Screening multi‐dimensional heterogeneous populations for infectious diseases under scarce testing resources, with application to <scp>COVID</scp> ‐19

Testing provides essential information for managing infectious disease outbreaks, such as the COVID-19 pandemic. When testing resources are scarce, an important managerial decision is who to test. This compounded by fact that potential subjects heterogeneous in multiple dimensions consider, including their likelihood of being disease-positive, and how much harm would be averted through subsequent interventions. To increase coverage, pooled can utilized, but this comes at a cost increased false-negatives when test imperfect. Then, problem partition population into three mutually exclusive sets: those individually tested, pool not tested. Additionally, tested must further partitioned pools, potentially containing different numbers subjects. The objectives include minimization (through detection mitigation) or maximization coverage. We develop data-driven optimization models algorithms design strategies, show, via contact tracing case study, proposed strategies substantially outperform current practice used (individually contacts with symptoms). Our results demonstrate substantial benefits optimizing design, while considering heterogeneity limited capacity.