Association Between the Community Prevalence of COVID-19 and Daily Unscheduled Absences of Anesthesiologists, Nurse Anesthetists, and Residents in an Academic Anesthesia Department

Abstract

Introduction An "unscheduled absence" refers to an occurrence when an employee does not appear for work and the absence was not approved in advance by an authorized supervisor. Daily unscheduled absences need to be forecasted when doing staff scheduling to maintain an acceptable risk of being unable to run all anesthetizing locations and operating rooms planned. The number of extra personnel to be scheduled needs to be at least twice as large as the mean number absent. In an earlier historical cohort study, we found that our department's modeled risks of being unavailable unexpectedly differed among types of anesthesia practitioners (e.g., anesthesiologists and nurse anesthetists) and among weekdays (i.e., Mondays, Fridays, and workdays adjacent to holidays versus other weekdays). In the current study, with two extra years of data, we examined the effect of the coronavirus COVID-19 pandemic on the frequency of unscheduled absences. Methods There were 50 four-week periods studied at a large teaching hospital in the United States, from August 30, 2018 to June 29, 2022. The sample size of 120,687 person-assignment days (i.e., a person assigned to work on a given day) included 322 anesthesia practitioners (86 anesthesiologists, 88 certified registered nurse anesthetists, 99 resident and fellow physicians, and 49 student nurse anesthetists). The community prevalence of COVID‑19 was estimated using the percentage positive among asymptomatic patients tested before surgery and other interventional procedures at the hospital. Results Each 1% increase in the prevalence of COVID-19 among asymptomatic patients was associated with a 1.131 increase in the odds of unscheduled absence (P < 0.0001, 99% confidence interval 1.086 to 1.178). Using an alternative model with prevalence categories, unscheduled absences were substantively more common when the COVID-19 prevalence exceeded 2.50%, P [Formula: see text] 0.0002. For example, there was a 1% unscheduled absence rate among anesthesiologists working Mondays and Fridays early in the pandemic when the prevalence of COVID-19 among asymptomatic patients was 1.3%. At a 1% unscheduled absence rate, 67 would be the minimum scheduled to maintain a <5.0% risk for being unable to run all 65 anesthetizing locations. In contrast, there was a 3% unscheduled absence rate among nurse anesthetists working Mondays and Fridays during the Omicron variant surge when the prevalence was 4.5%. At a 3% unscheduled absence rate, 70 would be the minimum scheduled to maintain the same risk of not being able to run 65 rooms. Conclusions Increases in the prevalence of COVID-19 asymptomatic tests were associated with more unscheduled absences, with no detected threshold. This quantitative understanding of the impact of communicable diseases on the workforce potentially has broad generalizability to other fields and infectious diseases.