Deepening Our Understanding of COVID-19 Vaccine Decision-Making amongst Healthcare Workers in Southwest Virginia, USA Using Exploratory and Confirmatory Factor Analysis.
Jesse Bendetson, Mandy C Swann, Alicia Lozano, Jennifer West, Alexandra L Hanlon, Ian Crandell, Maimuna Jatta, Charles J Schleupner, Anthony Baffoe-Bonnie
Author Information
Jesse Bendetson: Virginia Tech Carilion School of Medicine, Roanoke, VA 24016, USA. ORCID
Mandy C Swann: Infection Prevention and Control Section, Carilion Clinic, Roanoke, VA 24014, USA. ORCID
Alicia Lozano: Virginia Tech Center for Biostatistics and Health Data Science, Roanoke, VA 24016, USA.
Jennifer West: Virginia Tech Center for Biostatistics and Health Data Science, Roanoke, VA 24016, USA.
Alexandra L Hanlon: Virginia Tech Center for Biostatistics and Health Data Science, Roanoke, VA 24016, USA.
Ian Crandell: Virginia Tech Center for Biostatistics and Health Data Science, Roanoke, VA 24016, USA.
Maimuna Jatta: Infection Prevention and Control Section, Carilion Clinic, Roanoke, VA 24014, USA. ORCID
Charles J Schleupner: Virginia Tech Carilion School of Medicine, Roanoke, VA 24016, USA.
Anthony Baffoe-Bonnie: Virginia Tech Carilion School of Medicine, Roanoke, VA 24016, USA. ORCID
Vaccine hesitancy amongst healthcare workers (HCWs) has been a major challenge throughout the COVID-19 pandemic. While many studies have identified HCW characteristics and specific attitudes associated with COVID-19 vaccine hesitancy, researchers are still working towards developing a holistic understanding of the psychological constructs that influence COVID-19 vaccine decision-making in this population. Between 15 March and 29 March 2021, we distributed an online survey assessing individual characteristics and vaccine-related perceptions to employees of a not-for-profit healthcare system in Southwest Virginia (N = 2459). We then performed exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) to describe patterns of vaccine-related thought amongst HCWs and identify latent psychometric constructs involved in vaccine decision-making. The goodness of model fit was assessed using the Tucker-Lewis Index (TLI), the Comparative Fit Index (CFI), and the Root Mean Square Error of Approximation (RMSEA). Internal consistency and reliability of each factor were assessed using Cronbach's alpha. EFA identified four latent psychometric constructs: Lack of trust in the COVID-19 vaccine; Anti-science sentiment; Adverse side-effects; and Situational risk assessment. The goodness of EFA model fit was adequate (TLI > 0.90, RMSEA ≤ 0.08) with acceptable internal consistency and reliability for three of four factors (Cronbach's alpha > 0.70). The CFA model also had adequate goodness of fit (CFI > 0.90, RMSEA ≤ 0.08). We believe the psychometric constructs identified in this study can provide a useful framework for interventions to improve vaccine uptake amongst this critical population.
