Evaluation of the Johne's disease surveillance program in dairy herds in Hokkaido, Japan, using a stochastic individual-based simulation model.

Johne's disease (JD), caused by Mycobacterium avium subsp. paratuberculosis (MAP), imposes major economic burdens on dairy production. Japan's legally mandated surveillance includes regular surveillance (RS) every 5 years in MAP-negative herds and follow-up surveillance (FS) in MAP-positive herds, consisting of 5 tests over 3 years. However, the effectiveness and cost-effectiveness of this program have not been quantitatively assessed. We developed a stochastic, individual-based simulation model to evaluate MAP transmission, farm-level economic impacts, and government costs in dairy herds in Hokkaido, Japan's largest dairy-producing region. Eight scenarios were compared: no surveillance, the current program, and variations in RS interval, FS duration, and FS testing frequency. Results indicated that, compared with no surveillance, the current program slowed the increase in within-herd prevalence and increased the probability of MAP fade-out, thereby reducing economic losses. Shortening the RS interval to 3 years enhanced early detection and more efficiently achieved MAP fade-out, which in turn reduced government costs. Compared with the current program, shortening FS duration or reducing testing frequency lowered detection efficiency of the surveillance program and increased MAP persistence, while extending duration or increasing frequency provided little additional benefit. Overall, the current JD surveillance program in Hokkaido effectively promotes MAP fade-out within herds through early detection by RS combined with repeated high-frequency FS testing, although a shorter RS interval may further strengthen early detection and JD control.