This work aimed to study the role of different SARS-CoV-2 lineages in the epidemiology of multiple waves of the COVID-19 pandemic in Ribeirão Preto (São Paulo state), with comparison within Brazil and globally. Viral genomic sequencing was combined with clinical and sociodemographic information of 2,379 subjects at a large Brazilian hospital. On the whole 2,395 complete SARS-CoV-2 genomes were obtained from April 2020 to January 2022. We report variants of concern (VOC) and interest (VOI) dynamics and the role of Brazilian lineages. We identified three World Health Organization VOCs (Gamma, Delta, Omicron) and one VOI (Zeta), which caused distinct waves in this cohort. We also identified 47 distinct Pango lineages. Consistent with the high prevalence of Gamma in Brazil, Pango lineage P.1 dominated infections in this cohort for half of 2021. Each wave of infection largely consisted of a single variant group, with each new group quickly and completely rising to dominance. Despite increasing vaccination in Brazil starting in 2021, this pattern was observed throughout the study and is consistent with the hypothesis that herd immunity tends to be SARS-CoV-2 variant-specific and does not broadly protect against COVID-19.

In this paper we study a variation of the random $k$-SAT problem, called polarised random $k$-SAT, which contains both the classical random $k$-SAT model and the random version of monotone $k$-SAT another well-known NP-complete version of SAT. In this model there is a polarisation parameter $p$, and in half of the clauses each variable occurs negated with probability $p$ and pure otherwise, while in the other half the probabilities are interchanged. For $p=1/2$ we get the classical random $k$-SAT model, and at the other extreme we have the fully polarised model where $p=0$, or 1. Here there are only two types of clauses: clauses where all $k$ variables occur pure, and clauses where all $k$ variables occur negated. That is, for $p=0$, and $p=1$, we get an instance of random monotone $k$-SAT.

We show that the threshold of satisfiability does not decrease as $p$ moves away from $\frac{1}{2}$ and thus that the satisfiability threshold for polarised random $k$-SAT with $p\neq \frac{1}{2}$ is an upper bound on the threshold for random $k$-SAT. Hence the satisfiability threshold for random monotone $k$-SAT is at least as large as for random $k$-SAT, and we conjecture that asymptotically, for a fixed $k$, the two thresholds coincide.

Cyclosporiasis results from an infection of the small intestine by Cyclospora parasites after ingestion of contaminated food or water, often leading to gastrointestinal distress. Recent developments in temporally linking genetically related Cyclospora isolates demonstrated effectiveness in supporting epidemiological investigations. We used ‘temporal-genetic clusters’ (TGCs) to investigate reported cyclosporiasis cases in the United States during the 2021 peak-period (1 May – 31 August 2021). Our approach split 655 genotyped isolates into 55 genetic clusters and 31 TGCs. We linked two large multi-state epidemiological clusters (Epidemiologic Cluster 1 [n = 136 cases, 54 genotyped] and Epidemiologic Cluster 2 [n = 42 cases, 15 genotyped]) to consumption of lettuce varieties; however, product traceback did not identify a specific product for either cluster due to the lack of detailed product information. To evaluate the utility of TGCs, we performed a retrospective case study comparing investigation outcomes of outbreaks first detected using epidemiological methods with those of the same outbreaks had TGCs been used to first detect them. Our study results indicate that adjustments to routine epidemiological approaches could link additional cases to epidemiological clusters of cyclosporiasis. Overall, we show that CDC’s integrated genotyping and epidemiological investigations provide valuable insights into cyclosporiasis outbreaks in the United States.

A register-based retrospective observational study was conducted to describe SARS-CoV-2 cases and case-clusters in schoolchildren of Danish primary and lower secondary schools and identify which factors were associated with the occurrence of case-clusters in schools. The study period was the autumn school semester 2021. Clusters were defined as three or more cases in a school-class level within 14 days. Descriptive analysis was carried out and multivariable logistic regression analysis was performed to determine which factors were associated with case introductions (i.e., primary case) being linked to a cluster. More cases and clusters were identified in lower than in higher class levels. Out of 21,497 cases introduced into a school, 41.6% started a cluster. A higher assumed immunity level in a class level was significantly reducing the odds of a case introduction being linked to a cluster (e.g., assumed immunity of ≥80% vs <20%: OR: 0.28; 95%CI: 0.17–0.44). A previous infection (in the primary case) had a protective effect (OR: 0.58; 95%CI: 0.33–0.99). This study suggests that most cases appearing in schools did not induce clusters, but that once cluster occur sizes can be large. It further indicates that vaccination of children markedly reduces the risk of secondary infections.

The aim of this cross-sectional study was to identify post-COVID-19 condition (PCC) phenotypes and to investigate the health-related quality of life (HRQoL) and healthcare use per phenotype. We administered a questionnaire to a cohort of PCC patients that included items on socio-demographics, medical characteristics, health symptoms, healthcare use, and the EQ-5D-5L. A principal component analysis (PCA) of PCC symptoms was performed to identify symptom patterns. K-means clustering was used to identify phenotypes. In total, 8630 participants completed the survey. The median number of symptoms was 18, with the top 3 being fatigue, concentration problems, and decreased physical condition. Eight symptom patterns and three phenotypes were identified. Phenotype 1 comprised participants with a lower-than-average number of symptoms, phenotype 2 with an average number of symptoms, and phenotype 3 with a higher-than-average number of symptoms. Compared to participants in phenotypes 1 and 2, those in phenotype 3 consulted significantly more healthcare providers (median 4, 6, and 7, respectively, p < 0.001) and had a significantly worse HRQoL (p < 0.001). In conclusion, number of symptoms rather than type of symptom was the driver in the identification of PCC phenotypes. Experiencing a higher number of symptoms is associated with a lower HRQoL and more healthcare use.

HIV-1 molecular surveillance provides a new approach to explore transmission risks and targeted interventions. From January to June 2021, 663 newly reported HIV-1 cases were recruited in Zhaotong City, Yunnan Province, China. The distribution characteristics of HIV-1 subtypes and HIV-1 molecular network were analysed. Of 542 successfully subtyped samples, 12 HIV-1 strains were identified. The main strains were CRF08_BC (47.0%, 255/542), CRF01_AE (17.0%, 92/542), CRF07_BC (17.0%, 92/542), URFs (8.7%, 47/542), and CRF85_BC (6.5%, 35/542). CRF08_BC was commonly detected among Zhaotong natives, illiterates, and non-farmers and was mostly detected in Zhaoyang County. CRF01_AE was frequently detected among married and homosexual individuals and mostly detected in Weixin and Zhenxiong counties. Among the 516 pol sequences, 187 (36.2%) were clustered. Zhaotong natives, individuals aged ≥60 years, and illiterate individuals were more likely to be found in the network. Assortativity analysis showed that individuals were more likely to be genetically associated when stratified by age, education level, occupation, and reporting area. The genetic diversity of HIV-1 reflects the complexity of local HIV epidemics. Molecular network analyses revealed the subpopulations to focus on and the characteristics of the risk networks. The results will help optimise local prevention and control strategies.

In pricing insurance contracts based on the individual policyholder’s aggregate losses for non-life insurers, the literature has mainly focused on using detailed information from policies and closed claims. However, the information on open claims can reflect shifts in the distribution of the expected claim payments better than closed claims. Such shifts may be needed to be reflected in the ratemaking process earlier rather than later, especially when insurers are experiencing environmental changes. In practice, actuaries use ad hoc techniques to adjust data to current levels to determine premiums. This paper presents an intuitive ratemaking model, employing a marked Poisson process framework, which ensures that the multivariate risk analysis is done more routinely using all reported claims and makes an adjustment for Incurred But Not Reported claims. Utilizing data from the Wisconsin Local Government Property Insurance Fund, we find that by determining rates based on current data, the proposed ratemaking model leads to better alignment of premiums and provides insurers with a more financially sound portfolio.

The use of artificial intelligence and algorithmic decision-making in public policy processes is influenced by a range of diverse drivers. This article provides a comprehensive view of 13 drivers and their interrelationships, identified through empirical findings from the taxation and social security domains in Belgium. These drivers are organized into five hierarchical layers that policy designers need to focus on when introducing advanced analytics in fraud detection: (a) trust layer, (b) interoperability layer, (c) perceived benefits layer, (d) data governance layer, and (e) digital governance layer. The layered approach enables a holistic view of assessing adoption challenges concerning new digital technologies. The research uses thematic analysis and interpretive structural modeling.

Bribery for access to public goods and services remains a widespread and seemingly innocuous practice which disproportionately targets the poor and helps keep them poor. Furthermore, its aggregate effects erode the legitimacy of government institutions and their capacity to fairly administer public goods and services as well as protection under the law. Drawing on original evidence using social norms methodology, this research tests underlying beliefs and expectations which sustain persistent forms of bribery and draws attention to the presence of pluralistic ignorance and consequent collective action problems. With examples focused on bribery in traffic law enforcement, healthcare, and education—three critical areas where bribery is often identified as an entrenched practice—this article contributes new evidence of: (a) the presence of pluralistic ignorance, a common social comparison error, surrounding bribery behavior; (b) differing social evaluations of bribe-solicitation; and finally, (c) how this context might exacerbate collective action problems. This empirical case study of Nigeria shows that even though more people are likely to be directly affected by bribery during routine interactions with public officials and institutions and many believe this practice is wrong, most people incorrectly believe that others in their community tolerate or even accept bribery behavior.

The least squares Monte Carlo method has become a standard approach in the insurance and financial industries for evaluating a company’s exposure to market risk. However, the non-linear regression of simulated responses on risk factors poses a challenge in this procedure. This article presents a novel approach to address this issue by employing an a-priori segmentation of responses. Using a K-means algorithm, we identify clusters of responses that are then locally regressed on their corresponding risk factors. The global regression function is obtained by combining the local models with logistic regression. We demonstrate the effectiveness of the proposed local least squares Monte Carlo method through two case studies. The first case study investigates butterfly and bull trap options within a Heston stochastic volatility model, while the second case study examines the exposure to risks in a participating life insurance scenario.

The Minkowski functionals, including the Euler characteristic statistics, are standard tools for morphological analysis in cosmology. Motivated by cosmic research, we examine the Minkowski functional of the excursion set for an isotropic central limit random field, whose k-point correlation functions (kth-order cumulants) have the same structure as that assumed in cosmic research. Using 3- and 4-point correlation functions, we derive the asymptotic expansions of the Euler characteristic density, which is the building block of the Minkowski functional. The resulting formula reveals the types of non-Gaussianity that cannot be captured by the Minkowski functionals. As an example, we consider an isotropic chi-squared random field and confirm that the asymptotic expansion accurately approximates the true Euler characteristic density.

Salmonella spp. is a common zoonotic pathogen, causing gastrointestinal infections in people. Pigs and pig meat are a major source of infection. Although farm biosecurity is believed to be important for controlling Salmonella transmission, robust evidence is lacking on which measures are most effective. This study enrolled 250 pig farms across nine European countries. From each farm, 20 pooled faecal samples (or similar information) were collected and analysed for Salmonella presence. Based on the proportion of positive results, farms were categorised as at higher or lower Salmonella risk, and associations with variables from a comprehensive questionnaire investigated. Multivariable analysis indicated that farms were less likely to be in the higher-risk category if they had ‘<400 sows’; used rodent baits close to pig enclosures; isolated stay-behind (sick) pigs; did not answer that the hygiene lock/ anteroom was easy to clean; did not have a full perimeter fence; did apply downtime of at least 3 days between farrowing batches; and had fully slatted flooring in all fattener buildings. A principal components analysis assessed the sources of variation between farms, and correlation between variables. The study results suggest simple control measures that could be prioritised on European pig farms to control Salmonella.

This study aimed to assess the prevalence of anti-hepatitis E virus (HEV) immunoglobulin (Ig) M and elevated serum alanine aminotransferase (ALT) levels among employees in catering and public place industries. Blood samples were collected between January and December 2020 from 26,790 employees working in the Qinhuai district of Nanjing, China. Anti-HEV IgM in the serum samples was tested by the capture ELISA method and ALT was tested by the IFCC method. Samples positive for anti-HEV IgM or with ALT levels over 200 U/L were subjected to PCR screening of HEV RNA. The overall seroprevalence of anti-HEV IgM was 0.41%, and the seroprevalence was slightly higher in males (0.47%) than in females (0.37%); however, the difference was not substantial (p = 0.177). Seroprevalence of anti-HEV IgM increased with age, reaching its peak level after 48 years of age. The prevalence of elevated ALT levels was 4.24%, and males exhibited a higher prevalence than females (6.78% vs 2.65%, p < 0.001). Prevalence of elevated ALT levels differed in age groups and the 26–36-year-old group had the highest rate of elevated ALT levels. Employees with elevated ALT levels had a higher prevalence of positive anti-HEV IgM than those with normal ALT (0.57% vs 0.31%, p < 0.001). Positive HEV RNA was detected in one anti-HEV IgM-negative employee with ALT higher than 200 U/L. In our study, all the HEV RNA-positive and IgM-positive individuals are asymptomatic, and a combination of ALT tests, serological methods, and molecular methods is recommended to screen asymptomatic HEV carriers and reduce the risk of transmission.

Data-informed predictive maintenance planning largely relies on stochastic deterioration models. Monitoring information can be utilized to update sequentially the knowledge on model parameters. In this context, on-line (recursive) Bayesian filtering algorithms typically fail to properly quantify the full posterior uncertainty of time-invariant model parameters. Off-line (batch) algorithms are—in principle—better suited for the uncertainty quantification task, yet they are computationally prohibitive in sequential settings. In this work, we adapt and investigate selected Bayesian filters for parameter estimation: an on-line particle filter, an on-line iterated batch importance sampling filter, which performs Markov Chain Monte Carlo (MCMC) move steps, and an off-line MCMC-based sequential Monte Carlo filter. A Gaussian mixture model approximates the posterior distribution within the resampling process in all three filters. Two numerical examples provide the basis for a comparative assessment. The first example considers a low-dimensional, nonlinear, non-Gaussian probabilistic fatigue crack growth model that is updated with sequential monitoring measurements. The second high-dimensional, linear, Gaussian example employs a random field to model corrosion deterioration across a beam, which is updated with sequential sensor measurements. The numerical investigations provide insights into the performance of off-line and on-line filters in terms of the accuracy of posterior estimates and the computational cost, when applied to problems of different nature, increasing dimensionality and varying sensor information amount. Importantly, they show that a tailored implementation of the on-line particle filter proves competitive with the computationally demanding MCMC-based filters. Suggestions on the choice of the appropriate method in function of problem characteristics are provided.

In 2022, a case of paralysis was reported in an unvaccinated adult in Rockland County (RC), New York. Genetically linked detections of vaccine-derived poliovirus type 2 (VDPV2) were reported in multiple New York counties, England, Israel, and Canada. The aims of this qualitative study were to: i) review immediate public health responses in New York to assess the challenges in addressing gaps in vaccination coverage; ii) inform a longer-term strategy to improving vaccination coverage in under-vaccinated communities, and iii) collect data to support comparative evaluations of transnational poliovirus outbreaks. Twenty-three semi-structured interviews were conducted with public health professionals, healthcare professionals, and community partners. Results indicate that i) addressing suboptimal vaccination coverage in RC remains a significant challenge after recent disease outbreaks; ii) the poliovirus outbreak was not unexpected and effort should be invested to engage mothers, the key decision-makers on childhood vaccination; iii) healthcare providers (especially paediatricians) received technical support during the outbreak, and may require resources and guidance to effectively contribute to longer-term vaccine engagement strategies; vi) data systems strengthening is required to help track under-vaccinated children. Public health departments should prioritize long-term investments in appropriate communication strategies, countering misinformation, and promoting the importance of the routine immunization schedule.

This paper proposes a novel method of algorithmic subsampling (data sketching) for multiway cluster-dependent data. We establish a new uniform weak law of large numbers and a new central limit theorem for multiway algorithmic subsample means. We show that algorithmic subsampling allows for robustness against potential degeneracy, and even non-Gaussian degeneracy, of the asymptotic distribution under multiway clustering at the cost of efficiency and power loss due to algorithmic subsampling. Simulation studies support this novel result, and demonstrate that inference with algorithmic subsampling entails more accuracy than that without algorithmic subsampling. We derive the consistency and the asymptotic normality for multiway algorithmic subsampling generalized method of moments estimator and for multiway algorithmic subsampling M-estimator. We illustrate with an application to scanner data for the analysis of differentiated products markets.

Insurers and pension funds face the challenges of historically low-interest rates and high volatility in equity markets, that have been accentuated due to the COVID-19 pandemic. Recent advances in equity portfolio management with a target volatility have been shown to deliver improved on average risk-adjusted return, after transaction costs. This paper studies these targeted volatility portfolios in applications to equity, balanced, and target-date funds with varying constraints on leverage. Conservative leverage constraints are particularly relevant to pension funds and insurance companies, with more aggressive leverage levels appropriate for alternative investments. We show substantial improvements in fund performance for differing leverage levels, and of most interest to insurers and pension funds, we show that the highest Sharpe ratios and smallest drawdowns are in targeted volatility-balanced portfolios with equity and bond allocations. Furthermore, we demonstrate the outperformance of targeted volatility portfolios during major stock market crashes, including the crash from the COVID-19 pandemic.

This paper studies dynamic reinsurance contracting and competition problems under model ambiguity in a reinsurance market with one primary insurer and n reinsurers, who apply the variance premium principle and who are distinguished by their levels of ambiguity aversion. The insurer negotiates reinsurance policies with all reinsurers simultaneously, which leads to a reinsurance tree structure with full competition among the reinsurers. We model the reinsurance contracting problems between the insurer and reinsurers by Stackelberg differential games and the competition among the reinsurers by a non-cooperative Nash game. We derive equilibrium strategies in semi-closed form for all the companies, whose objective is to maximize their expected surpluses penalized by a squared-error divergence term that measures their ambiguity. We find that, in equilibrium, the insurer purchases a positive amount of proportional reinsurance from each reinsurer. We further show that the insurer always prefers the tree structure to the chain structure, in which the risk of the insurer is shared sequentially among all reinsurers.