Objectives: Bacille Calmette-Guérin (BCG) vaccine has immunomodulatory effects that may provide protection against unrelated infectious diseases. We aimed to determine whether BCG vaccination protects adults against COVID-19.

Design: Phase III double blind randomised controlled trial.

Setting: Healthcare centres in Australia, Brazil, the Netherlands, Spain, and the United Kingdom during the COVID-19 pandemic.

Participants: 3988 healthcare workers with no prior COVID-19 and no contraindication to BCG.

Intervention: Randomised 1:1 using a web-based procedure to receive a single 0.1mL intradermal dose of BCG-Denmark (BCG group, n=1999) or saline (placebo group, n=1989).

Main outcome measures: Difference in incidence of (i) symptomatic and (ii) severe COVID-19 during the 12-months following randomisation in the modified intention to treat (mITT) population (confirmed SARS-CoV-2 naïve at inclusion).

Results: Of the 3988 participants randomised, 3386 had a negative baseline SARS-CoV-2 test and were included in the mITT population. The 12-month adjusted estimated risk of symptomatic COVID-19 was higher in the BCG group (22.6%; 95% confidence interval [CI] 20.6% to 24.5%) compared with the placebo group (19.6%; 95%CI 17.6% to 21.5%); adjusted difference +3.0 percentage points (95%CI 0.2% to 5.8%; p=0.04). The 12-month adjusted estimated risk of severe COVID-19 (mainly comprising those reporting being unable to work for ≥3 consecutive days) was 11.0% in the BCG group (95%CI 9.5% to 12.4%) compared with 9.6% in the placebo group (95%CI 8.3% to 11.1%); adjusted difference +1.3 percentage points (95%CI -0.7% to 3.3%, p=0.2). Breakthrough COVID-19 (post COVID-19 vaccination), and asymptomatic SARS-CoV-2 infections were similar in the two groups. There were 18 hospitalisations due to COVID-19 (11 in BCG group, 7 in placebo group; adjusted hazard ratio 1.56, 95%CI 0.60 to 4.02, p=0.4) and two deaths due to COVID-19, both in the placebo group.

Conclusions: Compared to placebo, vaccination with BCG-Denmark increased the risk of symptomatic COVID-19 over 12 months among health care workers and did not decrease the risk of severe COVID-19 or post-vaccination breakthrough COVID-19.

Trial registration: ClinicalTrials.gov NCT04327206.