Limits...
Managerial and environmental determinants of clinical mastitis in Danish dairy herds.

Sato K, Bartlett PC, Alban L, Agger JF, Houe H - Acta Vet. Scand. (2008)

Bottom Line: 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.Our factor analysis identified one significant latent factor, which was related to labor quality on the farm.

View Article: PubMed Central - HTML - PubMed

Affiliation: Department of Veterinary Sciences, College of Agriculture, University of Wyoming, Laramie, WY, USA. bartlett@cvm.msu.edu

ABSTRACT

Background: 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.

Methods: 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.

Results: 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.

Conclusion: 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.

Show MeSH

Related in: MedlinePlus

Scree plot of Eigenvalues for 14 variables. Label on X-axis: Variable number, Label on Y-axis: Eigenvalue, Legends: Numbers inside the figure indicate the variable number.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC2268923&req=5

Figure 2: Scree plot of Eigenvalues for 14 variables. Label on X-axis: Variable number, Label on Y-axis: Eigenvalue, Legends: Numbers inside the figure indicate the variable number.

Mentions: Fourteen variables were found to have p-values < 0.10 in the initial bivariate analysis, and these variables were subjected to an exploratory factor analysis. The highest correlation was 0.61 between "who takes care of the cows?" (PASSER_1) and "hired labor used in cow house?" (FHJLP_73). The second and third highest squared correlations were 0.51 and 0.43 between PASSER_1 and TILSYN32, and between TILSYN32 and FHJLP_73 respectively. All other squared correlations were below 0.35. The eigenvalues of the correlation matrix (not shown in this report) showed that the percentage of common variance accounted for by factor 1, 2, and 3 were 16%, 10% and 10%, respectively. The factors 4 to 14 each accounted for less than 10% of the common variance. If we choose the number of latent factors based on eigenvalues greater than 1 (the Kaiser criterion [15]), there would be 6 latent factors for this data. The scree plot of the eigenvalues (Figure 2) shows a sizable gap between the factors with relatively large eigenvalues (factor 1–3) and those with smaller eigenvalues (factor 4–14). Therefore, three factors were contained in the model.


Managerial and environmental determinants of clinical mastitis in Danish dairy herds.

Sato K, Bartlett PC, Alban L, Agger JF, Houe H - Acta Vet. Scand. (2008)

Scree plot of Eigenvalues for 14 variables. Label on X-axis: Variable number, Label on Y-axis: Eigenvalue, Legends: Numbers inside the figure indicate the variable number.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC2268923&req=5

Figure 2: Scree plot of Eigenvalues for 14 variables. Label on X-axis: Variable number, Label on Y-axis: Eigenvalue, Legends: Numbers inside the figure indicate the variable number.
Mentions: Fourteen variables were found to have p-values < 0.10 in the initial bivariate analysis, and these variables were subjected to an exploratory factor analysis. The highest correlation was 0.61 between "who takes care of the cows?" (PASSER_1) and "hired labor used in cow house?" (FHJLP_73). The second and third highest squared correlations were 0.51 and 0.43 between PASSER_1 and TILSYN32, and between TILSYN32 and FHJLP_73 respectively. All other squared correlations were below 0.35. The eigenvalues of the correlation matrix (not shown in this report) showed that the percentage of common variance accounted for by factor 1, 2, and 3 were 16%, 10% and 10%, respectively. The factors 4 to 14 each accounted for less than 10% of the common variance. If we choose the number of latent factors based on eigenvalues greater than 1 (the Kaiser criterion [15]), there would be 6 latent factors for this data. The scree plot of the eigenvalues (Figure 2) shows a sizable gap between the factors with relatively large eigenvalues (factor 1–3) and those with smaller eigenvalues (factor 4–14). Therefore, three factors were contained in the model.

Bottom Line: 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.Our factor analysis identified one significant latent factor, which was related to labor quality on the farm.

View Article: PubMed Central - HTML - PubMed

Affiliation: Department of Veterinary Sciences, College of Agriculture, University of Wyoming, Laramie, WY, USA. bartlett@cvm.msu.edu

ABSTRACT

Background: 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.

Methods: 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.

Results: 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.

Conclusion: 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.

Show MeSH
Related in: MedlinePlus