“…Ideally, research into risks of COVID-19 by occupation would be based on cases ascertained through systematic prospective follow-up of large cohorts of workers, with regular ascertainment of symptoms, clinical signs, and polymerase chain reaction (PCR) testing. As that has not been feasible, researchers have been forced to use proxy measures of incidence based on: PCR testing (4)(5)(6)(7)(8)(9)(10)(11), measurements of specific immunoglobulins against SARS-CoV-2 (12)(13)(14), symptoms and sickness absence (6), COVID-19 related hospital admission (7,15,16), or mortality from the disease (7,15,(17)(18)(19)(20)(21)(22). In most studies to date, with broad coverage of occupations, mortality from COVID-19 has been the outcome measure (7,15,(17)(18)(19)(20)(21)(22).…”