Table 4.

Details of the study outcome measures related to COVID-19 VE

Study	Outcome measures	Controls	VE follow-up duration	VE calculation	Analysis method
Young-Xu Y, et al. [17]	VE: SARS-CoV-2 infection (positive RT-PCR or antigen test) VE: COVID-19-related hospitalization (admission/discharge diagnosis of COVID-19); not assessed in IC VE: COVID-19-related death; not assessed in IC	For SARS-CoV-2 infection, controls were patients with a negative SARS-CoV-2 RT-PCR test For endpoints 2 & 3, controls were patients with a positive SARS-CoV-2 RT-PCR test who were not hospitalized for COVID-19 or died during the study period	≥14 days after 2nd dose up to 3 months	(1− OR) X 100 OR: SARS-CoV-2 infection in vaccinated vs. unvaccinated	Logistic regression
Tenforde MW, et al. [18]	VE: Odds of prior SARS-CoV-2 vaccination in hospitalized cases with COVID-19 (symptoms and a positive RT-PCR/clinical test) vs. controls	Test-negative controls: Hospitalized patients who tested negative for SARS-CoV-2, but with acute respiratory illness (ARI) symptoms ARI syndrome-negative controls: Hospitalized patients who tested negative for SARS-CoV-2 and without ARI signs and symptoms	≥14 days before reference date (date of hospitalization for cases)	(1‒OR) X 100 OR: Prior vaccination in cases vs. controls	Logistic regression
Tenforde MW, et al. [19]	VE: Odds of prior SARS-CoV-2 vaccination in hospitalized cases with COVID-19 (symptoms and a positive RT-PCR/clinical test) vs. controls	Test-negative controls: Hospitalized patients who tested negative for SARS-CoV-2, but had COVID-19-like illness COVID-19-like illness-negative controls: Hospitalized patients who tested negative for SARS-CoV-2 and did not have COVID-19-like illness	≥14 days before reference date (date of hospitalization for cases) Full surveillance period and separately for March–May and June–July 2021, because of increased circulation of Delta variant in the US	(1‒OR) X 100 OR: Prior vaccination in cases vs. controls	Logistic regression
Khan N & Mahmud N. [20]	VE: SARS-CoV-2 infection (positive RT-PCR test) VE: Severe COVID-19 (hospitalization or death) VE: All-cause mortality (extracted from vital status file in VHA dataset)	Unvaccinated controls	≥7 days after 2nd dose Median fully vaccinated: 38 days Median unvaccinated: 123 days	(1 – incidence vaccinated/incidence unvaccinated) X 100	Inverse probability weighted (IPW) adjusted models Cox proportional hazards regression
Polinski JM, et al. [21]	VE: SARS-CoV-2 infection (medical claim with COVID-19 diagnosis code and/or a positive RT-PCR test) VE: COVID-19-related hospitalization (discharge diagnosis of COVID-19 or a recorded infection within 21 days before admission)	Unvaccinated controls	≥14 days after single dose up to 5 months	(1-HR) X 100 HR: COVID-19 infection/hospitalization in vaccinated vs. unvaccinated	Propensity score matching Incidence rates calculated per 1,000 person-years Cox regression models used to calculate hazard ratios VE estimates were corrected for under-recording in data source
Dagan N, et al.; Barda N, et al. [22,23]	VE: SARS-CoV-2 infection (positive RT-PCR test) VE: Symptomatic COVID-19 (documented in EMR) VE: Severe COVID-19 illness (according to NIH criteria [37]) VE: COVID-19-related hospitalization (admission records) VE: COVID-19-related death (admission records); not assessed in IC	Matched unvaccinated controls	7 days after 2nd dose up to 28 days: mean follow-up: 18 days in subgroup analyses	1 – RR RR: vaccinated vs. unvaccinated	Matching analysis Kaplan-Meier analysis to compute risk of outcomes and then calculated risk ratios
Chodick G, et al. [24]	VE: SARS-CoV-2 infection (≥1 positive RT-PCR test) VE: Symptomatic COVID-19 (documented in EMRs) VE: Risk of COVID-19-related hospitalization/death among infected (extracted from EMRs); VE not reported in IC	Between 7 to 27 days after 2nd dose (protection-period) vs. days 1 to 7 after the 1st dose (reference period)	Days 7–27 after 2nd dose vs. days 1–7 after 1st dose	(1 – Relative Risk) X 100 Relative Risk: Protection-period vs. reference-period	Logistic regression was used to estimate odds ratios for symptomatic COVID-19 among infected cases Cox proportional hazards model were used to calculate adjusted hazard ratios for COVID-19-related hospitalization/death.
Yelin I, et al. [25]	VE: SARS-CoV-2 infection (positive RT-PCR test) VE: Symptomatic COVID-19; not assessed in IC	Unvaccinated controls	1–11 days 12–28 days 29–50 days (IC) All post-vaccination	(1-OR) X 100 OR: Positive test/symptomatic infection in vaccinated vs. unvaccinated	Logistic regression
Whitaker HJ, et al. [26]	VE: Symptomatic COVID-19 (diagnosis of COVID-19 or clinical illness consistent with COVID-19 within 10 days before or after a positive RT-PCR test)	Test-negative controls: Patients with symptoms within 10 days of SARS-CoV-2 negative test	≥14 days after 2nd dose up to ~6.5 months	Adjusted VE OR OR: Cases in vaccinated vs. unvaccinated	Logistic regression
Chemaitelly H, et al. [27]	VE: SARS-CoV-2 infection (positive RT-PCR test) VE: Severe COVID-19 illness (according to WHO criteria [38])	Unvaccinated controls	≥14 days after 2nd dose ≥42 days after 2nd dose ≥56 days after 2nd dose	1 – HR HR: vaccinated vs. unvaccinated	A proportional hazards model was used to calculate adjusted hazard ratios

COVID-19: Coronavirus disease 2019; EMR: Electronic medical records; HR: Hazard ratio; IC: Immunocompromised; NIH: National Institutes of Health; OR: Odds ratio; RR: Risk ratio; RT-PCR: Reverse transcription polymerase chain reaction; SARS-CoV-2: Severe acute respiratory syndrome coronavirus 2; VE: Vaccine effectiveness; VHA: Veterans Health Administration; WHO: World Health Organization.