Efficacy of detecting Chronic Wasting Disease via sampling hunter-killed white-tailed deer

Surveillance programs for Chronic Wasting Disease (CWD) in free-ranging cervids often use a standard of being able to detect 1% prevalence when determining minimum sample sizes. However, 1% prevalence may represent >10,000 infected animals in a population of 1 million, and most wildlife managers would prefer to detect the presence of CWD when far fewer infected animals exist. We wanted to detect the presence of CWD in white-tailed deer (Odocoileus virginianus) in Pennsylvania when the disease was present in only 1 of 21 wildlife management units (WMUs) statewide. We used computer simulation to estimate the probability of detecting CWD based on a sampling design to detect the presence of CWD at 0.1% and 1.0% prevalence (23–76 and 225–762 infected deer, respectively) using tissue samples collected from hunter-killed deer. The probability of detection at 0.1% prevalence was <30% with sample sizes of ≤6,000 deer, and the probability of detection at 1.0% prevalence was 46–72% with statewide sample sizes of 2,000–6,000 deer. We believe that testing of hunter-killed deer is an essential part of any surveillance program for CWD, but our results demonstrated the importance of a multifaceted surveillance approach for CWD detection rather than sole reliance on testing hunter-killed deer.