Chronic Wasting Disease Surveillance Optimization Software

Abstract

The Chronic Wasting Disease Surveillance Optimization Software computes sampling strategy recommendations for state-wise use of surveillance funds by a state wildlife management agency when the goal of the disease surveillance program is to detect chronic wasting disease (CWD) in white-tailed deer (Odocoileus virginianus). Driven by a combinatorial optimization algorithm, the Chronic Wasting Disease Surveillance Optimization Software pinpoints the combination of samples that should be tested in each population segment of deer (adult male, adult female, yearling male, yearling female, fawn male, fawn female), in each collection category of deer (hunter harvested, non-hunter harvested), and in each county (or other administrative area) of the state to maximize the return-on-investment, while keeping the overall surveillance program within the predetermined budget. Inputs to the combinatorial optimization algorithm in the Chronic Wasting Disease Surveillance Optimization Software broadly include the Optimization Matrix and the Historical Data. The Optimization Matrix includes information such as the costs, benefits, and starting prevalence of each population segment, collection category, and county or other administrative unit. The Historical Data considers the previously enacted sampling strategy in each county (or other administrative area), then uses that data as comparison to the algorithmic recommendations. Agencies may further parameterize their algorithm to achieve specific management objectives. Objectives include monitoring known infections, searching for new infections, and providing detailed information to the public. The Chronic Wasting Disease Surveillance Optimization Software includes (redacted) Optimization Matrices and (redacted) Historical Data from Alabama, Arkansas, Connecticut, Florida, Georgia, Indiana, Iowa, Kentucky, Louisiana, Maryland, Michigan, Minnesota, Mississippi, New Hampshire, New York, North Carolina, Ohio, Pennsylvania, Rhode Island, Tennessee, Virginia, and Wisconsin, US, but the software may be adapted for use in other states and provinces.