As SARS-CoV-2 vaccines enter mass distribution, little is known about vaccinated individuals who nonetheless develop Covid-19 (symptomatic disease or “vaccine failure”), require hospitalization, and die. Pfizer-BioNTech reported 3.8 vaccine failure cases/1,000 person-years, meaning vaccine failure will be relatively common on a population level. Pfizer, Moderna, and Janssen vaccine efficacy trials involved 35,000 to 50,000 individuals from which there are insufficient numbers to study vaccine failure, calling for post-vaccination surveillance with observational cohort studies in real-world settings. Using the VHA Corporate Data Warehouse (CDW) and the Covid-19 Shared Data Resource, we can harness real-time data to design a forward-thinking, active surveillance system to pragmatically study the epidemiology of vaccine failure. Our overarching goals in the proposed VAccine Failure study are to characterize post-vaccination Covid-19 clinical outcomes and identify factors associated with vaccine failure. Our central hypothesis is that adverse Covid-19 clinical outcomes after vaccination will be increasingly common over time, driven by delayed vaccination, clinical factors associated with weak or rapidly declining immune responses, and viral lineages accumulating escape mutations. We propose to identify a national longitudinal cohort of all Veterans age >65 years who had a primary care visit in the VA within the past 2 years and received at least 1 dose of any SARS-CoV-2 vaccine (N>1.55 million Veterans). We will observe these Veterans for up to 3.5 years and ascertain diagnoses of Covid-19 incidence or vaccine failure (acute SARS-CoV-2 infection + at least 1 symptom), hospitalization, and mortality. We will leverage our extensive experience conducting observational cohort studies with VA and Medicare data to emulate a target trial approach and achieve the following Aims, rooted in the epidemiological triad: Aim 1. To describe and compare Covid-19 clinical outcomes by vaccine type and manufacturer in a population-based national cohort of SARS-CoV-2 vaccinated Veterans. Aim 2. To determine environmental factors associated with vaccine failure. Aim 3. To determine host factors associated with vaccine failure. Aim 4. To determine agent factors associated with vaccine failure. To achieve Aim 4, we will use a case-control study of 600 Veterans with Covid-19 to identify cases with the outcome of viral variant and perform density sampling of controls from our SFVAHCS/VISN-21 catchment area, where we have access to banked remnant SARS-CoV-2 PCR-positive samples. This project is uniquely positioned to provide the National VA Vaccine Program with real-time evidence to rapidly adapt its vaccine delivery for Veterans (e.g., vaccine boosters for at- risk groups; timing of next-generation vaccines) while filling the urgent national need for post-vaccination surveillance in real-world settings.