Risk Factors for COVID-19: Community Exposure and Mask-Wearing
19 Pages Posted: 8 Sep 2020
Date Written: August 18, 2020
Importance: Many studies have focused on characteristics of symptomatic COVID-19 patients and clinical risk factors. This study reports prevalence of COVID-19 in the general population and identifies factors that affect exposure to the virus.
Objective: To measure the prevalence of COVID-19 in a hospital service area and identify factors that may increase or decrease the risk of infection and exposure.
Design: This cohort study collected survey information relating to work and living situations, income, behavior, socio-demographic characteristics and pre-pandemic health characteristics. This data was combined with polymerase chain reaction (PCR) testing and two different serologic assays.
Setting and Participants: Our sampling frame was the primary care population of a Level 1 medical center in the Northeast U.S.. A random sample was drawn, stratified by age and gender. About 20 percent opted in and 1,694 completed the survey. Participants were invited to receive PCR and serologic testing. A total of 454 individuals provided samples between June 25th and June 28th, 2020.
Research Methods: Survey data was collected via Redcap. PCR testing was conducted at a community testing site using nasopharyngeal swabs. Serologic testing was done using two different methods to ensure the reliability of the results.
Main Outcomes: Positivity rate was used to calculate approximate prevalence, hospitalization rate and infection fatality rate (IFR). Survey data was used to analyze risk factors, including the number of contacts reported by study participants.
Results: We found a positivity rate of 2.2 percent, a hospitalization rate of 1.2 percent and an adjusted IFR of 0.55 percent. The number of contacts with adults and seniors increases the probability of becoming infected. Occupation, living in apartment versus a house, and wearing a facial mask outside work increased probability of COVID-19 infection.
Conclusions and Relevance: Based on the IFR and the number of deaths, estimations about the total number of infections in similar demographic areas with different infection rates can be made. Comparing this number with the number of officially reported infections leads to an estimate of unreported cases. Occupational, living-situation, and behavioral data may aid in the identification of non-clinical factors affecting SARS-CoV-2 exposure and infection.
Note: Funding: Serology tests performed at UVM were funded by a pilot grant to SAD from the UVM Translational Global Infectious Disease Research Center (National Institute of Health grant P20GM125498). Additional funding was from NIH grant U01AI141997 to SAD, and the University of Vermont Larner College of Medicine Departments Emergency Medicine and Surgery. The University of Vermont Department of Radiology funded serologic tests performed at Mayo. The Vermont Department of Health funded PCR tests performed at the state laboratory. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Declaration of Interest: None to declare
Ethical Approval Statement: This study has been approved by the University of Vermont and State Agricultural College Institutional Review Board: FWA 00000723, Expiration Date: March 14, 2024 and the UVM Medical Center: FWA 00000727, Expiration Date: March 14, 2024; IRB number 00000485.
Keywords: COVID-19, random sample testing approach, population health research, mask wearing, risk factors, social distancing
JEL Classification: I12, I18, I19, H51
Suggested Citation: Suggested Citation