Background
To rapidly characterize COVID–19 epidemiology, numerous population-based studies have been undertaken to model the risk of SARS–CoV–2 infection. Less is known about what may drive the probability to undergo testing. Understanding how much testing is driven by contextual or individual conditions provides the basis for unbiased representation of testing figures and shaping of public health interventions and resource allocation.
Methods
In the Val Venosta/Vinschgau district (South Tyrol, Italy), we conducted a population-representative longitudinal study on 697 individuals who completed 4,512 repeated online questionnaires at four week intervals between September 2020 and May 2021. Mixed–effects logistic regression models were fitted to investigate associations between self-reported SARS–CoV–2 testing and both individual characteristics (social, demographic, and biological) and contextual determinants.
Results
Testing was associated with month of reporting, reflecting the timing of both the pandemic intensity and public health interventions, COVID–19–related symptoms (odds ratio, OR:8.26; 95% confidence interval, CI:6.04–11.31), contacts with infected individuals within home (OR:7.47, 95%CI:3.81–14.62) or outside home (OR:9.87, 95%CI:5.78–16.85), and being retired (OR:0.50, 95%CI:0.34–0.73). Within– and outside–home contacts were the leading determinants in the first and second half of the period, respectively, possibly reflecting later relaxation of social–distancing measures. Testing was not associated with age, sex, education, comorbidities, or lifestyle factors.
Conclusions
In the study area, contextual determinants reflecting the course of the pandemic were predominant compared to individual sociodemographic characteristics in explaining the SARS–CoV–2 probability testing. Decision makers should evaluate whether the intended target groups were correctly prioritized by the testing campaign.