Abstract: In response to the COVID-19 epidemic, Egypt established a unique care model based on quarantine hospitals where only externally-referred confirmed COVID-19 patients were admitted, and healthcare workers resided continuously over 1- to 2-week working shifts. While the COVID-19 risk for HCWs has been widely reported in standard healthcare settings, it has not been evaluated yet in quarantine hospitals. Here, we relied on longitudinal data, including results of routine RT-PCR tests, collected within three quarantine hospitals located in Cairo and Fayoum, Egypt. Using a mathematical model accounting for the time-since-exposure variation in false-negative rates of RT-PCR tests, we computed the incidence rate of SARS-CoV-2 infection among HCWs. Our results thus unveil the proportion of infections remaining undiagnosed despite routine testing. We estimated that the risk for an HCW to be infected during a working shift lied within the range of risk levels previously documented in standard healthcare settings for two quarantine hospitals, whereas it was > 3-fold higher for the third hospital. This large variation suggests that HCWs from quarantine hospitals may face a high occupational risk of infection, but that, with sufficient infection control measures, this risk can be brought down to levels similar to those observed in standard healthcare settings.