org/download/strobe-checklist-cohort-studies-pdf Accessed August 5, 2022. Jacobson KB, Rao M, Bonilla H, Subramanian A, Hack I, Madrigal M, Singh U, Jagannathan P, Grant P (2021) Patients with uncomplicated COVID-19 have long-term persistent symptoms and functional impairment similar to patients with severe COVID19: a cautionary tale during a global pandemic. Clinical Infectious Diseases. 10.1093/cid/ciab103. Hernandez-Romieu AC, Leung S, Mbanya A, Jackson BR, Cope JR, Bushman D, Dixon M, Brown J, McLeod T, Saydah S (2021) Health care utilization and clinical characteristics of nonhospitalized adults in an integrated health care system 28–180 days after COVID-19 diagnosis—Georgia, May 2020–March 2021. Morbidity and Mortality Weekly Report. 70(17):644. Jiang R, Janssen M, Pickard AS (2021) US population norms for the EQ-5D-5L and comparison of norms from face-to-face and online samples. Quality of Life Research. 30(3):803-16. Lawson A, Tan AC, Naylor J, Harris IA (2020) Is retrospective assessment of healthrelated quality of life valid? BMC Musculoskeletal Disorders. 21(1):1-10. 22 medRxiv preprint doi: https://doi.org/10.1101/2022.08.31.22279264; this version posted September 6, 2022. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 28. 29. 30. 31. Rajan SS, Wang M, Singh N, Jacob AP, Parker SA, Czap AL, Bowry R, Grotta JC, Yamal JM (2021) Retrospectively Collected EQ-5D-5L Data as Valid Proxies for Imputing Missing Information in Longitudinal Studies. Value Health. 24(12):1720-7. 10.1016/j.jval.2021.07.007. Tundia N, Hass S, Fuldeore M, Wang LL, Cavanaugh T, Boone J, Heaton P (2015) Validation and US population norms of health-related productivity questionnaire. Value in Health. 18(3):A24. Romano C, Fehnel S, Stoddard J, Sadoff J, Lewis S, McNulty P, Chan EK, Evans E, Jamieson C, Slagle AF (2022) Development of a novel patient-reported outcome measure to assess signs and symptoms of COVID-19. Journal of Patient-Reported Outcomes. 6(1):1-12. Hughes SE, Haroon S, Subramanian A, McMullan C, Aiyegbusi OL, Turner GM, Jackson L, Davies EH, Frost C, McNamara G (2022) Development and validation of the symptom burden questionnaire for long covid (SBQ-LC): Rasch analysis. BMJ. 377. 23 medRxiv preprint doi: https://doi.org/10.1101/2022.08.31.22279264; this version posted September 6, 2022. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 537 Figure 1. Study design a 538 539 540 a QoL refers to the EQ-5D-5L survey 24 medRxiv preprint doi: https://doi.org/10.1101/2022.08.31.22279264; this version posted September 6, 2022. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 541 Figure 2. Study flow diagram 542 543 25 medRxiv preprint doi: https://doi.org/10.1101/2022.08.31.22279264; this version posted September 6, 2022. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 544 Table 1 Patient Characteristics on Index Day Total, n Age, years Mean, SD 18-29 30-49 50-64 ≥65 Gender Female Male Race / Ethnicity White or Caucasian (not Hispanic or Latino) Black or African American Hispanic Asian Patient Refused Other CMS Geographic Region (n, %) Region 1: ME, NH, VT, MA, CT, RI Region 2: NY, NJ, PR, VI Region 3: PA, DE, MD, DC, WV, VA Region 4: KY, TN, NC, SC, GA, MS, AL, FL Region 5: MN, WI, IL, MI, IN, OH Region 6: NM, OK, AR, TX, LA Region 7: NE, IA, KS, MO Region 8 : MT, ND, SD, WY, UT, CO Region 9: CA, NV, AZ, GU Region 10: AK, WA, OR, ID U.S. Geographic Region Northeast South Midwest West Previously Tested Positive All BNT162b2 Unvaccinated 430 233 197 42.4 (14.3) 43.7 (15.3) 40.9 (12.9) 87 (20.2%) 213 (49.5%) 94 (21.9%) 49 (21.0%) 100 (42.9%) 60 (25.8%) 38 (19.3%) 113 (57.4%) 34 (17.3%) 36 (8.4%) 24 (10.3%) P-value a 0.049 0.011 12 (6.1%) 327 (76.0%) 103 (24.0%) 177 (76.0%) 56 (24.0%) 150 (76.1%) 47 (23.9%) 295 (68.6%) 166 (71.2%) 129 (65.5%) 20 (4.7%) 61 (14.2%) 22 (5.1%) 13 (3.0%) 19 (4.4%) 7 (3.0%) 35 (15.0%) 15 (6.4%) 5 (2.2%) 5 (2.2%) 13 (6.6%) 26 (13.2%) 7 (3.6%) 8 (4.1%) 14 (7.1%) 19 (4.4%) 10 (4.3%) 9 (4.6%) 11 (2.6%) 7 (3.0%) 4 (2.0%) 37 (8.6%) 22 (9.4%) 15 (7.6%) 156 (36.3%) 81 (34.8%) 75 (38.1%) 58 (13.5%) 31 (13.3%) 27 (13.7%) 82 (19.1%) 19 (4.4%) 56 (24.0%) 11 (4.7%) 26 (13.2%) 8 (4.1%) 1 (0.2%) 1 (0.4%) 0 (0.0%) 46 (10.7%) 1 (0.2%) 13 (5.6%) 1 (0.4%) 33 (16.8%) 0 (0.0%) 53 (12.2%) 254 (58.7%) 77 (17.8%) 49 (11.3%) 167 (38.8%) 29 (12.3%) 150 (63.6%) 42 (17.8%) 15 (6.4%) 89 (38.2%) 24 (12.2%) 104 (52.8%) 35 (17.8%) 34 (17.3%) 78 (39.6%) 0.966 0.026 0.009 0.005 0.589 26 medRxiv preprint doi: https://doi.org/10.1101/2022.08.31.22279264; this version posted September 6, 2022. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . All 545 BNT162b2 Unvaccinated Work in healthcare 47 (10.9%) 29 (12.4%) 18 (9.1%) Work in high-risk setting 44 (10.2%) 30 (12.9%) 14 (7.1%) Live in high-risk setting 22 (5.1%) 12 (5.2%) 10 (5.1%) Social vulnerability index Mean (SD) 0.44 (0.22) 0.40 (0.22) 0.49 (0.21) Median (Q1, Q3) 0.41 (0.27, 0.59) 0.37 (0.23, 0.55) 0.47 (0.31, 0.63) Self-Reported Comorbidity Asthma or Chronic Lung 34 (7.9%) 21 (9.0%) 13 (6.6%) Disease Cirrhosis of the liver 1 (0.2%) 1 (0.4%) 0 (0.0%) Immunocompromised 19 (4.4%) 12 (5.2%) 7 (3.6%) Conditions or Weakened Immune System c Diabetes 20 (4.7%) 13 (5.6%) 7 (3.6%) Heart Conditions or 52 (12.1%) 30 (12.9%) 22 (11.2%) Hypertension Overweight or obesity 19 (4.4%) 12 (5.2%) 7 (3.6%) At least 1 comorbidity 104 (24.2%) 61 (26.2%) 43 (21.8%) Number of comorbidities, 0.34 (0.68) 0.38 (0.75) 0.28 (0.58) Mean (SD) Index day b acute COVID-19 symptoms Systemic symptoms Fever 93 (47.2%) 164 (38.1%) 71 (30.5%) Chills 113 (57.4%) 213 (49.5%) 100 (42.9%) Muscle or Body Aches 117 (59.4%) 232 (54.0%) 115 (49.4%) Headache 140 (71.1%) 293 (68.1%) 153 (65.7%) Fatigue 125 (63.5%) 266 (61.9%) 141 (60.5%) Respiratory symptoms 29 (14.7%) Shortness of Breath or 54 (12.6%) 25 (10.7%) Difficulty Breathing Cough 141 (71.6%) 309 (71.9%) 168 (72.1%) Sore Throat 104 (52.8%) 238 (55.3%) 134 (57.5%) 22 (11.2%) New/Recent Loss of Taste or 45 (10.5%) 23 (9.9%) Smell Congestion or Runny Nose 134 (68.0%) 322 (74.9%) 188 (80.7%) GI symptoms Nausea or Vomiting 31 (15.7%) 55 (12.8%) 24 (10.3%) Diarrhea 51 (25.9%) 88 (20.5%) 37 (15.9%) Number of acute COVID-19 5.3 (2.6) 5.1 (2.4) 5.6 (2.7) symptoms, Mean (SD) a P value refers to the comparison between BNT162B2 and Unvaccinated. P-value a 0.309 0.158 0.553 <0.001 <0.001 0.355 1.0000 0.422 0.320 0.588 0.422 0.294 0.138 <0.001 0.003 0.038 0.231 0.532 0.213 0.903 0.327 0.662 0.003 0.093 0.010 0.034 27 medRxiv preprint doi: https://doi.org/10.1101/2022.08.31.22279264; this version posted September 6, 2022. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 546 547 548 549 550 b COVID-19 test nasal swab day Immunocompromised conditions includes compromised immune system (such as from immuno-compromising drugs, solid organ or blood stem cell transplant, HIV, or other conditions), conditions that result in a weakened immune system, including cancer treatment, and kidney failure or end stage renal disease c 28 medRxiv preprint doi: https://doi.org/10.1101/2022.08.31.22279264; this version posted September 6, 2022. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 551 Table 2 Post-COVID-19 Symptoms at Week 4 Symptom 552 553 554 All BNT162b2 P-value a Unvaccinated General symptoms Tiredness or fatigue 71 (40.6%) 65 (41.9%) 136 (41.2%) Symptoms that get worse after physical or 50 (15.2%) 18 (10.3%) 32 (20.6%) mental activities Fever 1 (0.3%) 0 (0.0%) 1 (0.7%) General pain/discomfort 50 (15.2%) 20 (11.4%) 30 (19.4%) Respiratory and cardiac Difficulty breathing or shortness of breath 58 (17.6%) 26 (14.9%) 32 (20.6%) Cough 86 (26.1%) 40 (22.9%) 46 (29.7%) Chest or stomach pain 32 (9.7%) 14 (8.0%) 18 (11.6%) Fast-beating or pounding heart (also 38 (11.5%) 17 (9.7%) 21 (13.5%) known as heart palpitations) Neurologic Change in smell or taste 51 (15.5%) 19 (10.9%) 32 (20.6%) Headache 67 (20.3%) 28 (16.0%) 39 (25.2%) Dizziness on standing (lightheadedness) 45 (13.6%) 20 (11.4%) 25 (16.1%) Difficulty thinking or concentrating 86 (26.1%) 43 (24.6%) 43 (27.7%) (sometimes referred to as “brain fog”) Pins-and-needles feeling 24 (7.3%) 10 (5.7%) 14 (9.0%) Sleep problems 81 (24.5%) 35 (20.0%) 46 (29.7%) Mood changes 36 (10.9%) 13 (7.4%) 23 (14.8%) Memory loss 38 (11.5%) 11 (6.3%) 27 (17.4%) Other Diarrhea 23 (7.0%) 6 (3.4%) 17 (11.0%) Joint or muscle pain 67 (20.3%) 29 (16.6%) 38 (24.5%) Rash 11 (3.3%) 3 (1.7%) 8 (5.2%) Changes in period cycles 28 (11.5%) 12 (9.4%) 16 (13.8%) Number of post-COVID-19 symptoms, Mean 3.1 (3.6) 2.5 (3.0) 3.7 (4.1) (SD) 0 100 (30.3%) 54 (30.9%) 46 (29.7%) 1-2 100 (30.3%) 59 (33.7%) 41 (26.5%) 3-5 61 (18.5%) 40 (22.9%) 21 (13.5%) 6-8 32 (9.7%) 9 (5.1%) 23 (14.8%) ≥9 37 (11.2%) 13 (7.4%) 24 (15.5%) a P-values of t-test for number of symptoms, chi-square or Fisher’s exact tests when any one cell has an expected frequency less than 5 for individual symptoms and number of symptom category comparing the BNT162b2 cohort and the unvaccinated cohort. 29 0.802 0.009 0.470 0.045 0.168 0.159 0.268 0.276 0.014 0.039 0.214 0.513 0.247 0.042 0.031 0.002 0.007 0.073 0.082 0.280 0.002 0.001 555 556 Table 3 Summary of EQ-5D-5L and WPAI-GH Scoresa and Their Changes from Baseline by Assessment Time Score n Mean (SD) EQ-5D-5L Visual analogue scale (VAS) Baseline g 233 86.9 (10.7) Day 3 233 73.9 (15.6) Week 4 171 82.9 (13.8) Utility Index (US weight) 0.924 Baselinee 233 (0.117) 0.820 Day 3 233 (0.193) 0.882 Week 4 176 (0.145) WPAI-GH Absenteeism Baselinee 153 2.8 (11.8) Week 1 Week 4 Presenteeism Baselinee Week 1 Week 4 Work productivity loss Baselinee Week 1 BNT162b2 Cohort Change from Baseline b n Mean (SD) P-value c ESwd 233 -13.0 (12.5) <0.001 171 -3.7 (10.7) <0.001 -1.04 -0.35 Unvaccinated Cohort Difference in Change from Baseline Between Cohorts Score Change from Baseline n Mean (SD) n Mean (SD) P-value c ESwd Mean (SD) P-value e ESbf 197 87.8 (11.0) 194 71.8 (19.6) 194 -16.1 (17.1) <0.001 -0.94 3.1 (14.8) 151 80.9 (15.7) 151 -7.2 (13.7) <0.001 -0.53 3.5 (12.2) 197 233 176 -0.105 (0.159) -0.039 (0.120) <0.001 -0.66 197 <0.001 -0.33 155 0.918 (0.117) 0.762 (0.252) 0.838 (0.197) 0.414 h 0.033 0.21 0.011 0.28 0.547 h 197 155 -0.155 (0.228) -0.074 (0.156) <0.001 -0.68 <0.001 -0.47 129 11.7 (26.4) 0.050 (0.194) 0.035 (0.138) -13.0 (39.0) -0.27 9.0 (24.1) 0.007 0.26 0.023 0.25 <0.001 h 153 44.9 (38.3) 147 42.5 (38.8) <0.001 1.09 129 66.7 (37.0) 127 55.5 (39.2) <0.001 1.42 0.006 -0.33 128 4.6 (17.2) 122 0.460 0.07 106 3.3 (12.2) 0.006 0.37 1.13 0.06 124 9.4 (20.4) 82 47.8 (34.2) 105 18.7 (23.4) 0.958 h 81 38.2 (35.9) <0.001 1.06 -5.1 (32.1) 0.268 -0.16 96 10.7 (24.1) <0.001 0.45 -9.3 (24.3) 0.006 -0.38 1.35 123 14.2 (24.7) 82 66.9 (32.4) 0.175 h 80 53.2 (35.3) <0.001 1.51 -6.4 (34.9) 0.203 1.4 (20.1) 153 9.5 (18.5) 123 41.7 (27.4) 119 33.0 (29.1) 125 11.7 (18.8) 119 1.4 (24.5) 152 10.5 (20.0) 123 56.6 (31.5) 118 46.7 (34.5) <0.001 0.527 <0.001 102 -7.7 (28.1) 0.007 30 -0.18 557 558 559 560 561 562 563 564 565 566 Week 4 Activity impairment Baselinee Week 1 Week 4 a 125 12.9 (20.7) 118 1.4 (27.5) 203 12.5 (22.2) 202 48.6 (29.8) 202 36.1 (31.9) 175 15.5 (21.7) 175 2.3 (26.3) 0.589 <0.001 0.242 0.05 105 19.8 (24.6) 96 7.7 (30.6) 0.016 0.25 -6.3 (29.0) 0.113 1.13 0.09 176 17.2 (27.2) 0.065 h 176 54.2 (32.7) 176 37.0 (37.5) <0.001 0.99 -0.9 (34.6) 0.801 -0.03 155 25.9 (28.5) 155 8.8 (33.8) 0.002 0.26 -6.4 (30.1) 0.053 -0.21 Score ranges: EQ-5D-5L VAS 0 to 100, EQ-5D-5L UI (the United States weights) -0.573 to 1; WPAI-GH (absenteeism, presenteeism, work productivity loss, and activity impairment) 0 to 100 percent. b Baseline refers to pre-COVID-19 symptom onset. c P-value of t-test comparing mean score changes from baseline and 0 within BNT162b2 or Unvaccinated cohorts. d ESw refers to effect size for score changes from baseline within BNT162b2 or Unvaccinated cohorts. e P-value of t-test comparing mean score changes from baseline between BNT162b2 and Unvaccinated cohorts. f ESb refers to effect size for score changes from baseline between BNT162b2 and Unvaccinated cohorts. g Prior to symptom onset (pre-COVID). h P-values of t-tests comparing pre-COVID-19 baseline mean scores between BNT162b2 and Unvaccinated cohorts. 31 -0.22 567 Table 4 Least-Square Mean Estimate and 95% Confidence Interval for EQ-5D-5L and WPAI-GH Scores a Variable BNT162b2 Cohort Unvaccinated Cohort Score Change from Baseline Score Change from Baseline c LSE (95% CI) LSE (95% CI) PESw LSE (95% CI) LSE (95% CI) PESwc b b value value EQ-5D-5L Visual analogue scale (VAS) -11.1 (-14.1, <0.001 -0.89 8.2) Day 3 76.2 (73.3, 79.2) Week 4 85.0 (82.1, 87.9) -2.3 (-5.2, 0.6) Utility Index (US weight) 0.842 (0.805, Day 3 0.879) 0.903 (0.868, Week 4 0.938) WPAI-GH Absenteeism Week 1 45.6 (38.0, 53.1) Week 4 5.3 (0.0, 10.7) Presenteeism Week 1 38.4 (30.3, 46.5) Week 4 6.8 (0, 13.9) f Work productivity loss Week 1 53.8 (45.1, 62.4) Week 4 8.6 (0.9, 16.2) Activity impairment Week 1 43.9 (37.7, 50.1) 0.119 -0.22 72.6 (69.7, 75.4) 81.6 (78.8, 84.4) -0.077 (-0.115, 0.773 (0.736, <0.001 -0.49 -0.040) 0.809) -0.016 (-0.051, 0.859 (0.826, 0.369 -0.13 0.019) 0.892) 39.1 (31.6, 65.0 (57.5, <0.001 1.01 46.7) 72.5) -1.1 (-6.5, 4.2) 0.676 -0.06 2.2 (0.0, 7.1) f 50.2 (44.2, 56.2) P-value d ESbe <0.001 -0.86 3.6 (0.8, 6.5) 0.013 0.25 -5.7 (-8.5, -2.9) <0.001 -0.42 3.4 (0.6, 6.2) 0.016 0.28 -0.147 (-0.183, <0.001 -0.64 0.110) -0.060 (-0.093, <0.001 -0.38 0.027) 0.069 (0.032, 0.107) 0.044 (0.013, 0.075) <0.001 0.36 0.005 0.32 58.6 (51.1, 66.0) <0.001 1.49 -19.4 (-28.3, -10.6) <0.001 -0.50 -4.3 (-9.2, 0.6) 0.086 -0.15 42.4 (33.8, 65.0 (56.2, <0.001 1.23 53.6 (44.8, 62.4) <0.001 1.52 51.0) 73.7) -2.8 (-10.4, 4.9) 0.476 -0.10 17.0 (9.9, 24.0) 5.7 (-1.4, 12.7) 0.116 0.18 <0.001 0.91 LSE (95% CI) -14.8 (-17.6, 11.9) 29.0 (20.8, 46.8 (38.6, <0.001 1.00 37.4 (29.1, 45.6) <0.001 1.04 37.1) 55.1) -2.7 (-9.7, 4.4) 0.458 -0.11 16.0 (9.4, 22.5) 6.5 (0.0, 13.0) 0.050 0.27 29.0 (22.8, 35.3) Between Cohort Difference 35.4 (29.3, 41.4) <0.001 0.94 3.2 (-1.1, 7.4) 0.145 0.13 -8.4 (-16.7, -0.1) 0.047 -0.26 -9.2 (-14.7, -3.6) 0.001 -0.38 -11.2 (-19.9, -2.5) 0.012 -0.32 -8.4 (-14.5, -2.3) 0.007 -0.29 -6.3 (-12.4, -0.2) 0.044 -0.18 32 Week 4 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 11.0 (5.1, 16.9) -3.9 (-9.7, 2.0) 0.194 -0.15 21.3 (15.7, 26.9) 6.4 (0.8, 12.0) 0.025 0.19 -10.3 (-15.6, -5.0) <0.001 -0.34 Abbreviations: LSE = Least-Square Mean Estimate; CI = Confidence Interval. a Multivariate models include variables for time, vaccination status and interaction of time by vaccination status, as well as covariates of participant pre-COVID-19 symptom onset score, sociodemographic characteristics (age, sex, regions, social vulnerability, race/ethnicity, high risk occupations), previously tested positive for COVID-19, severity of acute illness (number of symptoms reported on index date), and immunocompromised status. Parameter estimates are presented in Supplemental Table 3. b P-value refers to the comparison of lease-square mean estimates score changes from baseline and 0 within BNT162b2 or Unvaccinated cohorts c ESw, within-cohort effect size, was calculated as the least square estimate of mean change from divided by the observed standard deviation of change scores from baseline to follow-up. d ESb, between-cohort effect size, was calculated as the difference in least square estimates of mean changes from baseline between cohorts, divided by the observed pooled standard deviation of change scores e P-value refers to the difference in lease-square mean estimates between BNT162b2 and Unvaccinated cohorts. f Lower limit of 95% CI was truncated from -2.7 to 0 for absenteeism and -0.3 to 0 for presenteeism at Week 4 because the valid range is 0 to 100. 33 599 Figure 3 Mean Responses of EQ-5D-5L Dimensions by Timepoint 600 601 602 603 604 605 606 607 Mean dimension scores range from 1 for no problem to 5 for extreme / unable. The blue and red solid lines indicate that vaccinated and unvaccinated were similar at the pre-COVID baseline. At Day 3 and Week 4 post-index date, vaccinated cohort was less impacted (lower scores) than unvaccinated by COVID on anxiety/depression, pain/discomfort, and usual activities (dotted lines for Day 3, dashed lines for Week 4). 34 medRxiv preprint doi: https://doi.org/10.1101/2022.08.31.22279264; this version posted September 6, 2022. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . Supplemental Table 1 Patient Characteristics by Enrollment Status Patient characteristics Vaccinated Yes No Missing Age, years Mean, SD Median, Q1, Q3 Age Category 18-29 30-49 50-64 ≥65 Gender, n (%) Male Female Race / Ethnicity, n (%) White or Caucasian (non-Hispanic or Latino) Black or African American Hispanic or Latino Asian Patient Refused Other US Geographic Region Northeast South Midwest West Number of acute COVID-19 symptoms, Mean (SD) Number of comorbidities, Mean (SD) ≥1 comorbidity Enrolled N=676 Not-Enrolled N=39,213 P value <0.001 465 (68.8%) 211 (31.2%) 22,125 (56.4%) 17,087 (43.6%) 43.2 (14.7) 41 (31--55) 42.1 (15.5) 40 (29--54) 0.076 0.010 134 (19.8%) 317 (46.9%) 156 (23.1%) 69 (10.2%) 10,095 (25.7%) 16,706 (42.6%) 8,561 (21.8%) 3,850 (9.8%) 0.005 <0.001 181 (26.8%) 495 (73.2%) 17,843 (45.5%) 21,370 (54.5%) <0.001 486 (71.9%) 32 (4.7%) 85 (12.6%) 35 (5.2%) 16 (2.4%) 22 (3.3%) 22,647 (57.8%) 4,162 (10.6%) 6,878 (17.5%) 2,551 (6.5%) 1,535 (3.9%) 1,439 (3.7%) 0.258 92 (13.6%) 402 (59.5%) 118 (17.5%) 64 (9.5%) 5.2 (2.5) 4,387 (11.2%) 24,305 (62.0%) 6,414 (16.4%) 4,105 (10.5%) 5.1 (2.5) 0.732 0.3 (0.7) 164 (24.3%) 0.3 (0.6) 7,677 (19.6%) 0.021 0.002 35 medRxiv preprint doi: https://doi.org/10.1101/2022.08.31.22279264; this version posted September 6, 2022. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . Supplemental Table 2 Summary of EQ-5D-5L Dimensions Dimension Pre-COVID-19 Baseline Mobility No problems Slight problems Moderate problems Severe problems Unable Missing Self-Care No problems Slight problems Moderate problems Severe problems Unable Missing Usual Activities No problems Slight problems Moderate problems Severe problems Unable Missing Pain / Discomfort No Slight Moderate Severe Extreme Missing Anxiety / Depression No Slightly Moderately Severely Extremely Missing Days 3 Mobility All BNT162b2 Unvaccinated P-value a 0.841 396 (92.1%) 30 (7.0%) 4 (0.9%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 213 (91.4%) 18 (7.7%) 2 (0.9%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 183 (92.9%) 12 (6.1%) 2 (1.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 419 (97.4%) 11 (2.6%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 226 (97.0%) 7 (3.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 193 (98.0%) 4 (2.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 394 (91.6%) 23 (5.4%) 12 (2.8%) 1 (0.2%) 0 (0.0%) 0 (0.0%) 214 (91.8%) 11 (4.7%) 7 (3.0%) 1 (0.4%) 0 (0.0%) 0 (0.0%) 180 (91.4%) 12 (6.1%) 5 (2.5%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 307 (71.4%) 93 (21.6%) 27 (6.3%) 3 (0.7%) 0 (0.0%) 0 (0.0%) 172 (73.8%) 46 (19.7%) 13 (5.6%) 2 (0.9%) 0 (0.0%) 0 (0.0%) 135 (68.5%) 47 (23.9%) 14 (7.1%) 1 (0.5%) 0 (0.0%) 0 (0.0%) 230 (53.5%) 140 (32.6%) 49 (11.4%) 10 (2.3%) 1 (0.2%) 0 (0.0%) 124 (53.2%) 80 (34.3%) 26 (11.2%) 3 (1.3%) 0 (0.0%) 0 (0.0%) 106 (53.8%) 60 (30.5%) 23 (11.7%) 7 (3.6%) 1 (0.5%) 0 (0.0%) 0.560 0.904 0.614 0.395 0.131 36 medRxiv preprint doi: https://doi.org/10.1101/2022.08.31.22279264; this version posted September 6, 2022. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . No problems Slight problems Moderate problems Severe problems Unable Missing Self-Care No problems Slight problems Moderate problems Severe problems Unable Missing Usual Activities No problems Slight problems Moderate problems Severe problems Unable Missing Pain / Discomfort No Slight Moderate Severe Extreme Missing Anxiety / Depression No Slightly Moderately Severely Extremely Missing Week 4 Mobility No problems Slight problems Moderate problems Severe problems Unable 343 (79.8%) 68 (15.8%) 17 (4.0%) 2 (0.5%) 0 (0.0%) 0 (0.0%) 193 (82.8%) 31 (13.3%) 7 (3.0%) 2 (0.9%) 0 (0.0%) 0 (0.0%) 150 (76.1%) 37 (18.8%) 10 (5.1%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 370 (86.0%) 47 (10.9%) 11 (2.6%) 2 (0.5%) 0 (0.0%) 0 (0.0%) 205 (88.0%) 21 (9.0%) 6 (2.6%) 1 (0.4%) 0 (0.0%) 0 (0.0%) 165 (83.8%) 26 (13.2%) 5 (2.5%) 1 (0.5%) 0 (0.0%) 0 (0.0%) 207 (48.1%) 132 (30.7%) 64 (14.9%) 21 (4.9%) 6 (1.4%) 0 (0.0%) 117 (50.2%) 75 (32.2%) 30 (12.9%) 9 (3.9%) 2 (0.9%) 0 (0.0%) 90 (45.7%) 57 (28.9%) 34 (17.3%) 12 (6.1%) 4 (2.0%) 0 (0.0%) 145 (33.7%) 183 (42.6%) 82 (19.1%) 17 (4.0%) 3 (0.7%) 0 (0.0%) 85 (36.5%) 99 (42.5%) 43 (18.5%) 4 (1.7%) 2 (0.9%) 0 (0.0%) 60 (30.5%) 84 (42.6%) 39 (19.8%) 13 (6.6%) 1 (0.5%) 0 (0.0%) 190 (44.2%) 135 (31.4%) 75 (17.4%) 26 (6.1%) 4 (0.9%) 0 (0.0%) 108 (46.4%) 79 (33.9%) 37 (15.9%) 8 (3.4%) 1 (0.4%) 0 (0.0%) 82 (41.6%) 56 (28.4%) 38 (19.3%) 18 (9.1%) 3 (1.5%) 0 (0.0%) 0.571 0.361 0.080 0.047 0.278 270 (81.3%) 49 (14.8%) 12 (3.6%) 1 (0.3%) 0 (0.0%) 149 (84.7%) 22 (12.5%) 5 (2.8%) 0 (0.0%) 0 (0.0%) 121 (77.6%) 27 (17.3%) 7 (4.5%) 1 (0.6%) 0 (0.0%) 37 medRxiv preprint doi: https://doi.org/10.1101/2022.08.31.22279264; this version posted September 6, 2022. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . Missing Self-Care No problems Slight problems Moderate problems Severe problems Unable Missing Usual Activities No problems Slight problems Moderate problems Severe problems Unable Missing Pain / Discomfort No Slight Moderate Severe Extreme Missing Anxiety / Depression No Slightly Moderately Severely Extremely Missing 1 (0.0%) 1 (0.0%) 0 (0.0%) 308 (93.1%) 17 (5.1%) 5 (1.5%) 1 (0.3%) 0 (0.0%) 2 (0.0%) 167 (94.9%) 5 (2.8%) 4 (2.3%) 0 (0.0%) 0 (0.0%) 1 (0.0%) 141 (91.0%) 12 (7.7%) 1 (0.7%) 1 (0.7%) 0 (0.0%) 1 (0.0%) 232 (70.1%) 75 (22.7%) 20 (6.0%) 4 (1.2%) 0 (0.0%) 2 (0.0%) 132 (75.0%) 35 (19.9%) 8 (4.6%) 1 (0.6%) 0 (0.0%) 1 (0.0%) 100 (64.5%) 40 (25.8%) 12 (7.7%) 3 (1.9%) 0 (0.0%) 1 (0.0%) 170 (51.4%) 118 (35.6%) 37 (11.2%) 5 (1.5%) 1 (0.3%) 2 (0.0%) 95 (54.0%) 66 (37.5%) 13 (7.4%) 2 (1.1%) 0 (0.0%) 1 (0.0%) 75 (48.4%) 52 (33.5%) 24 (15.5%) 3 (1.9%) 1 (0.7%) 1 (0.0%) 167 (50.5%) 98 (29.6%) 51 (15.4%) 10 (3.0%) 5 (1.5%) 2 (0.0%) 92 (52.3%) 53 (30.1%) 25 (14.2%) 5 (2.8%) 1 (0.6%) 1 (0.0%) 75 (48.4%) 45 (29.0%) 26 (16.8%) 5 (3.2%) 4 (2.6%) 1 (0.0%) 0.068 0.152 0.097 0.605 a P values of Freeman-Halton tests comparing BNT162b2 cohort and unvaccinated cohort, excluding category of missing. 38 Supplemental Table 3 Mixed Models for Repeated Measurements EQ-5D-5L and WPAI-GH Scores Work productivity EQ-5D-5L Utility Index Absenteeism Presenteeism Activity impairment (U.S. weights) loss Coeff (SE) P value Coeff (SE) P value Coeff (SE) P value Coeff (SE) P value Coeff (SE) P value Coeff (SE) P value EQ VAS Intercept 4.51 (5.36) Vaccinated BNT162B2BNT162b2 3.65 (1.46) No Reference Assessment Time Day 2-4 / Week 1 Reference Week 4 9.04 (1.23) Assessment Time * Vaccinated Day 2-4 / Week 1 * Reference BNT162b2 Week 4 * BNT162B2 -0.25 (1.69) Pre-COVID-19 Baseline -0.22 (0.05) Score Age, years 18-29 Reference 30-49 -1.82 (1.45) 50-64 -2.84 (1.74) ≥65 -1.86 (2.29) Gender Male 4.07 (1.30) Female Reference Race / Ethnicity White (non-Hispanic) Reference Black/African 2.17 (2.73) American Hispanic or Latino 0.65 (1.68) 0.401 0.038 (0.062) 0.546 53.6 (4.9) <0.001 41.4 (5.7) <0.001 57.7 (6.0) <0.001 0.013 0.069 (0.019) <0.001 -19.4 (4.5) <0.001 -8.4 (4.2) Reference Reference Reference 0.047 -11.2 (4.4) 0.012 Reference 41.2 (4.9) <0.001 -6.3 (3.1) Reference 0.044 Reference Reference Reference Reference Reference <0.001 0.086 (0.014) <0.001 -62.9 (3.4) <0.001 -30.9 (3.3) <0.001 -48.0 (3.6) <0.001 -28.9 (2.5) Reference Reference Reference Reference 22.6 (4.8) <0.001 -0.8 (4.4) 0.561 -4.0 (3.5) 0.257 <0.001 -0.177 (0.057) 0.002 -0.8 (0.0) <0.001 -0.7 (0.1) <0.001 -0.8 (0.1) <0.001 -0.7 (0.0) <0.001 Reference 0.211 -0.011 (0.018) 0.532 0.102 -0.048 (0.021) 0.025 0.417 -0.025 (0.028) 0.367 Reference 3.9 (2.4) 0.102 10.8 (2.9) <0.001 1.7 (5.5) 0.760 Reference 0.065 4.1 (3.2) 0.204 0.074 7.3 (4.1) 0.075 0.964 -1.4 (7.9) 0.863 Reference 7.8 (2.9) 13.7 (3.5) 6.9 (4.7) 0.006 <0.001 0.139 0.002 -2.1 (2.2) 0.360 -6.8 (2.9) Reference Reference 0.019 -5.8 (3.1) 0.057 Reference -9.8 (2.6) Reference Reference Reference 0.028 (0.016) Reference 0.080 Reference Reference Reference 2.8 (4.8) Reference 0.884 -0.025 (0.020) 0.209 Reference 5.6 (3.0) 6.8 (3.8) 0.3 (7.4) 0.863 <0.001 <0.001 0.428 0.017 (0.033) 0.613 -1.8 (5.1) 0.726 0.8 (6.4) 0.899 0.1 (7.1) 0.994 -1.9 (5.6) 0.730 0.701 0.005 (0.021) 0.818 -1.9 (3.0) 0.530 4.3 (3.9) 0.268 4.8 (4.1) 0.243 -2.3 (3.4) 0.494 39 Asian 4.24 (2.54) Patient refused -0.30 (3.26) Other 2.90 (2.81) Region Northeast 1.91 (2.00) South 0.87 (1.51) Midwest Reference West 0.49 (2.08) Social Vulnerability Index <0.25 Reference ≥0.25 and <0.5 1.21 (1.45) ≥0.5 and <0.75 0.93 (1.62) ≥0.75 -2.80 (2.14) Previously tested 0.99 (1.13) positive Work in healthcare -0.90 (1.80) Work in high-risk -2.83 (1.96) setting Live in high-risk setting 1.36 (2.53) Immunocompromised 1.22 (2.61) Number of Symptoms -0.59 (0.23) on index day 0.096 0.927 0.303 0.032 (0.031) 0.031 (0.040) 0.019 (0.035) 0.316 0.438 0.582 -1.7 (4.7) -3.3 (6.1) 3.9 (4.8) 0.713 0.589 0.418 1.4 (5.8) 1.1 (8.4) -6.9 (6.5) 0.807 0.897 0.288 2.4 (6.3) 0.705 -2.3 (9.0) 0.800 -3.1 (6.9) 0.661 0.9 (5.2) -6.4 (7.0) -7.5 (5.8) 0.858 0.359 0.198 0.341 0.015 (0.025) 0.534 0.568 -0.002 (0.019) 0.905 Reference 0.815 0.019 (0.026) 0.448 6.2 (3.3) 0.057 -1.9 (4.2) -1.0 (2.6) 0.698 -2.4 (3.3) Reference Reference -0.7 (3.4) 0.830 -0.8 (4.5) 0.648 0.466 -0.5 (4.5) 0.908 -2.4 (3.5) 0.507 Reference 0.857 -1.3 (4.8) 0.793 -2.3 (3.9) -2.9 (3.0) Reference -3.7 (4.2) 0.551 0.328 0.403 0.568 0.192 Reference 0.034 (0.018) 0.020 (0.020) 0.009 (0.026) 0.060 0.321 0.740 Reference Reference 1.3 (2.4) 0.579 -3.9 (3.0) 4.7 (2.8) 0.090 -4.8 (3.6) 2.3 (3.7) 0.531 -9.3 (4.7) Reference 0.196 -4.6 (3.3) 0.158 0.180 -6.0 (3.9) 0.124 0.048 -7.8 (5.0) 0.121 Reference -5.2 (2.9) -3.6 (3.3) -5.0 (4.3) 0.076 0.276 0.247 0.381 0.003 (0.014) 0.816 -1.9 (2.0) 0.333 -0.5 (2.6) 0.851 -2.7 (2.8) 0.329 -1.2 (2.3) 0.591 0.616 -0.018 (0.022) 0.424 3.1 (2.8) 0.272 0.8 (3.6) 0.818 -0.7 (4.0) 0.851 3.7 (3.6) 0.300 0.150 -0.006 (0.024) 0.790 4.7 (3.4) 0.166 10.3 (4.3) 0.017 13.3 (4.6) 0.004 5.7 (4.0) 0.155 0.591 -0.028 (0.031) 0.370 0.639 -0.018 (0.032) 0.560 3.2 (4.7) 0.502 13.1 (6.1) 0.033 4.8 (6.0) 9.9 (8.4) 0.431 0.243 7.2 (6.5) 7.9 (9.1) 0.268 0.385 12.1 (5.0) 5.5 (5.1) 0.017 0.282 0.009 -0.010 (0.003) 0.001 0.8 (0.4) 1.3 (0.5) 0.008 1.9 (0.5) 0.001 1.8 (0.4) <0.001 0.048 0.374 40 Supplemental Figure 1 Least-Square Estimates and 95% Confidence Intervals of EQ-5D-5L Scores a a Score ranges: EQ-5D-5L VAS 0 to 100, EQ-5D-5L UI (the United States weights) -0.573 to 1. 41 Supplemental Figure 2 Summary Results of EQ-5D-5L scores across time periods a Score ranges: EQ-5D-5L VAS 0 to 100, EQ-5D-5L UI (the United States weights) -0.573 to 1. Dots are the mean values and whiskers are the 95% Confidence Intervals. The Pre-COVID values represent the pooled means with 95% Confidence Internals. The Day 3 and Week 4 values are the least square estimate scores. a 42 Supplemental Figure 3 Least-Square Estimates and 95% Confidence Intervals of WPAI-GH Scores a a WPAI-GH score (absenteeism, presenteeism, work productivity loss, and activity impairment) ranges from 0 to 100 percent. 43 Supplemental Figure 4 Summary Results of WPAI-GH scores across time periods a a WPAI-GH score (absenteeism, presenteeism, work productivity loss, and activity impairment) ranges from 0 to 100 percent. Dots are the mean values and whiskers are the 95% Confidence Intervals. The Pre-COVID values represent the pooled means with 95% Confidence Internals. The Week 1 and Week 4 values are the least square estimate scores. 44 45 Supplemental Table 4 Model Predicting the Missingness at Week 1 and Week 4 Intercept Assessment Time Week 4 Week 1 Vaccination status BNT162b2 Unvaccinated EQ VAS on Day 3 EQ VAS change from preCOVID-19 baseline to Day 3 Age, years 18-29 30-49 50-64 ≥65 Gender Male Female Race / Ethnicity White (non-Hispanic) Black/African American Hispanic or Latino Asian Patient refused Summary: % (n) / Mean (SD) Total Week 1 Missing Week 4 Missing 430 11.9% (51) 23.0% (99) Coeff (SE) -3.14 (1.19) Model P value 0.009 Odds Ratio 23.0% (99) 0.85 (0.12) Reference <0.001 . 2.3 (1.9, 2.9) 1.0 12.9% (30) 10.7% (21) 24.5% (57) 21.3% (42) 68.9 (19.1) -18.2 (18.9) 71.7 (17.3) -16.0 (16.6) 0.36 (0.26) Reference 0.00 (0.01) 0.166 . 0.967 1.4 (0.9, 2.4) 1.0 1.0 (1.0, 1.0) -0.01 (0.01) 0.311 1.0 (1.0, 1.0) 87 213 94 36 8.0% (7) 10.3% (22) 16.0% (15) 19.4% (7) 16.1% (14) 23.9% (51) 27.7% (26) 22.2% (8) Reference 0.69 (0.36) 0.94 (0.40) 0.72 (0.52) . 0.052 0.019 0.161 1.0 2.0 (1.0, 4.0) 2.6 (1.2, 5.6) 2.1 (0.8, 5.6) 103 327 4.9% (5) 14.1% (46) 16.5% (17) 25.1% (82) -0.80 (0.32) Reference 0.013 . 0.4 (0.2, 0.8) 1.0 295 20 61 22 13 9.8% (29) 20.0% (4) 11.5% (7) 18.2% (4) 30.8% (4) 19.7% (58) 35.0% (7) 27.9% (17) 27.3% (6) 30.8% (4) Reference 0.67 (0.50) 0.21 (0.36) 0.37 (0.52) 0.94 (0.62) . 0.184 0.565 0.480 0.128 1.0 2.0 (0.7, 5.3) 1.2 (0.6, 2.5) 1.4 (0.5, 4.0) 2.6 (0.8, 8.6) 430 430 233 197 72.9 (17.5) -14.4 (14.8) 11.9% (51) 46 Other Region Northeast South Midwest West Social Vulnerability Index <0.25 ≥0.25 and <0.5 ≥0.5 and <0.75 ≥0.75 Previously tested positive Work in healthcare Work in high-risk setting Live in high-risk setting Immunocompromised Number of Symptoms on index day 19 15.8% (3) 36.8% (7) 1.22 (0.54) 0.024 3.4 (1.2, 9.8) 52 252 77 49 7.7% (4) 13.5% (34) 6.5% (5) 16.3% (8) 21.2% (11) 24.6% (62) 16.9% (13) 26.5% (13) 0.50 (0.48) 0.32 (0.37) Reference 0.44 (0.47) 0.295 0.378 . 0.343 1.7 (0.6, 4.2) 1.4 (0.7, 2.8) 1.0 1.6 (0.6, 3.9) 98 164 120 48 167 47 44 22 19 12.2% (12) 9.8% (16) 12.5% (15) 16.7% (8) 15.6% (26) 12.8% (6) 13.6% (6) 9.1% (2) 10.5% (2) 22.4% (22) 19.5% (32) 25.0% (30) 31.3% (15) 26.9% (45) 19.1% (9) 22.7% (10) 27.3% (6) 15.8% (3) Reference . 1.0 -0.22 (0.33) 0.07 (0.35) 0.38 (0.45) 0.45 (0.25) -0.24 (0.42) -0.25 (0.43) 0.38 (0.54) -0.33 (0.67) 0.496 0.850 0.397 0.069 0.567 0.559 0.484 0.623 0.8 (0.4, 1.5) 1.1 (0.5, 2.1) 1.5 (0.6, 3.5) 1.6 (1.0, 2.5) 0.8 (0.3, 1.8) 0.8 (0.3, 1.8) 1.5 (0.5, 4.3) 0.7 (0.2, 2.7) 5.3 (2.6) 5.4 (2.6) 5.0 (2.5) -0.07 (0.05) 0.188 0.9 (0.8, 1.0) 47