Abstract Introduction COVID-19 caused a high burden of sick leave worldwide. Long-term sick leave for COVID-19 may be longer than for other influenza-like syndromes. The real impact of COVID on absenteeism remains uncertain. The aim of our study was to investigate the burden of sick leave, especially > 12 weeks, in Belgian workers with a positive PCR test for SARS-CoV-2 from July 2020 to September 2021 and to compare these figures with sick leave for other infectious diseases. Methods We coupled a database of SARS-CoV-2–positive workers and workers who were absent for other infections with objective absence data. Predictors of prolonged sickness were evaluated by negative binomial regression, Cox proportional-hazards regression and ordinal logistic regression. Results The study population involved 2569 workers who tested positive for SARS-CoV-2 and 392 workers who were absent for other infectious diseases. In total, 16% (95%CI: 14-17%) of workers with a positive SARS-CoV-2 test had no sick leave registered. The prevalence of long-term sick leave was 1.39% (95%CI: 0.94-1.97) in workers with COVID-19 and 4.34% (95%CI: 2.55-6.85) in workers with other infectious diseases. When including PCR-positive workers without sick leave, the prevalence of long-term sick leave decreased to 1.17% (95%CI: 0.79-1.66). Long sick leave was associated with older age, high previous sick leave and low educational level. Discussion and conclusion The prevalence of long-term sick leave was lower than estimated in earlier investigations, regardless of worrying reports about post-COVID-19 syndrome.

Some occupational sectors, such as human health and care, food service, cultural and sport activities, have been associated with a higher risk of SARS-CoV-2 infection than other sectors. To curb the spread of SARS-CoV-2, it is preferable to apply targeted non-pharmaceutical interventions on selected economic sectors, rather than a full lockdown. However, the effect of these general and sector-specific interventions on the virus circulation has only been sparsely studied. We assess the COVID-19 incidence under different levels of non-pharmaceutical interventions per economic activity during the autumn 2020 wave in Belgium. The 14-day incidence of confirmed COVID-19 cases per the Statistical Classification of Economic Activities in the European Community (NACE–BEL) sector is modelled by a longitudinal Gaussian-Gaussian two-stage approach. This is based on exhaustive data on all employees in all sectors. In the presence of sanitary protocols and minimal non-pharmaceutical interventions, many sectors with close contact with others show considerably higher COVID-19 14-day incidences than other sectors. The effect of stricter non-pharmaceutical interventions in the general population and non-essential sectors is seen in the timing of the peak incidence and the width and height of the post-peak incidence. In most sectors incidences returned to higher levels after the peak than before and this decrease took longer for the health and care sector. Sanitary protocols for close proximity occupations may be sufficient during periods of low-level virus circulation, but progressively less with increasing circulation. Stricter general and sector-specific non-pharmaceutical interventions adequately decrease COVID-19 incidences, even in close proximity in essential sectors under solely sanitary protocols.