Receiver-operating characteristic curves for somatic cell scores and California mastitis test in Valle del Belice dairy sheep

Baldassare Portolano, Valentina Riggio, Lorenzo L. Pesce

Risultato della ricerca: Article

7 Citazioni (Scopus)

Abstract

Using receiver-operating characteristic (ROC) curve methodology this study was designed to assess the diagnostic effectiveness of somatic cell count (SCC) and the California mastitis test (CMT) in Valle del Belice sheep, and to propose and evaluate threshold values for those tests that would optimally discriminate between healthy and infected udders. Milk samples (n = 1357) were collected from 684 sheep in four flocks. The prevalence of infection, as determined by positive bacterial culture was 0.36, 87.7% of which were minor and 12.3% major pathogens. Of the culture negative samples, 83.7% had an SCC < 500,000/mL and 97.4% had <1,000,000 cells/mL. When the associations between SC score (SCS) and whole sample status (culture negative vs. infected), minor pathogen status (culture negative vs. infected with minor pathogens), major pathogen status (culture negative vs. infected with major pathogens), and CMT results were evaluated, the estimated area under the ROC curve was greater for glands infected with major compared to minor pathogens (0.88 vs. 0.73), whereas the area under the curve considering all pathogens was similar to the one for minor pathogens (0.75). The estimated optimal thresholds were 3.00 (CMT), 2.81 (SCS for the whole sample), 2.81 (SCS for minor pathogens), and 3.33 (SCS for major pathogens). These correctly classified, respectively, 69.0%, 73.5%, 72.6% and 91.0% of infected udders in the samples. The CMT appeared only to discriminate udders infected with major pathogens. In this population, SCS appeared to be the best indirect test of the bacteriological status of the udder.
Lingua originaleEnglish
pagine (da-a)528-532
Numero di pagine5
RivistaTHE VETERINARY JOURNAL
Volume196
Stato di pubblicazionePublished - 2013

Fingerprint

Animal Mammary Glands
dairy sheep
Mastitis
somatic cells
ROC Curve
Sheep
pathogens
Cell Count
udders
Area Under Curve
Milk
somatic cell count
sampling
California mastitis test
Infection
sheep
Population
flocks
testing

All Science Journal Classification (ASJC) codes

  • Animal Science and Zoology
  • veterinary(all)

Cita questo

@article{0d029dcc3cbf4e9685bd6c9516eaf19e,
title = "Receiver-operating characteristic curves for somatic cell scores and California mastitis test in Valle del Belice dairy sheep",
abstract = "Using receiver-operating characteristic (ROC) curve methodology this study was designed to assess the diagnostic effectiveness of somatic cell count (SCC) and the California mastitis test (CMT) in Valle del Belice sheep, and to propose and evaluate threshold values for those tests that would optimally discriminate between healthy and infected udders. Milk samples (n = 1357) were collected from 684 sheep in four flocks. The prevalence of infection, as determined by positive bacterial culture was 0.36, 87.7{\%} of which were minor and 12.3{\%} major pathogens. Of the culture negative samples, 83.7{\%} had an SCC < 500,000/mL and 97.4{\%} had <1,000,000 cells/mL. When the associations between SC score (SCS) and whole sample status (culture negative vs. infected), minor pathogen status (culture negative vs. infected with minor pathogens), major pathogen status (culture negative vs. infected with major pathogens), and CMT results were evaluated, the estimated area under the ROC curve was greater for glands infected with major compared to minor pathogens (0.88 vs. 0.73), whereas the area under the curve considering all pathogens was similar to the one for minor pathogens (0.75). The estimated optimal thresholds were 3.00 (CMT), 2.81 (SCS for the whole sample), 2.81 (SCS for minor pathogens), and 3.33 (SCS for major pathogens). These correctly classified, respectively, 69.0{\%}, 73.5{\%}, 72.6{\%} and 91.0{\%} of infected udders in the samples. The CMT appeared only to discriminate udders infected with major pathogens. In this population, SCS appeared to be the best indirect test of the bacteriological status of the udder.",
author = "Baldassare Portolano and Valentina Riggio and Pesce, {Lorenzo L.}",
year = "2013",
language = "English",
volume = "196",
pages = "528--532",
journal = "Veterinary Journal",
issn = "1090-0233",
publisher = "Bailliere Tindall Ltd",

}

TY - JOUR

T1 - Receiver-operating characteristic curves for somatic cell scores and California mastitis test in Valle del Belice dairy sheep

AU - Portolano, Baldassare

AU - Riggio, Valentina

AU - Pesce, Lorenzo L.

PY - 2013

Y1 - 2013

N2 - Using receiver-operating characteristic (ROC) curve methodology this study was designed to assess the diagnostic effectiveness of somatic cell count (SCC) and the California mastitis test (CMT) in Valle del Belice sheep, and to propose and evaluate threshold values for those tests that would optimally discriminate between healthy and infected udders. Milk samples (n = 1357) were collected from 684 sheep in four flocks. The prevalence of infection, as determined by positive bacterial culture was 0.36, 87.7% of which were minor and 12.3% major pathogens. Of the culture negative samples, 83.7% had an SCC < 500,000/mL and 97.4% had <1,000,000 cells/mL. When the associations between SC score (SCS) and whole sample status (culture negative vs. infected), minor pathogen status (culture negative vs. infected with minor pathogens), major pathogen status (culture negative vs. infected with major pathogens), and CMT results were evaluated, the estimated area under the ROC curve was greater for glands infected with major compared to minor pathogens (0.88 vs. 0.73), whereas the area under the curve considering all pathogens was similar to the one for minor pathogens (0.75). The estimated optimal thresholds were 3.00 (CMT), 2.81 (SCS for the whole sample), 2.81 (SCS for minor pathogens), and 3.33 (SCS for major pathogens). These correctly classified, respectively, 69.0%, 73.5%, 72.6% and 91.0% of infected udders in the samples. The CMT appeared only to discriminate udders infected with major pathogens. In this population, SCS appeared to be the best indirect test of the bacteriological status of the udder.

AB - Using receiver-operating characteristic (ROC) curve methodology this study was designed to assess the diagnostic effectiveness of somatic cell count (SCC) and the California mastitis test (CMT) in Valle del Belice sheep, and to propose and evaluate threshold values for those tests that would optimally discriminate between healthy and infected udders. Milk samples (n = 1357) were collected from 684 sheep in four flocks. The prevalence of infection, as determined by positive bacterial culture was 0.36, 87.7% of which were minor and 12.3% major pathogens. Of the culture negative samples, 83.7% had an SCC < 500,000/mL and 97.4% had <1,000,000 cells/mL. When the associations between SC score (SCS) and whole sample status (culture negative vs. infected), minor pathogen status (culture negative vs. infected with minor pathogens), major pathogen status (culture negative vs. infected with major pathogens), and CMT results were evaluated, the estimated area under the ROC curve was greater for glands infected with major compared to minor pathogens (0.88 vs. 0.73), whereas the area under the curve considering all pathogens was similar to the one for minor pathogens (0.75). The estimated optimal thresholds were 3.00 (CMT), 2.81 (SCS for the whole sample), 2.81 (SCS for minor pathogens), and 3.33 (SCS for major pathogens). These correctly classified, respectively, 69.0%, 73.5%, 72.6% and 91.0% of infected udders in the samples. The CMT appeared only to discriminate udders infected with major pathogens. In this population, SCS appeared to be the best indirect test of the bacteriological status of the udder.

UR - http://hdl.handle.net/10447/104234

M3 - Article

VL - 196

SP - 528

EP - 532

JO - Veterinary Journal

JF - Veterinary Journal

SN - 1090-0233

ER -