Risk assessment of subclinical mastitis in Holstein cows in Jiangsu Province of China

Information from the abstract

Introduction Milk somatic cell count (SCC) and somatic cell score (SCS) are good indicators of the occurrence of subclinical mastitis (SCM) in dairy cows. Methods In this study, multi-factor ANOVA model was used to analyze the fixed effects of 6 factors, including farm size, management, parity, calving season, lactation stage, and previous SCS (PSCS), on the current month SCS in milk of dairy cows in a training dataset containing 858,546 dairy herd improvement (DHI) records from 12 dairy farms in Jiangsu Province of China. The Logistic regression models of SCS or SCM were established for each factor and and then valiated with a separate dataset containing 27,575 DHI records from 6 other dairy farms. Results The results showed that the six factors had significant effects on SCS and SCM incidence in dairy cows ( p < 0.01). Among them, milk SCS was significantly higher in the farms with fewer than 2,000 cows and fermented bed breeding, and in the cows having more than 5 parities, calving in summer, in late lactation, and when PSCS was 9. The odds ratios (OR) at different levels of each factor were consistent with the actual incidence of SCM. Among them, SCM incidence and OR of cows were lowest when farms had more than 10,000 cows, free stalls, when cows were in first parity, and during the autumn and mid lactation, and when the PSCS was 0. Prediction accuracy and balanced accuracy of risk of SCM during validation were 87.5 and 56.0%. The multivariable logistic regression model demonstrated fair discriminative ability (AUC = 0.771). However, due to its low sensitivity at the default threshold, the model should not be recommended as a standalone farm-level early warning tool without threshold calibration and external validation. Conclusion The model may be useful for risk stratification (e.g., identifying high-risk subgroups), but prospective validation is needed before field application.

Why this record is monitored

This record has an Impact Signal of 70/100 based on recency, source, collaboration, and bibliographic signals. It prioritizes monitoring and is not a judgment of research quality.

Related topics: Milk Quality and Mastitis in Dairy Cows · Animal Behavior and Welfare Studies · Reproductive Physiology in Livestock