Use of data mining techniques to investigate disease risk classification as a proxy for compromised biosecurity of cattle herds in Wales
2008

Investigating Disease Risk in Cattle Herds in Wales

Sample size: 15845 publication 10 minutes Evidence: moderate

Author Information

Author(s): Ángel Ortiz-Pelaez, Dirk U. Pfeiffer

Primary Institution: The Royal Veterinary College, University of London

Hypothesis

Can data mining techniques classify cattle herds based on their risk of disease presence as a proxy for compromised biosecurity?

Conclusion

The study shows that data mining can effectively classify cattle herds by their biosecurity risk based on existing data.

Supporting Evidence

  • High-risk holdings are large cattle herds in high-density areas with frequent movements.
  • Data mining can classify herds based on existing data without additional on-farm data collection.
  • The study identified specific risk factors associated with disease presence in cattle.

Takeaway

This study looked at how to tell if cattle farms are at risk for diseases using data that’s already been collected, like where the farms are and how many cows they have.

Methodology

The study used logistic regression, classification trees, and factor analysis to analyze data from cattle holdings.

Potential Biases

There may be selection bias due to reliance on voluntary disease reporting from veterinarians.

Limitations

The study faced issues with missing data and potential biases in disease reporting.

Participant Demographics

Cattle holdings in Wales, including both dairy and beef farms.

Statistical Information

P-Value

<0.001

Confidence Interval

95% CI: 0.046–0.068

Statistical Significance

p<0.001

Digital Object Identifier (DOI)

10.1186/1746-6148-4-24

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication