No abstract

OBJECTIVE: To estimate the seroprevalence of antibodies to SARS-CoV-2 among blood donors in the state of Rio de Janeiro, Brazil. METHODS: Data were collected on 2,857 blood donors from April 14 to 27, 2020. This study reports crude prevalence of antibodies to SARS-CoV-2, population weighted prevalence for the state, and prevalence adjusted for test sensitivity and specificity. Logistic regression models were used to establish the correlates of SARS-CoV-2 prevalence. For the analysis, we considered collection period and site, sociodemographic characteristics, and place of residence. RESULTS: The proportion of positive tests for SARS-Cov-2, without any adjustment, was 4.0% (95%CI 3.3–4.7%), and the weighted prevalence was 3.8% (95%CI 3.1–4.5%). We found lower estimates after adjusting for test sensitivity and specificity: 3.6% (95%CI 2.7–4.4%) for the non-weighted prevalence, and 3.3% (95%CI 2.6–4.1%) for the weighted prevalence. Collection period was the variable most significantly associated with crude prevalence: the later the period, the higher the prevalence. Regarding sociodemographic characteristics, the younger the blood donor, the higher the prevalence, and the lower the education level, the higher the odds of testing positive for SARS-Cov-2 antibody. We found similar results for weighted prevalence. CONCLUSIONS: Our findings comply with some basic premises: the increasing trend over time, as the epidemic curve in the state is still on the rise; and the higher prevalence among both the youngest, for moving around more than older age groups, and the less educated, for encountering more difficulties in following social distancing recommendations. Despite the study limitations, we may infer that Rio de Janeiro is far from reaching the required levels of herd immunity against SARS-CoV-2.

4.25 % (95%CI: 2.82; 71), 5.74% (95%CI: 3.80; 7.71), and 2.29% (95%CI: 0.86; 3.73)

A new algorithm for adjusting correlation matrices and for comparison with Finger's algorithm, which is used to compute Value-at-Risk in RiskMetrics for stress test scenarios. The solution proposed by the new methodology is always better than Finger's approach in the sense that it alters as little as possible those correlations that one would wish not to alter, but they change in order to obtain a consistent Finger correlation matrix.Correlation Matrix, Kuhn-TUCKER Conditions, Eigenvalue, Value-AT-RISK,

No abstract