Cost-effectiveness of diagnostic strategies using quantitative real-time PCR and bacterial culture to identify contagious mastitis cases in large dairy herds.

Diagnostic strategies to detect contagious mastitis caused by Mycoplasma bovis, Staphylococcus aureus, and Streptococcus agalactiae in dairy herds during an outbreak have been minimally studied with regard to cost and diagnostic sensitivity. The objective of this cross-sectional study was to compare the cost-effectiveness of diagnostic strategies for identification of infected cows in two California dairy herds during contagious mastitis outbreaks. M. bovis was investigated in a subset of a herd (n=1210 cows) with an estimated prevalence of 2.8% (95% CI=1.9, 3.7), whereas Staph. aureus and Strep. agalactiae were studied in a second herd (n=351 cows) with an estimated prevalence of 3.4% (95% CI=1.5, 5.3) and 16.8% (95% CI=12.9, 20.7), respectively. Diagnostic strategies involved a combination of testing stages that utilized bacterial culture, quantitative real-time PCR (qPCR), or both. Strategies were applied to individual or pooled samples of 5, 10, 50 or 100 samples. Culture was considered the gold standard for sensitivity estimation of each strategy. The reference strategy was the strategy with the lowest cost per culture-positive cow which for both M. bovis and Strep. agalactiae consisted of 2 stages, culture of samples in pools of 5 followed by culture of individual samples in positive pools with a sensitivity of 73.5% (95% CI: 55.6, 87.1) and 96.6% (95% CI: 27.7, 84.8), respectively. The reference strategy for Staph. aureus consisted of 3 stages, culture of individual samples in pools of 100 (stage 1), culture constituents of those positive from stage 1 in pools of 5 (stage 2), culture constituents of those positive from stage 2 individually (stage 3) which resulted in a sensitivity of 58.3% (95% CI: 88.3, 99.6). The most cost-effective alternative to the reference strategy was whole herd milk culture for all 3 pathogens. QPCR testing was a component of the second most cost-effective alternative for M. bovis and the third most cost-effective alternatives for the 3 pathogens. A stochastic model was used to assess the effect of prevalence or herd size on the cost-effectiveness of diagnostic strategies. In the current study, increasing the prevalence of mastitis did not alter the ranking of strategies by cost-effectiveness. However, larger herds could benefit from testing larger pools such as 50 or 100 samples to improve cost-effectiveness. Several diagnostic strategy options exist to identify contagious mastitis in herds, decisions should be based on cost and sensitivity of the strategies available.