RESUMEN
BACKGROUND: We sought to evaluate the impact of changes in estimates of COVID-19 vaccine effectiveness on the incidence of laboratory-confirmed infection among frontline workers at high risk for SARS-CoV-2. METHODS: We analyzed data from a prospective frontline worker cohort to estimate the incidence of COVID-19 by month as well as the association of COVID-19 vaccination, occupation, demographics, physical distancing, and mask use with infection risk. Participants completed baseline and quarterly surveys, and each week self-collected mid-turbinate nasal swabs and reported symptoms. RESULTS: Among 1018 unvaccinated and 3531 fully vaccinated workers, the monthly incidence of laboratory-confirmed SARS-CoV-2 infection in January 2021 was 13.9 (95% confidence interval [CI]: 10.4-17.4), declining to 0.5 (95% CI -0.4-1.4) per 1000 person-weeks in June. By September 2021, when the Delta variant predominated, incidence had once again risen to 13.6 (95% CI 7.8-19.4) per 1000 person-weeks. In contrast, there was no reportable incidence among fully vaccinated participants at the end of January 2021, and incidence remained low until September 2021 when it rose modestly to 4.1 (95% CI 1.9-3.8) per 1000. Below average facemask use was associated with a higher risk of infection for unvaccinated participants during exposure to persons who may have COVID-19 and vaccinated participants during hours in the community. CONCLUSIONS: COVID-19 vaccination was significantly associated with a lower risk of SARS-CoV-2 infection despite Delta variant predominance. Our data demonstrate the added protective benefit of facemask use among both unvaccinated and vaccinated frontline workers.
Asunto(s)
COVID-19 , Socorristas , COVID-19/epidemiología , COVID-19/prevención & control , Vacunas contra la COVID-19 , Atención a la Salud , Humanos , Incidencia , Estudios Prospectivos , SARS-CoV-2/genética , VacunaciónRESUMEN
When analyzing data, researchers may have multiple reasonable options for the many decisions they must make about the data-for example, how to code a variable or which participants to exclude. Therefore, there exists a multiverse of possible data sets. A classic multiverse analysis involves performing a given analysis on every potential data set in this multiverse to examine how each data decision affects the results. However, a limitation of the multiverse analysis is that it addresses only data cleaning and analytic decisions, yet researcher decisions that affect results also happen at the data-collection stage. I propose an adaptation of the multiverse method in which the multiverse of data sets is composed of real data sets from studies varying in data-collection methods of interest. I walk through an example analysis applying the approach to 19 studies on shooting decisions to demonstrate the usefulness of this approach and conclude with a further discussion of the limitations and applications of this method.