Impact of healthcare worker shift scheduling on workforce preservation during the COVID-19 pandemic

Dan M Kluger; Yariv Aizenbud; Ariel Jaffe; Lilach Aizenbud; Fabio Parisi; Eyal Minsky-Fenick; Jonathan M Kluger; Shelli Farhadian; Harriet M Kluger; Yuval Kluger

We simulated the spread of SARS-CoV-2 in hospital wards using five scheduling designs with various choices of model input parameters. Universal model parameters for COVID-19 included incubation period distribution (time from exposure to first symptom) and latent period distribution (time from exposure to becoming infectious.) Situation-dependent COVID-19 model parameters included pre-admission infection probability of an admitted patient, team member infection probability at start of simulation, physician-to-patient, nurse-to-patient, patient-tophysician, patient-to-nurse, and HCW-to-HCW transmission probabilities, team member days of absence after symptom onset, daily SARS-CoV-2 exposure probability of team members (e.g. via elevator use, exposure to other staff), length of patient stays after showing COVID-19 symptoms, and length of simulation time. Model parameters that varied by hospital setting and service type included average team patient census, average patient hospitalization length, and the number of physicians and nurses on a team and on duty at all times.