Effective screening strategies for safe opening of universities under Omicron and Delta variants of COVID-19.

As new COVID-19 variants emerge, and disease and population characteristics change, screening strategies may also need to change. We develop a decision-making model that can assist a college to determine an optimal screening strategy based on their characteristics and resources, considering COVID-19 infections/hospitalizations/deaths; peak daily hospitalizations; and the tests required. We also use this tool to generate screening guidelines for the safe opening of college campuses. Our compartmental model simulates disease spread on a hypothetical college campus under co-circulating variants with different disease dynamics, considering: (i) the heterogeneity in disease transmission and outcomes for faculty/staff and students based on vaccination status and level of natural immunity; and (ii) variant- and dose-dependent vaccine efficacy. Using the Spring 2022 academic semester as a case study, we study routine screening strategies, and find that screening the faculty/staff less frequently than the students, and/or the boosted and vaccinated less frequently than the unvaccinated, may avert a higher number of infections per test, compared to universal screening of the entire population at a common frequency. We also discuss key policy issues, including the need to revisit the mitigation objective over time, effective strategies that are informed by booster coverage, and if and when screening alone can compensate for low booster coverage.

A new perspective on breast cancer diagnostic guidelines to reduce overdiagnosis.

Overdiagnosis of breast cancer, defined as diagnosing a cancer that would otherwise not cause symptoms or death in a patient's lifetime, costs U.S. health care system over $1.2 billion annually. Overdiagnosis rates, estimated to be around 10%-40%, may be reduced if indolent breast findings can be identified and followed with noninvasive imaging rather than biopsy. However, there are no validated guidelines for radiologists to decide when to choose imaging options recognizing cancer grades and types. The aim of this study is to optimize breast cancer diagnostic decisions based on cancer types using a large-scale finite-horizon Markov decision process (MDP) model with 4.6 million states to help reduce overdiagnosis. We prove the optimality of a divide-and-search algorithm that relies on tight upper bounds on the optimal decision thresholds to find an exact optimal solution. We project the high-dimensional MDP onto two lower dimensional MDPs and obtain feasible upper bounds on the optimal decision thresholds. We use real data from two private mammography databases and demonstrate our model performance through a previously validated simulation model that has been used by the policy makers to set the national screening guidelines in the United States. We find that a decision-analytical framework optimizing diagnostic decisions while accounting for breast cancer types has a strong potential to improve the quality of life and alleviate the immense costs of overdiagnosis. Our model leads to a 20 % reduction in overdiagnosis on the screening population, which translates into an annual savings of approximately $300 million for the U.S. health care system.

Benefits of integrated screening and vaccination for infection control.

IMPORTANCE: Screening and vaccination are essential in the fight against infectious diseases, but need to be integrated and customized based on community and disease characteristics. OBJECTIVE: To develop effective screening and vaccination strategies, customized for a college campus, to reduce COVID-19 infections, hospitalizations, deaths, and peak hospitalizations. DESIGN, SETTING, AND PARTICIPANTS: We construct a compartmental model of disease spread under vaccination and routine screening, and study the efficacy of four mitigation strategies (routine screening only, vaccination only, vaccination with partial or full routine screening), and a no-intervention strategy. The study setting is a hypothetical college campus of 5,000 students and 455 faculty members during the Fall 2021 academic semester, when the Delta variant was the predominant strain. For sensitivity analysis, we vary the screening frequency, daily vaccination rate, initial vaccine coverage, and screening and vaccination compliance; and consider scenarios that represent low/medium/high transmission and test efficacy. Model parameters come from publicly available or published sources. RESULTS: With low initial vaccine coverage (30% in our study), even aggressive vaccination and screening result in a high number of infections: 1,020 to 2,040 (1,530 to 2,480) with routine daily (every other day) screening of the unvaccinated; 280 to 900 with daily screening extended to the newly vaccinated in base- and worst-case scenarios, which respectively consider reproduction numbers of 4.75 and 6.75 for the Delta variant. CONCLUSION: Integrated vaccination and routine screening can allow for a safe opening of a college when both the vaccine effectiveness and the initial vaccine coverage are sufficiently high. The interventions need to be customized considering the initial vaccine coverage, estimated compliance, screening and vaccination capacity, disease transmission and adverse outcome rates, and the number of infections/peak hospitalizations the college is willing to tolerate.

A Deterministic Analysis of an Online Convex Mixture of Experts Algorithm.

We analyze an online learning algorithm that adaptively combines outputs of two constituent algorithms (or the experts) running in parallel to estimate an unknown desired signal. This online learning algorithm is shown to achieve and in some cases outperform the mean-square error (MSE) performance of the best constituent algorithm in the steady state. However, the MSE analysis of this algorithm in the literature uses approximations and relies on statistical models on the underlying signals. Hence, such an analysis may not be useful or valid for signals generated by various real-life systems that show high degrees of nonstationarity, limit cycles and that are even chaotic in many cases. In this brief, we produce results in an individual sequence manner. In particular, we relate the time-accumulated squared estimation error of this online algorithm at any time over any interval to the one of the optimal convex mixture of the constituent algorithms directly tuned to the underlying signal in a deterministic sense without any statistical assumptions. In this sense, our analysis provides the transient, steady-state, and tracking behavior of this algorithm in a strong sense without any approximations in the derivations or statistical assumptions on the underlying signals such that our results are guaranteed to hold. We illustrate the introduced results through examples.