TB restricted herds in England by county

In the above graph the percentage of restricted herds was calculated by dividing the number of herds under TB movement restriction (non-OTF) by the total number of herds.

The total number of herds for 2009 and 2010 were calculated by extrapolation using the total number of herds in 1998 and in 2008 for all counties except for Cambridgeshire. In Cambridgeshire, the total number of herds in 2007 were used instead of the total number of herds in 2008. All this data is shown in Fig 74 below.

Data is not shown for 2001 because an insufficient number of herds were tested in that year due to Foot and Mouth.


The same data is plotted in the above graph except that it is smoothed. According to Reference 1, testing in the following counties changed on 1st January 2013 so that the counties went on to one-year testing. These counties are listed below in descending order of TB level in 2008 after smoothing.

• Derbyshire
• Leicestershire
• Oxfordshire
• Cheshire
• Buckinghamshire
• East Sussex
• Northamptonshire
• Hampshire
• Nottinghamshire
• Berkshire

The next ones on the list continuing in descending order which are not currently due (Date of edit: 31Aug2013) to go to one-year testing would be as follows:

• Cambridgeshire
• Lincolnshire

Animal county data fails to give this clear picture.

Areas where the cattle population is too small to give a result with high statistical confidence are excluded from the above list. For example the Isle of White only had 4 TB restricted herds in 2010 but in terms of a proportion exceeded that of Cambridgeshire and Licolnshire in 2008 after smoothing.


In the above graph the total number of herds in each county are shown. Numbers for years 2009 and 2010 were calculated as described above.

DEFRA's ability to extend its annual county herd data beyond 2010 is in doubt

The following is an extract taken from email correspondence with DEFRA in 2010. It gives some background regarding the problems faced by DEFRA when analysing bovine TB data. This pre-dates SAM the new data processing system which replaced VETNET. Work to replace VETNET is referred to at the end of the extract.

[START OF EXTRACT]
The problem is that the VetNet system dates from the early 1980s, and has limited storage space. So the 'number of herds under restriction' is overwritten each week, meaning you can have a snapshot of the proportion of herds restricted, which we publish each month - as you say, it is a good indicator of the impact on the industry - but [does] not track which herds became restricted and unrestricted over time.

It is indeed less than ideal, not least because of the time it takes my team to extract and manipulate all the data each month. We're currently testing a new automatic reporting system which should make querying VetNet much easier and eliminate human error. The Animal Health agency are working on a successor to VetNet - some modules are in place but the TB module is the most complex and work has only recently started on that.
[END OF EXTRACT]

It appears from this that it was relatively easy to produce a snapshot of the number of herds under restriction each month. However due to lack of storage space it was not possible to keep track of which herds became restricted and unrestricted over time without overwritting data each week.

So what does this mean?

Such tracking would be required to calculate the ANNUAL number of herds from 1998 to 2012. Unlike Northern Ireland and the Irish Republic who have provided historic annual county data^7,8, DEFRA have only provided a monthly "snapshot" of recent county data on their web site. In addition to this no annual county data has been provided in response to recent Freedom of Information requests^9,4.

During the transition period from VETNET to SAM, due to teething problems with SAM, DEFRA had to resort to making paper records of some data as reported in Reference 3 which reports the European Commission's audit made in 2011. An extract is shown below.

[START OF EXTRACT]
AHVLA is seeking to reduce the reliance on a number of old "unstable" IT systems through the development and implementation of new software (linking the main bodies responsible for TB controls, including the laboratory network). In relation to TB controls, the roll out of the new TB module for the AHVLA system "SAM" release 6 scheduled for 5 September 2011 was postponed. During the audit, it was noted that case management was impeded due to the incomplete/delayed access to information during this transitional period. Some case officers maintained paper files for important dossiers.
[END OF EXTRACT]

This raises the question of whether data was lost during the transition from VETNET to SAM and, if data was lost, which years this affected (or affects, if problems are still ongoing).

Why is annual county herd data important?

Annual county HERD data is important because it gives a clear indication of how bovine TB is spreading into new counties as shown in Fig 76 above. Annual county ANIMAL data does not give this clear picture. In addition to this, the picture given by MONTHLY HERD data is less clear because it is subject to the strong cyclic nature of herd testing and how testing rises and falls throughout the year as shown by the monthly data shown in Reference 10. As such annual data is valuable as it helps to clearly identify counties which are newly emerging as high risk.

So what is DEFRA currently doing?

According to the reply to Reference 4 (dated 6th March 2013), DEFRA will shortly be carrying out a public consultation on the way that they publish regional statistics. They also state in the above reply that continuity of herd data back to 1996 has not been lost⁴. Whether or not continuity of their COUNTY herd data has been lost remains to be seen. (Date of edit: 2nd September 2013). Unfortunately a request sent in July 2013 to DEFRA asking them to update data, which they sent in 2009 to the author, was not successful.

David Heath, the Minister of State for Agriculture and Food (Date of edit: 31Aug2013) said in a reply dated 29Oct12 in Hansard Written Answers and Statements that quality assurance work is currently being carried out on historic country and county level bovine tuberculosis statistics and will be published during the next few months². As of writing this (Date of edit: 31st Aug 2013), this statement was made 10 months ago so hopefully historical county data will be published very shortly. In what form this takes, remains to be seen.

References

1. Changes to Bovine TB Surveillance. DEFRA. Bovine TB Information Note 04/12. October 2012.
2. House of Commons Hansard Written Answers for 31 Oct 2012
3. Final report of an audit carried out in the United Kingdom from 05 to 16 September 2011 in order to evaluate the operation of the Bovine Tuberculosis Eradication Programme. European Commission Health and Consumers Directorate.
4. Historical county herd data in England. www.defra.gov.uk. RFI 5713. 13th July 2013.
5. COUNTY HERD STATISTICS - GB REGION TOTAL
6. Issues associated with DEFRA's TB data processing module. Correspondence with England and Wales staff in 2010 concerning how to interpret data.
7. TB Statistics. DAFF.
8. TB restricted herds in Northern Ireland by county from 1995 to 2012. Data supplied by DARDNI in August 2013.
9. Freedom of Information request made to DEFRA on 6th March 2013. [Reply was not treated under FoI?]
10. Number of cattle tested and slaughtered for bovine TB