Background: At the start of the coronavirus disease 2019 (COVID-19) pandemic, many national health organizations emphasized nonpharmacologic interventions, such as quarantining or physical distancing. In the United Kingdom, strict self-isolation (“shielding”) was advised for those deemed to be clinically extremely vulnerable on the basis of the presence of selected medical conditions or at the discretion of their general practitioners.

Down syndrome features on neither the UK shielding list nor the US Centers for Disease Control and Prevention list of groups at “increased risk.” However, it is associated with immune dysfunction, congenital heart disease, and pulmonary pathology and, given its prevalence, may be a relevant albeit unconfirmed risk factor for severe COVID-19 (1).

Objective: To evaluate Down syndrome as a risk factor for death from COVID-19 through a comprehensive analysis of individual-level data in a cohort study of 8.26 million adults (aged> 19 years), as part of a wider COVID-19 risk prediction project commissioned by the UK government (2).

Methods and Findings: We used QResearch, a population-level primary care database that has collected data for more than 35 million persons in England since 1998 and is linked at the individual patient level to severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) testing results from Public Health England, hospital episode statistics, and the Office of National Statistics death registry. Data extracted included age, sex, ethnicity, alcohol intake, smoking status, body mass index (BMI), a range of preexisting comorbid conditions, and concurrent medications. The primary outcome of interest was COVID-19 mortality in or out of the hospital, defined as confirmed or suspected COVID-19 on the death certificate or death within 28 days of a confirmed SARS-CoV-2 infection in the study period. The secondary outcome of interest was hospital admission related to COVID-19. The study period was 24 January 2020 (first confirmed SARS-CoV-2 infection in the United Kingdom) to 30 June 2020. We used Cox proportional hazards models to estimate adjusted hazard ratios (HRs) with 95% CIs, accounting for death from non–COVID-19 causes as a competing event by censoring all persons who did not have the outcome of interest at the study end date. We tested for interactions between Down syndrome and age, BMI, and sex.