From Contextual Risk to Preventive Health Behavior in a Pandemic: A Serial Mediation Analysis

• In Event: 41021 - Student Forum Refereed Roundtables
  In Refereed Roundtable (90 minute): Roundtable 10: Health and Illness

Abstract

Background:
While research shows that contextual characteristics affect health outcomes, their effect on preventive behaviors is understudied. Therefore, this study examines the relationship between regional infection risk and adherence to COVID-19 preventive behavior, examining frequency of information about the local situation (from news and social contacts) and perceived susceptibility as mediators. Given the mixed empirical evidence on the role of perceived susceptibility, we also test a new operationalization conditional on taking no preventative measures.

Methods:
We combined official data on regional seven-day COVID-19 incidence rates (county-level) as an indicator of contextual risk with data from a large nationally representative web-based panel survey of the adult population in Germany (N=9,023). This survey assessed the frequency of receiving information about the regional COVID-19 situation from news and social contacts, perceived susceptibility of contracting COVID-19, and adherence to preventative measures. We tested a serial mediation model using structural equation modeling (SEM). Further robustness checks included repeating the analyses with multiple imputation.

Results:
Individuals in regions with higher contextual risk of infection have higher average rates of adherence to protective behaviors. Moreover, the relationship between incidence rates and adherence is serially mediated by information frequency (from news and social contacts) and perceived susceptibility.

Conclusions:
In sum, this study elucidates the complex pathways by which the proximal environment influences health-related information, cognitions, and behavior. Furthermore, it demonstrates the importance of perceived susceptibility by using a novel measure of the construct.

Authors

• Shannon Taflinger, University of Cologne

• Sebastian Sattler, Bielefeld University