Sato Kenji, Bartlett Paul C, Alban Lis, Agger Jens F, Houe Hans
Department of Veterinary Sciences, College of Agriculture, University of Wyoming, Laramie, WY, USA.
Acta Vet Scand. 2008 Feb 7;50(1):4. doi: 10.1186/1751-0147-50-4.
Several management and environmental factors are known as contributory causes of clinical mastitis in dairy herd. The study objectives were to describe the structure of herd-specific mastitis management and environmental factors and to assess the relevance of these herd-specific indicators to mastitis incidence rate.
Disease reports from the Danish Cattle Data Base and a management questionnaire from 2,146 herds in three Danish regions were analyzed to identify and characterize risk factors of clinical mastitis. A total of 94 (18 continuous and 76 discrete) management and production variables were screened in separate bivariate regression models. Variables associated with mastitis incidence rate at a p-value < 0.10 were examined with a factor analysis to assess the construct of data. Separately, a multivariable regression model was used to estimate the association of management variables with herd mastitis rate.
Three latent factors (quality of labor, region of Denmark and claw trimming, and quality of outdoor holding area) were identified from 14 variables. Daily milk production per cow, claw disease, quality of labor and region of Denmark were found to be significantly associated with mastitis incidence rate. A common multiple regression analysis with backward and forward selection procedures indicated there were 9 herd-specific risk factors.
Though risk factors ascertained by farmer-completed surveys explained a small percentage of the among-herd variability in crude herd-specific mastitis rates, the study suggested that farmer attitudes toward mastitis and lameness treatment were important determinants for mastitis incidence rate. Our factor analysis identified one significant latent factor, which was related to labor quality on the farm.
多种管理和环境因素被认为是奶牛场临床乳腺炎的促成原因。本研究的目的是描述特定牛群乳腺炎管理和环境因素的结构，并评估这些特定牛群指标与乳腺炎发病率的相关性。
分析了来自丹麦牛数据库的疾病报告以及来自丹麦三个地区2146个牛群的管理调查问卷，以识别和表征临床乳腺炎的风险因素。在单独的双变量回归模型中筛选了总共94个（18个连续变量和76个离散变量）管理和生产变量。对p值<0.10时与乳腺炎发病率相关的变量进行因子分析，以评估数据结构。另外，使用多变量回归模型来估计管理变量与牛群乳腺炎发病率之间的关联。
从14个变量中识别出三个潜在因素（劳动力质量、丹麦地区和蹄部修剪以及户外饲养区质量）。发现每头奶牛的日产奶量、蹄部疾病、劳动力质量和丹麦地区与乳腺炎发病率显著相关。采用向后和向前选择程序的普通多元回归分析表明存在9个特定牛群的风险因素。
尽管通过农民完成的调查确定的风险因素在特定牛群乳腺炎粗发病率的牛群间变异性中所占比例较小，但该研究表明农民对乳腺炎和跛行治疗的态度是乳腺炎发病率的重要决定因素。我们的因子分析确定了一个重要的潜在因素，它与农场的劳动力质量有关。
