It looks like they looked at both infected and not infected individuals: >>The crude prevalence of persistent physical symptoms was first calculated for 4 grou

It looks like they looked at both infected and not infected individuals: >>The crude prevalence of persistent physical symptoms was first calculated for 4 groups of participants according to both belief (ie, self-reported COVID-19 infection) and serology test results: belief negative and serology negative; belief positive and serology negative; belief negative and serology positive; and belief positive and serology positive.