key: cord-0291818-8niu5hyi
authors: Alali, W. Q.; Ali, L. A.; AlSeaidan, M.; Al-Rashidi, M.
title: Effectiveness of BNT162b2 and ChAdOx1 vaccines against symptomatic COVID-19 among Healthcare Workers in Kuwait: A retrospective cohort study
date: 2021-07-29
journal: nan
DOI: 10.1101/2021.07.25.21261083
sha: abfaa0558dd63da4865ed8a70e0f99aea7690d3e
doc_id: 291818
cord_uid: 8niu5hyi

Background: The COVID-19 BNT162b2 vaccination rollout in Kuwait started on 24 December 2020 followed by ChAdOx1 on 3 February 2021. The study objectives were to assess the factors associated with vaccine coverage and determine vaccine effectiveness (VE) against SARSCoV2 infection in a healthcare worker (HCW) population. Methods: This retrospective cohort study was conducted among HCW working at a public secondary hospital in Kuwait. The follow-up period was from 24 December 2020 to 15 June 2021. The primary outcomes were vaccine coverage and PCR-confirmed SARS-CoV-2 infection for the VE analysis. Data on new SARS-CoV-2 infections (with or without symptoms) during study period in addition to HCWs characteristics (sex, age, nationality, and occupation) were extracted from the hospital records. The vaccine coverage and PCR confirmed SARS-CoV-2 infections were cross-tabulated by the HCWs characteristics. Furthermore, we used Cox regression to estimate time to infection hazard ratios in vaccinated (first and second dose) compared to unvaccinated HCWs. Only one ChAdOx1 dose was given during the study period. Results: There were 3246 HCWs included in the analysis. The median age was 38 years (IQR = 33 - 44), 63.4% were females, 46.8% aged 31 - 40, and 82.3% were non-Kuwaitis. Overall, 82.1% of HCWs received at least one vaccine dose (50.4% received only one dose of ChAdOx1, 3.3% received one dose of BNT162b2, and 28.3% received two doses of BNT162b2). 17.9% of HCWs remained unvaccinated by the end of the study. A significantly lower vaccination coverage percentage was amongst female HCWs, younger age group (20 - 30 years old), and administrative/executive staff. Symptomatic SARS-CoV-2 PCR confirmed infection prevalence was 7.3%. No asymptomatic infections were reported. The SARS-CoV-2 infection incidence rate was 126 per 100,000 person-days in the unvaccinated group; the incidence rates in partially vaccinated groups ([≥] 28 days after ChAdOx1 first dose) and ([≥] 14 days after receiving BNT162b2 through receipt of second dose) were 31.4 and 10.9 per 100,000 person-days, respectively. In the fully vaccinated group ([≥] 14 days after BNT162b2 second dose), the incidence rate was 6.3 per 100,000 person-days. The estimated adjusted vaccine effectiveness of fully vaccinated was 94.5% (95% confidence interval [CI] = 89.4%-97.2%). The VE of partially vaccinated for ChAdOx1 and BNT162b2 was 75.4% (95% CI = 67.2%-81.6%) and 91.4% (95% CI = 65.1% - 97.9%), respectively. Conclusions: Both BNT162b2 and ChAdOx1 vaccines prevented most symptomatic infections in this population across age groups, nationalities, and occupations. A significant proportion (17.9%) of HCWs were unvaccinated despite the vaccine accessibility. The findings complement other VE studies and demonstrate the vaccine benefit for HCWs.

The two anti-COVID-19 vaccines (BNT162b2 mRNA [Pfizer-BioNTech] and ChAdOx1 confirmed infection prior to the study's starting date and additional 403 HCWs who had missing 120 or incomplete vaccination data or symptom data were also excluded. Hence, a total of 3246 121

HCWs eligible for this study. 122

The primary outcome for vaccine effectiveness analysis was the SARS-CoV-2 PCR-124 confirmed infection among unvaccinated or vaccinated at any time during the study (i.e., during 125 the follow-up time) irrespective of symptom status. Infections were described as symptomatic if 126 their symptom status was seven days before or seven days of their PCR positive test date. 127

The primary outcome for the vaccine coverage analysis was the vaccination status (first 128 

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted July 29, 2021. ; https://doi.org/10.1101/2021.07.25.21261083 doi: medRxiv preprint Data were stratified by sociodemographic and occupation factors (i.e., covariates). These 142 were: age group (20 -30, 31-40, 41 -50, and >50), sex (male or female), nationality (Kuwaiti 143 national and non-Kuwaiti resident), staff group (Administrative or Executive; Nursing or Health-144 care assistant; Doctor; Specialist Staff; Estates, Porters, or Security; and Pharmacist) and 145 occupation settings (categorized into six groups: 1) office or laboratory, 2) hospital pharmacy, 3) 146 outpatient including radiology, day ward, general practice, or renal dialysis unit, 4) inpatient 147 ward, theatres, emergency department, maternity unit or labor ward, or ambulance, 5) intensive 148 care, and 6) other (e.g., plaster and observational rooms). 149

For vaccine coverage analysis, we cross-tabulated three vaccination statuses as 150 unvaccinated; vaccinated with one ChAdOx1 dose; and vaccinated with one or two doses of 151

BNT162b2 with the study covariates. The relationship between vaccine coverage status and 152 covariates were assessed via chi-square statistic using STATA software version 16.1 (College 153 Station, Texas, USA). Furthermore, we also cross-tabulated SARS-CoV-2 PCR-confirmed 154 infection by vaccination status and each level of the covariates. Similarly, chi-square statistic 155 was used to assess relationships. 156

We used retrospective cohort study design to estimate the vaccine effectiveness in HCWs 157 population after the first and second dose. For the purpose of vaccine effectiveness analysis, the 158 HCWs were defined as unvaccinated (if they had not received any doses of either vaccine), fully 159 vaccinated (if at least 14 days had passed since receiving the second dose of BNT162b2), and 160 partial vaccination (if at least 28 days passed since receiving ChAdOx1 first dose or at least 14 161 days after receiving BNT162b2 first dose but before receiving second dose). The BNT162b2 13 162 person-days between receiving vaccine first dose and partial or full vaccination were considered 163 excluded from the analysis as at-risk person-time because immunity was considered 164 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted July 29, 2021. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted July 29, 2021. ; https://doi.org/10.1101/2021.07.25.21261083 doi: medRxiv preprint vaccine and 28.3% received two doses of BNT162b2 vaccine. Those who received only one dose 188 of BNT162b2 (3.3%) did not received their second dose because they were SARS-CoV-2 189 infected after the first dose (only two HCWs) or the study ended before they received it. 190

The percentage of HCWs classified as partially vaccinated (i.e., ≥ 28 days after receiving 191 one dose of ChAdOx1 or ≥ 14 days after receiving BNT162b2 first dose through receipt of 192 second dose) was 50.2% and 2.8%, respectively. However, the percentage of HCWs classified 193 as fully vaccinated (≥ 14 days after BNT162b2 second dose) was 28.2%. 194

The characteristics of unvaccinated and vaccinated HCWs by the two types of vaccine 195 are shown in Table 1 . Twenty percent of females were unvaccinated compared to 10.6% of 196 males (P <0.001). For age groups, 28.2% of HCWs aged 20-30 were unvaccinated, significantly 197 higher than other age groups (P <0.001); whereas, within those received one or two doses of 198 BNT162b2 vaccine, the percentage of vaccinated HCWs in age groups (20 -30 and > 50) was 199 higher than that in other age groups (P <0.001). Interestingly, the percentages of unvaccinated 200

Kuwaitis (20.7%) and non-Kuwaitis (17.3%) HCWs were not significantly different (P = 0.054); 201 however, within those received one or two doses of BNT162b2, 60.7% of Kuwaiti HCWs were 202 vaccinated compared to 25.8% for non-Kuwaitis (P <0.001). Among the different occupation 203 settings, 23.3% of HCWs who worked in office or laboratories were unvaccinated, significantly 204 higher compared to the remaining settings (P <0.001) except for 'other'. As for the HCW staff 205 groups, 31.5% of administrative or executive staff were unvaccinated, significantly higher than 206 other groups (P <0.001). In addition, 58.4% of doctors received one or two doses of BNT162b2 207 vaccine significantly higher than other staff groups (P <0.001). 208 209 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted July 29, 2021. ; https://doi.org/10.1101/2021.07.25.21261083 doi: medRxiv preprint SARS-CoV-2 PCR-confirmed infection prevalence with reported symptoms was 7.3% 210 (237/3246) during the study period. There were two additional SARS-CoV-2 PCR-confirmed 211 HCW infections with missing symptomatic status; hence, they were excluded from the analysis. 212 Therefore, all the 237 SARS-CoV-2 PCR-confirmed infection were classified as symptomatic. 213

As shown in Table 2 , the infection prevalence was significantly higher among unvaccinated 214 female HCWs (20.8%) compared to 7.1% in those vaccinated with ChAdOx1 and 1.96% in those 215 vaccinated with one or two doses of BNT162b2. Similar findings were observed among male 216

HCWs. SARS-CoV-2 PCR-confirmed prevalence in the different age-groups and by nationality 217 were significantly higher in unvaccinated compared to the vaccinated groups. Furthermore, the 218 infection prevalence was significantly higher across the unvaccinated occupation settings 219 compared to those in vaccinated occupation settings except for "other" where the differences 220

were not significant (P = 0.508). The infection prevalence in staff groups were also significantly 221 higher in unvaccinated compared to vaccinated except for pharmacists (P = 0.866) ( Table 2) . 222 223 There were 114 SARS-CoV-2 infection during the 90,484 person-days of follow-up in 224 the unvaccinated group, an incidence rate of 126 per 100,000 person-days (Table 3 ). In the 225 partially vaccinated group, ≥ 28 days after ChAdOx1 first dose, there were 87 infections 226 (incidence rate of 31.4 per 100,000 person-days). Moreover, in the partially vaccinated group (≥ 227 14 days after receiving BNT162b2 vaccine through receipt of second dose), there were two 228 infections (incidence rate of 10.9 per 100,000 person-days). In the fully vaccinated group (≥ 14 229 days after BNT162b2 second dose) there were 10 infections (incidence rate of 6.3 per 100,000 230 person-days). 231 232 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted July 29, 2021. This retrospective cohort study was conducted between 24 December 2020 and 15 June 246 2021 (i.e., 173 days) at a secondary hospital in Kuwait shows that full vaccination (i.e., 247 immunization) via BNT162b2 is highly effective against symptomatic COVID-19 among this 248 HCW population. Furthermore, ChAdOx1 one dose was relatively effective (Table 3) . 249

Vaccine coverage with at least one dose among HCWs after 173 days (about 5.8 months) 250 was 82.1% of HCWs including 28.3% who received two doses. However, still there were 17.9% 251 of HCWs unvaccinated by the end of the study which is a concern. Healthcare workers have 252 been given the priority for vaccination in Kuwait as most countries; therefore, efforts are needed 253 to better understand reasons for vaccine hesitancy in this high-risk exposure group. Other studies 254 have reported that most HCWs were vaccinated with at least one dose within two to three months 255 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted July 29, 2021. ; https://doi.org/10.1101/2021.07.25.21261083 doi: medRxiv preprint in Kuwait in late June 2021 (after the study ended); hence, it did not confound the effectiveness 301

This study has several limitations. First, the study was based on one public secondary 303 hospital and might not be generalizable to HCWs in other public hospitals in Kuwait. However, 304 this hospital is one of the major healthcare facilities in Kuwait and serves over a quarter of the 305 country's population. Second, the identification of HCWs SARS-CoV-2 PCR-confirmed 306 infections was based on passive reporting to the hospital management due to lack of active 307 laboratory surveillance. However, it was/is required by all HCWs to report PCR-confirmed 308 infections to their upper management within each hospital's department. shown in other studies, it is important for HCWs to continue to exercise physical distancing, 322

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted July 29, 2021. ; https://doi.org/10.1101/2021.07.25.21261083 doi: medRxiv preprint 

Safety and Efficacy of the BNT162b2 mRNA Covid-19 Vaccine

Interim Estimates of Vaccine Effectiveness of BNT162b2 and mRNA-1273 COVID-19

Vaccines in Preventing SARS-CoV-2 Infection Among Health Care Personnel

Effectiveness of First Dose of COVID-19 Vaccines Against Hospital Admissions in

National Prospective Cohort Study of 5.4 Million People

Safety and efficacy of the ChAdOx1 nCoV-19 vaccine (AZD1222)

SARS-CoV-2: an interim analysis of four randomised controlled trials in Brazil

South Africa, and the UK. The Lancet

P-values (comparing the column percentages of vaccinate status by sociodemographic, occupation setting, and staff group categories were calculated using

Pearson's chi-square test (cells with ≥5 observations) or Fisher's exact test (cells with <5 observations) in STATA statistical software

Hospital pharmacy, 3: outpatient including radiology, day ward, general practice, or renal dialysis 397 unit; 4: inpatient ward, theatres, emergency department, maternity unit or labor ward, or ambulance; 5: intensive care