Limits...
Summary of current knowledge of the size and spatial distribution of the horse population within Great Britain.

Boden LA, Parkin TD, Yates J, Mellor D, Kao RR - BMC Vet. Res. (2012)

Bottom Line: Robust demographic information is important to understanding the risk of introduction and spread of exotic diseases as well as the development of effective disease control strategies, but is often based on datasets collected for other purposes.Otherwise, at a regional resolution, there are few differences between the datasets.Despite this, the similarity in distributions of these datasets is re-assuring, suggesting that there are few regional biases in the NED.

View Article: PubMed Central - HTML - PubMed

Affiliation: Boyd Orr Centre for Population and Ecosystem Health, Institute of Biodiversity, Animal Health and Comparative Medicine, College of Medical, Veterinary & Life Sciences, University of Glasgow, 464 Bearsden Road, Glasgow G61 1QH, UK. Lisa.Boden@glasgow.ac.uk

ABSTRACT

Background: Robust demographic information is important to understanding the risk of introduction and spread of exotic diseases as well as the development of effective disease control strategies, but is often based on datasets collected for other purposes. Thus, it is important to validate, or at least cross-reference these datasets to other sources to assess whether they are being used appropriately. The aim of this study was to use horse location data collected from different contributing industry sectors ("Stakeholder horse data") to calibrate the spatial distribution of horses as indicated by owner locations registered in the National Equine Database (the NED).

Results: A conservative estimate for the accurately geo-located NED horse population within GB is approximately 840,000 horses. This is likely to be an underestimate because of the exclusion of horses due to age or location criteria. In both datasets, horse density was higher in England and Wales than in Scotland. The high density of horses located in urban areas as indicated in the NED is consistent with previous reports indicating that owner location cannot always be viewed as a direct substitute for horse location. Otherwise, at a regional resolution, there are few differences between the datasets. There are inevitable biases in the stakeholder data, and leisure horses that are unaffiliated to major stakeholders are not included in these data. Despite this, the similarity in distributions of these datasets is re-assuring, suggesting that there are few regional biases in the NED.

Conclusions: Our analyses suggest that stakeholder data could be used to monitor possible changes in horse demographics. Given such changes in horse demographics and the advantages of stakeholder data (which include annual updates and accurate horse location), it may be appropriate to use these data for future disease modelling in conjunction with, if not in place of the NED.

Show MeSH
Bland-Altman plots showing the limits of agreement between the numbers of horses per postcode area and by region in the NED owner and Stakeholder horse datasets. These plots assume a relationship between the mean number of horses and the difference in the number of horses in each postcode area or region. The plots illustrate the agreement between the relative percentages of horses in each postcode area/region in the Stakeholder horse and the NED owner data. The owner location data provided by the NED has been restricted to include only those horses less than 30 years old. There are comparatively fewer horses in the Stakeholder horse dataset compared to the NED owner dataset in Dartford (DA) and London. Scotland appears to have comparatively greater numbers of horses in the Stakeholder horse dataset compared to the NED owner dataset.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC3351363&req=5

Figure 3: Bland-Altman plots showing the limits of agreement between the numbers of horses per postcode area and by region in the NED owner and Stakeholder horse datasets. These plots assume a relationship between the mean number of horses and the difference in the number of horses in each postcode area or region. The plots illustrate the agreement between the relative percentages of horses in each postcode area/region in the Stakeholder horse and the NED owner data. The owner location data provided by the NED has been restricted to include only those horses less than 30 years old. There are comparatively fewer horses in the Stakeholder horse dataset compared to the NED owner dataset in Dartford (DA) and London. Scotland appears to have comparatively greater numbers of horses in the Stakeholder horse dataset compared to the NED owner dataset.

Mentions: Bland-Altman limits of agreement plots illustrating the uncertainty around the mean number of horses in the two datasets per postcode area and per region are described in Figure 3. There were comparatively fewer horses in the Stakeholder horse dataset in the Dartford (DA) and London postcode areas. At a regional resolution, the distribution of numbers of horses in England and Wales was similar. In addition to the difference in horse numbers in London, the most significant difference between the two datasets was the greater number of horses in Scotland in the Stakeholder horse dataset compared to the NED owner dataset.


Summary of current knowledge of the size and spatial distribution of the horse population within Great Britain.

Boden LA, Parkin TD, Yates J, Mellor D, Kao RR - BMC Vet. Res. (2012)

Bland-Altman plots showing the limits of agreement between the numbers of horses per postcode area and by region in the NED owner and Stakeholder horse datasets. These plots assume a relationship between the mean number of horses and the difference in the number of horses in each postcode area or region. The plots illustrate the agreement between the relative percentages of horses in each postcode area/region in the Stakeholder horse and the NED owner data. The owner location data provided by the NED has been restricted to include only those horses less than 30 years old. There are comparatively fewer horses in the Stakeholder horse dataset compared to the NED owner dataset in Dartford (DA) and London. Scotland appears to have comparatively greater numbers of horses in the Stakeholder horse dataset compared to the NED owner dataset.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC3351363&req=5

Figure 3: Bland-Altman plots showing the limits of agreement between the numbers of horses per postcode area and by region in the NED owner and Stakeholder horse datasets. These plots assume a relationship between the mean number of horses and the difference in the number of horses in each postcode area or region. The plots illustrate the agreement between the relative percentages of horses in each postcode area/region in the Stakeholder horse and the NED owner data. The owner location data provided by the NED has been restricted to include only those horses less than 30 years old. There are comparatively fewer horses in the Stakeholder horse dataset compared to the NED owner dataset in Dartford (DA) and London. Scotland appears to have comparatively greater numbers of horses in the Stakeholder horse dataset compared to the NED owner dataset.
Mentions: Bland-Altman limits of agreement plots illustrating the uncertainty around the mean number of horses in the two datasets per postcode area and per region are described in Figure 3. There were comparatively fewer horses in the Stakeholder horse dataset in the Dartford (DA) and London postcode areas. At a regional resolution, the distribution of numbers of horses in England and Wales was similar. In addition to the difference in horse numbers in London, the most significant difference between the two datasets was the greater number of horses in Scotland in the Stakeholder horse dataset compared to the NED owner dataset.

Bottom Line: Robust demographic information is important to understanding the risk of introduction and spread of exotic diseases as well as the development of effective disease control strategies, but is often based on datasets collected for other purposes.Otherwise, at a regional resolution, there are few differences between the datasets.Despite this, the similarity in distributions of these datasets is re-assuring, suggesting that there are few regional biases in the NED.

View Article: PubMed Central - HTML - PubMed

Affiliation: Boyd Orr Centre for Population and Ecosystem Health, Institute of Biodiversity, Animal Health and Comparative Medicine, College of Medical, Veterinary & Life Sciences, University of Glasgow, 464 Bearsden Road, Glasgow G61 1QH, UK. Lisa.Boden@glasgow.ac.uk

ABSTRACT

Background: Robust demographic information is important to understanding the risk of introduction and spread of exotic diseases as well as the development of effective disease control strategies, but is often based on datasets collected for other purposes. Thus, it is important to validate, or at least cross-reference these datasets to other sources to assess whether they are being used appropriately. The aim of this study was to use horse location data collected from different contributing industry sectors ("Stakeholder horse data") to calibrate the spatial distribution of horses as indicated by owner locations registered in the National Equine Database (the NED).

Results: A conservative estimate for the accurately geo-located NED horse population within GB is approximately 840,000 horses. This is likely to be an underestimate because of the exclusion of horses due to age or location criteria. In both datasets, horse density was higher in England and Wales than in Scotland. The high density of horses located in urban areas as indicated in the NED is consistent with previous reports indicating that owner location cannot always be viewed as a direct substitute for horse location. Otherwise, at a regional resolution, there are few differences between the datasets. There are inevitable biases in the stakeholder data, and leisure horses that are unaffiliated to major stakeholders are not included in these data. Despite this, the similarity in distributions of these datasets is re-assuring, suggesting that there are few regional biases in the NED.

Conclusions: Our analyses suggest that stakeholder data could be used to monitor possible changes in horse demographics. Given such changes in horse demographics and the advantages of stakeholder data (which include annual updates and accurate horse location), it may be appropriate to use these data for future disease modelling in conjunction with, if not in place of the NED.

Show MeSH