Data on effectiveness and protection duration of Covid-19 vaccines and previous infection against general SARS-CoV-2 infection in general populations are limited. Here we evaluate protection from Covid-19 vaccination (primary series) and previous infection in 21,261 university students undergoing repeated surveillance testing between 8/8/2021–12/04/2021, during which B.1.617 (delta) was the dominant SARS-CoV-2 variant. Estimated mRNA-1273, BNT162b2, and AD26.COV2.S effectiveness against any SARS-CoV-2 infection is 75.4% (95% CI: 70.5-79.5), 65.7% (95% CI: 61.1-69.8), and 42.8% (95% CI: 26.1–55.8), respectively. Among previously infected individuals, protection is 72.9% when unvaccinated (95% CI: 66.1–78.4) and increased by 22.1% with full vaccination (95% CI: 15.8–28.7). Statistically significant decline in protection is observed for mRNA-1273 (P < .001), BNT162b2 (P < .001), but not Ad26.CoV2.S (P = 0.40) or previous infection (P = 0.12). mRNA vaccine protection dropped 29.7% (95% CI: 17.9–41.6) six months post- vaccination, from 83.2% to 53.5%. We conclude that the 2-dose mRNA vaccine series initially offers strong protection against general SARS-CoV-2 infection caused by the delta variant in young adults, but protection substantially decreases over time. These findings indicate that vaccinated individuals may still contribute to community spread. While previous SARS-CoV-2 infection consistently provides moderately strong protection against repeat infection from delta, vaccination yields a substantial increase in protection.

Existing conversational recommendation (CR) systems usually suffer from insufficient item information when conducted on short dialogue history and unfamiliar items. Incorporating external information (e.g., reviews) is a potential solution to alleviate this problem. Given that reviews often provide a rich and detailed user experience on different interests, they are potential ideal resources for providing high-quality recommendations within an informative conversation. In this paper, we design a novel end-to-end framework, namely, Review-augmented Conversational Recommender (RevCore), where reviews are seamlessly incorporated to enrich item information and assist in generating both coherent and informative responses. In detail, we extract sentiment-consistent reviews, perform review-enriched and entity-based recommendations for item suggestions, as well as use a review-attentive encoder-decoder for response generation. Experimental results demonstrate the superiority of our approach in yielding better performance on both recommendation and conversation responding. 1

Existing conversational recommendation (CR) systems usually suffer from insufficient item information when conducted on short dialogue history and unfamiliar items. Incorporating external information (e.g., reviews) is a potential solution to alleviate this problem. Given that reviews often provide a rich and detailed user experience on different interests, they are potential ideal resources for providing high-quality recommendations within an informative conversation. In this paper, we design a novel end-to-end framework, namely, Review-augmented Conversational Recommender (RevCore), where reviews are seamlessly incorporated to enrich item information and assist in generating both coherent and informative responses. In detail, we extract sentiment-consistent reviews, perform review-enriched and entity-based recommendations for item suggestions, as well as use a review-attentive encoder-decoder for response generation. Experimental results demonstrate the superiority of our approach in yielding better performance on both recommendation and conversation responding. 1

We aim to estimate the effectiveness of 2-dose and 3-dose mRNA vaccination (BNT162b2 and mRNA-1273) against general Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) infection (asymptomatic or symptomatic) caused by the omicron BA.1 variant. This propensity-score matched retrospective cohort study takes place in a large public university undergoing weekly Coronavirus Disease 2019 (Covid-19) testing in South Carolina, USA. The population consists of 24,145 university students and employees undergoing weekly Covid-19 testing between January 3rd and January 31st, 2022. The analytic sample was constructed via propensity score matching on vaccination status: unvaccinated, completion of 2-dose mRNA series (BNT162b2 or mRNA-1273) within the previous 5 months, and receipt of mRNA booster dose (BNT162b2 or mRNA-1273) within the previous 5 months. The resulting analytic sample consists of 1,944 university students (mean [SD] age, 19.64 [1.42] years, 66.4% female, 81.3% non-Hispanic White) and 658 university employees (mean [SD] age, 43.05 [12.22] years, 64.7% female, 83.3% non-Hispanic White). Booster protection against any SARS-CoV-2 infection was 66.4% among employees (95% CI: 46.1–79.0%; P < .001) and 45.4% among students (95% CI: 30.0–57.4%; P < .001). Compared to the 2-dose mRNA series, estimated increase in protection from the booster dose was 40.8% among employees (P = .024) and 37.7% among students (P = .001). We did not have enough evidence to conclude a statistically significant protective effect of the 2-dose mRNA vaccination series, nor did we have enough evidence to conclude that protection waned in the 5-month period after receipt of the 2nd or 3rd mRNA dose. Furthermore, we did not find evidence that protection varied by manufacturer. We conclude that in adults 18–65 years of age, Covid-19 mRNA booster doses offer moderate protection against general SARS-CoV-2 infection caused by the omicron variant and provide a substantial increase in protection relative to the 2-dose mRNA vaccination series.

Multivariate count data are common in many disciplines. The variables in such data often exhibit complex positive or negative dependency structures. We propose three Bayesian approaches to modeling bivariate count data by simultaneously considering covariate-dependent means and correlation. A direct approach utilizes a bivariate negative binomial probability mass function developed in Famoye (2010, Journal of Applied Statistics). The second approach fits bivariate count data indirectly using a bivariate Poisson-gamma mixture model. The third approach is a bivariate Gaussian copula model. Based on the results from simulation analyses, the indirect and copula approaches perform better overall than the direct approach in terms of model fitting and identifying covariate-dependent association. The proposed approaches are applied to two RNA-sequencing data sets for studying breast cancer and melanoma (BRCA-US and SKCM-US), respectively, obtained through the International Cancer Genome Consortium.