Abstract: Predictors of COVID-19 Risk Behaviors: A Longitudinal Study of Veterans and non-Veterans

◆ Lingzi Zhong, University of Utah School of Medicine
◆ Alistair Thorpe, University of Utah School of Medicine
◆ Laura D Scherer, University of Colorado School of Medicine
◆ Frank A Drews, University of Utah
◆ Nicole Burpo, University of North Carolina at Chapel Hill
◆ Angela Fagerlin, University of Utah

Introduction. The COVID-19 pandemic has had a devastating impact on communities world-wide. Communicating the importance of avoiding risk-increasing behaviors and adopting protective health behaviors has been critical for limiting the spread and impact of COVID-19. Identifying individual factors that affect the public’s engagement with health behaviors (whether risk-increasing or protective) is crucial for the design and communication of public health guidelines for COVID-19 and future health crises. Thus, our aim was to assess the demographic, structural, and psychological predictors of avoiding risk-increasing behaviors and adopting protective health behaviors between December-2020 and March-2021. Method. This study used data from an online longitudinal, three-wave COVID-19 survey regarding the behaviors, attitudes, and experiences of US Veteran and non-Veteran adults. Three surveys were conducted between December-2020 and March-2021 (December 2-27 (N=2,085); January 21-February 6 (N=1,257), and March 8-23 (N=1,075)). Primary outcomes were self-reported frequency of engaging in risk-increasing behaviors (e.g., Going to gatherings of 10 people or more) and wearing a mask when in a public place (6-point Likert scale, 1=“Never”, 2=“Very rarely”, 3=“Rarely”, 4=“Occasionally”, 5=“Frequently”, 6=“Very frequently”). Linear multiple regression was used to model the data. Predictors used in these models were age, location of residence (rural vs. urban), Veteran status, health status, health literacy, subjective numeracy, Race/Ethnicity, COVID-19 worry and risk perception, COVID-19 risk perception, facilitators of distancing (e.g., getting groceries delivered), trust in health care, belief in science, belief in conspiracies, political views, and attitudes towards government response to COVID-19. Results. Not being able to get groceries delivered emerged as the strongest predictor of more frequent risk-increasing behavior across all timepoints. Other consistent predictors of more frequent risk-increasing behavior and less mask wearing included less worry about getting COVID-19, lack of belief in science, more belief in conspiracy theories, and anger/disagreement with the state response. No demographic factor consistently predicted risk-increasing behavior or mask wearing throughout the study, though different demographic predictors emerged for more frequent risk-increasing behaviors (younger age, higher health literacy, and identifying as non-Hispanic White) and more frequent mask-wearing (older age, urban residence, lower health literacy, and higher subjective numeracy) at certain timepoints. The most frequently endorsed reasons for having contact with others reflected both health-related (food, medical care, and exercise) and social needs (desires to see friends/family in person and feelings of boredom). Discussion. Public behavior continues to play a critical role in the spread and the impact of the COVID-19 pandemic. The current research adds to our understanding of the individual determinants of risk-increasing behaviors and mask wearing during the COVID-19 pandemic that encompass demographic, psychological, structural, and political factors. These findings highlight the plethora of factors associated with important health behaviors during crises such as the COVID-19 pandemic and the need for more tailored interventions to support public health experts and health communicators promote uptake of protective behaviors and address barriers to engaging in risk-reducing behaviors.