Search

Background: Asymptomatic SARS-CoV-2 infection occurring in RT-PCR negative individuals represent a poorly characterized cohort with important infection control connotations. While household and community-based studies have evaluated seroprevalence of antibody and transmission dynamics in this group, workplace-based data is currently unavailable. Methods: A cohort study was carried out in July 2021, during and immediately following the peak of the 3rd wave of COVID-19 in Sri Lanka, prior to mass vaccination. A total of 92 unvaccinated individuals between the ages of 17–65 years were purposively sampled from an office and two factory settings. The selected cohort that had been exposed to RT-PCR positive cases in the workplace was tested RT-PCR negative. Serological samples collected six weeks post exposure were tested for anti-SARS-CoV-2 neutralizing antibody. Results: The seroprevalence for SARS-CoV-2 specific neutralizing antibodies in the overall cohort was 63.04% (58/92). Seroprevalences in the office setting, factory setting 1 and factory setting 2 were 69.2% (9/13), 55.7% (34/61) and 83.33% (15/18), respectively. Primary risk factor associated with seropositivity was face to face contact with no mask for > 15 min (p < 0.024, Odds Ratio (OR); 5.58, 95%CI;1.292– 25.65). Individuals with workspace exposure had significantly higher levels of neutralizing antibodies than those who did not (percentage neutralization in assay 63.3% (SD:21)vs 45.7% (SD:20), p = 0.0042), as did individuals who engaged socially without protective measures (62.4 (SD:21.6)% vs 49.7 (SD:21)%, p = 0.026). Conclusion: There was a high seroprevalence for SARS-CoV-2 specific neutralizing antibodies among RT-PCR negative contacts in workplace settings in Sri Lanka. Higher levels of transmission of SARS-CoV-2 infection than estimated based on RT-PCR positive contact data indicate need for targeted infection control measures in these settings during future outbreaks. [ABSTRACT FROM AUTHOR]

The categorical Gini correlation, ρ g , was proposed by Dang et al. (2021) to measure the dependence between a categorical variable, Y , and a numerical variable, X. It has been shown that ρ g has more appealing properties than current existing dependence measurements. In this paper, we develop the jackknife empirical likelihood (JEL) method for ρ g. Confidence intervals for the Gini correlation are constructed without estimating the asymptotic variance. Adjusted and weighted JEL are explored to improve the performance of the standard JEL. Simulation studies show that our methods are competitive to existing methods in terms of coverage accuracy and shortness of confidence intervals. The proposed methods are illustrated in an application on two real datasets. • Confidence intervals for the categorical Gini correlation are constructed without knowing the underlying distributions. • The proposed project is easy to implement without estimating the complicated asymptotic variance. • The proposed project is computational efficient. [ABSTRACT FROM AUTHOR]