bovine tuberculosis – The Badger Crowd

Defra’s last stand?

Posted on 18/10/2023 by The Badger Crowd

APHA show a little bit of their hand at last

In March 2022, a peer-reviewed analysis of badger culling data was published in a top veterinary journal. It used all publicly available data. It showed no detectable difference in bovine TB breakdown between badger culled areas and unculled areas since 2013 (1). Now, after a government information black-out following their look at the data up to 2017, the Defra’s Animal and Plant Health Agency (APHA) have at long last posted a pre-print of their own creation looking at their secretly held data up to 2021.

It is not clear why it has taken so long to release this new analysis, a draft of which was presented at a conference in August 2022 (2).

Cynics might say there has been a search for obscure statistical methods to get the results needed to try to show a badger culling has not been a 100% failure. And of course, it supported the pro-cull rationale to hold data secret and keep quiet about any weakness in policy. This served to enable the 2022 and 2023 killings of over 50,000 mostly healthy badgers, planned and executed by the willing and obedient hands of Natural England (3).

The new pre-print, posted 6^th September in ‘bioarchive’ (bioRxiv), is by six members of APHA staff. It tries to use an obscure technique: “Difference in Differences analysis evaluates the effects of the Badger Control Policy on Bovine Tuberculosis in England” to look at Bovine TB breakdown incidence (Withdrawn herds: OTFW) between 2009 and 2020 (4). There are many aspects of this new analysis which are problematic, but here are a few of the more obvious.

A new statistical method of analysis – ‘Difference in differences’ (DID)

This is the first time the DID approach has been used in badger cull analyses and it is simply inappropriate. DID compares the changes in outcomes over time between a population that is enrolled in a programme (a group subject to an intervention) and a population that is not (the control group).

The attempt here is to use the same areas in a ‘before and after’ way. There is no clearly defined control group, it is just the fragmented cull areas before they are licensed to cull. The 52 cull areas are surrounded by and influencing each other during a wide range of tightening testing interventions. They are, therefore, not adequately discrete in space and time, breaking a fundamental DID requirement. So not only is this (DID) an unorthodox approach, but it does not follow the basic rules of the unorthodox approach. It is looking a bit desperate already.

Notably, in the period 2016- 2020, pre-cull and cull areas closely juxta position. Pre-cull and cull cattle testing interventions of course influence changes in adjacent pre-cull and cull areas in an irregular and unpredictable manner, via the constant movement of cattle with different testing and control histories. It’s a mess, and surprising that this has been presented as an appropriate statistical method, considering the subject and the epidemiology involved.

Analysis does not distinguish between different causes of disease decline.

Does the analysis decisively link badger culling to decline in bovine TB in cattle? Not at all. Any possible effect of badger culling is confounded with the continuous multi-faceted and complex tightening of cattle testing that took place in an uneven manner over different and irregular time periods. Some of the very important bTB testing controls used, and the timing of their introduction, are incorrectly described and some of the most important ones are not even mentioned.

The Badger Culling Policy invention

There are places in the manuscript where a claim is made for a link to badger culling, ‘The effect of badger culling…’, but then later this is adjusted to ‘However, this data analysis cannot explicitly distinguish…….’ APHA have stated more than once in their recent epidemiology reports that the cattle breakdown data alone are insufficient to show an effect of culling.

Frankly there is no such thing as the ‘Badger Culling Policy’ (BCP) as framed by this report, beyond just killing badgers. It is an invention trying to characterize extensive cattle testing and movement control measures as just a part of the badger culling bolt-on, rolled out six years after testing control measures had been continuously tightened year on year. To pretend the extensive suite of testing and movement control measures is somehow a part of a whole is just a repeat of the ‘all tools in a box’ nonsense and it deceives. True, some of the more essential measures were increased from the first year of badger culling, but most of them were independent of it and ramped up gradually. So trying to pull them apart in a post-facto try-on will not impress the independent scientist, even if an obliging journal prints it, and industry and Defra promote it with their coterie of academic cheerleaders. Badger culling has just been a pointless distraction to cattle surveillance and control measures to address rampant cattle-to-cattle transmission.

So again, no, the analysis is not able to show causation by badger culling of any change in disease, and Defra and APHA’s judgement could be seen ill-judged by trying to convince anyone that this is credible. It is as clumsy as the Chief Vet’s apologies and efforts on Radio 4’s Farming Today in 2022. Defra boss Richard Benyon has been claiming big figures for badger cull benefits for years, but this does not provide any evidence for them.

What is the ‘true burden of disease’?

The new analysis has a surprising and unconventional interpretation of the ‘true burden of the disease’ or disease ‘prevalence’ in a population. It suggests that ‘incidence’ is the better indicator of true burden, which they must surely understand flies against first principles. It is ‘old thinking’ based on the out-dated views of farm vets who claimed that only visibly lesioned cows are infectious. Yes, incidence has been one unit of disease measurement, and it was the unit used by the RBCT to claim a benefit for badger culling. But science has moved on, and this is a strange attempt to live in the past.

The interpretation contrasts with APHA’s reports and understanding of disease controls if you look closely. It contrasts with the conclusions drawn from better comprehension of the SICCT (skin) test sensitivity and specificity, which has been clarified in recent years. Gamma IFyN testing has revealed a very significant reservoir of undetected reactors (diseased animals) in bTB infected herds, pointing towards undisclosed infection in undetected herds too. This ‘hidden’ reservoir in cattle that remains is what really matters for disease understanding and control. The true burden is bigger than the known breakdowns and identified reactors, why else would the disease persist? Both here and in Republic of Ireland where badgers have been relentlessly persecuted and SICCT and Gamma testing has failed for decades.

So why exactly did APHA attempt this clumsy redefinition of the ‘true burden of disease’. Perhaps because it needs to try to defend the original RBCT analysis which by controversial statistical modelling (5), managed to suggest a relationship between incidence and badger culling, having failed to find a relationship with disease prevalence. The effort applied to carry on killing badgers is deeply disturbing.

But wasn’t bovine TB already reducing before badger culling began?

Levels of Bovine TB were falling in many areas soon after annual SICCT testing was introduced in 2010. And well before culling started in most areas. However, the new APHA analysis reduces the analysis of pre-cull data to one data point. This conveniently helps to conceal the significant pre-cull decline and masks the true disease trajectory and results. A logarithmic scale is also used to distort visual effect. Not unlike methods used in the earlier Brunton and Downs reports.

Recent Edge Area data added into the analysis for what purposes?

The new analysis mixes data from 46 High Risk Area study areas with 6 from the Edge Area. These have very different epidemiological and disease control history profiles and the reason for mixing them is not explained. It looks like a deliberate attempt to manipulate the data. Pre-cull gamma testing was intense in the Edge Area. While the manuscript mentions additional gamma testing from 2017, gamma testing was erratic between areas and over time. There were considerable numbers of gamma reactors in many of the cull areas in cull years 1 and 2 of culling with similar disclosures levels to years 3 and 4, but pre-cull use in the HRA was generally low. Mixing them may give you a result you like, but it can be seen through for what it is.

In both HRA and Edge areas, bTB incidence was declining when many of the additional disease control rules were intensified, and badger culling introduced. The addition of this extra data makes it even more impossible to distinguish effects of badger culling and disease control measures that are known to drive down bTB.

Attempt to dismiss peer-reviewed 2022 study published without mentioning it

A comment in the text of the new pre-print says that it is not possible to match badger culled and control areas. This is not only incorrect but unevidenced, with unsubstantiated claims. The 2022 published and peer reviewed study that used this methodology effectively has not been cited. There is no mention of the even more obvious alternative to their complex DID approach, which would be to simply match a series of individual farms in culled and unculled areas. This approach is only available to APHA, which holds all individual farm records confidentially. Such simple monitoring could and should have been an expected outcome of a High Court ruling that required the government to ‘adapt and learn’ from badger culling. Why has it not been done? Perhaps it has been done but gave the wrong results?

Distinct lack of clarity

There is a definite need for clarity in the analysis presented, including the Appendix. This does not seem to be the output from the constrained DID analysis, but something else that is not fully described in the methodology. Analysis uses the lesser known system called STATA (nearly all analysts use R-code). The data is not supplied, and the Stata code is not supplied, so no one can replicate what has been done here. An author, when contacted, said these would be available following publication of the paper. Too late for others to comment on?

Does this draft show that badger culling has reduced bovine TB In cattle?

No, it doesn’t. Because the methods are inappropriate, the results are flawed, and so the conclusions are wrong. There is no way to distinguish between different interventions and change in herd breakdowns since 2013 with the approach taken. Might these results be misleading and deceive if published? Yes. Why has such a flawed analysis been produced now? Is it a try-on to justify Natural England’s new autumn badger blood-fest? Will Defra contractors, grant recipients and a friendly journal whisk it through peer review with recommended friendly reviewers? Probably.

A consultation on culling all badgers over wide areas is being cooked up despite previous policy promises. One could speculate that APHA have been told both by bosses and lawyers that they need to produce evidence that the culling policy has not been a complete waste of life, public funds, and other resources including masses of police time. They are desperate to do this, and get it out right in front of the move to ‘carry on culling’. This preprint aims to be the new truth about badger culling, but it’s all smoke and mirrors.

References

BBC misleading again on badger culling?

Posted on 05/07/2023 by The Badger Crowd

10 years on and the BBC repeats acknowledged errors, misleading the public on badger culling

On 13^th May 2013, the BBC posted an article entitled “How did the Irish badger cull play out?”.

On 1^st June 2013, this report was produced on the impact of the badger cull in Ireland.

Both of these pieces were by BBC Journalist David Gregory-Kumar (DGK). Both these stories, and others beside support the controversial government view that badgers play a significant role in bovine TB control in cattle. Complaints to the BBC by ecologist Tom Langton eventually led to the following BBC ruling:

Given that both the statistics and scientific studies about the link between badger culling and bovine TB levels were inconclusive, the Committee examined whether the article had used clear and precise language to make this apparent to the audience. In doing so, it observed that the badger cull had become a highly divisive issue with those for and against the cull using the findings of different trials to bolster their respective causes and so it was essential that the BBC provided the greatest clarity possible on the subject.

In this context, the Committee believed that the language used in the article had not been sufficiently precise as it suggested that the badger cull might be a factor in helping control the disease when this was scientifically unproven. It considered that, while the data did show a decline in the number of cattle infected with TB in Ireland, there was no conclusive evidence to show that the badger cull had been categorically responsible for any of this decline and so it was inaccurate to say that, along with other measures, it can help control the disease.

Exactly the same error was repeated by the same journalist in a piece on BBC’s Farming Today on July 4th 2023, 10 years later, this time regarding the English culls. DGK plays a clip in which farmer James Griffiths says “badger culling has undoubtedly made a difference, no one can deny it” DGK follows this by saying “the latest data seems to support this.” The latest data does not support this. The latest data shows bTB falling generally because of cattle measures but the role of badger culling, if any, is not known. Why is DGK doing this we wonder?

DGK forgot to mention that there were more bTB breakdowns in Gloucetreshire pilot 1 area in 2022 than in 2013 when badger culling began.

And again, in ‘Has a decade of badger culling worked?’ published on the same day, the bias continues.

There is no mention of the peer-reviewed paper published last year by Vet Record, “Analysis of the impact of badger culling on bovine tuberculosis in cattle in the high-risk area of England, 2009–2020” by Tom Langton, Mark W. Jones, and Iain McGill. This examination of government data obtained over a wide area and a long time period failed to identify a meaningful effect of badger culling on bTB in English cattle herds.

Defra has failed to publish any analysis that shows that badger culling has produced any disease benefits since its much caveated publication in 2017, which was based on limited data over a limited period of time. DGK says that to show that badger culling is working you would need to “compare areas with a cull to those without and we can’t do that because the culls are now everywhere.” This is not the case. Defra holds all the data for all farms, culled and unculled, and it would be a relatively simple job to extract data from culled farms and compare them with unculled farms using a range of available techniques. They may have already done this. If not why not?

Overall, the bTB disease figures from Wales (where there is virtually no culling) and England (where 210,555+ badgers have been culled) have been very closely aligned. This is, perhaps, the best evidence that badger culling plays no role in bTB control.

Incidentally, Jeff Sim at Staffs Wildlife Trust who was interviewed said that he was vaccinating badgers to reduce bTB in cattle. But in fact such a relationship is not known and is purely speculative. Wales gave up on vaccinating badgers a long time ago.

Finally DGK’s idea that badger cull benefit lasts 4 years is not solid science. The confidence intervals on that old data indicate nothing scientifically significant. The late statistician David Cox who was involved with the RBCT is reported to have said at a conference later that he wished they had presented it as trends rather than as significant as the RBCT confidence intervals were also mostly very large.

The real story is that 6-monthly cattle testing in the High Risk and Edge Area are now starting to work.

NEW REPORT: A Bovine Tuberculosis Policy Conundrum in 2023

Posted on 12/05/2023 by The Badger Crowd

A team of researchers and veterinarians have concluded an urgent need for rapid change to bTB policy for England, warning the present focus on badgers is undermining efforts to combat the disease.

You can read the report HERE:

The Independent story on the report “Government’s badger cull plan ‘deeply flawed’, damning report warns” is available here.

The Vet Times story on the new report, 11.05.2023: “New report blasts ‘weak science and huge supposition’ behind bTB policy” is available here.

Updated Vet Times story, 22.05.2023: “Immediate rethink sought as report condemns badger cull” is available here.

Human TB – Bovine TB. Lessons to be learnt and three simple questions for DEFRA

Posted on 15/08/2022 by The Badger Crowd

While Defra hold on to a now disproven view that badger culling is needed for control of bTB in cattle, there remain important omissions, contradictions & unanswered questions in their approach to the current management policy.

For a start, the management of Bovine TB (bTB) is still plagued by the inaccuracies of currently used tests and testing systems. Importantly, none of the primary tests currently used can identify the presence of live Mycobacteria which is core to the central dogma of bacterial diagnosis developed by Robert Koch, one of the main founders of modern bacteriology. He discovered the causative organisms of anthrax (1877), septicæmia, tuberculosis (1882) and cholera (1884). Koch’s dogma or principle, specifies that definitive diagnosis is dependent on the identification of the presence of the infectious organism.

There are major concerns about elements of the front-line bTB tuberculin skin-test. Firstly, it misses a substantial proportion of infected animals, thus allowing undetected infection to circulate in and between herds. Secondly, repeated injection with the tuberculin injection used for the skin test may potentially render an animal skin-test positive, despite there never having been any infection. Thirdly, the implementation of the skin-test is laborious and time-consuming, and requires multiple visits to the farm by a vet, hence while it creates work for vets, it is expensive.

In Human TB (hTB) the basic biology is similar. Recently, however, the human TB clinical fraternity have become increasingly concerned about the TB skin-test. In fact the concern has been such that the Centres for Disease Control (CDC), based in the US, now recommends the use of the TB blood tests (IGRAs), such as QFT-Plus (similar to Bovigam) over the TB skin test (TST) for most risk groups.

To illustrate the importance of this change of approach, clinical studies on the phage-based test ‘Actiphage’ have shown for the first time that live bacteria can be detected in the blood of people with incipient TB infection, including contacts of infected patients. This success was mirrored by the data produced by vet Dick Sibley at Gatcombe who used Actiphage and Bovigam (gamma interferon) to effectively eradicate bTB from the farm on two occasions. Those studies showed clearly that the best way to identify infected animals was the combined use of Bovigam and Actiphage, and not skin-testing. They indicated very strongly that the biggest problem in the eradication of bTB is the inability of skin-testing to identify infected carriers which actually maintain the infectious burden within the herd. They also imply that there is no significant involvement of external non-bovine hosts in the propagation of the disease.

So, the questions we would like to put to George Eustice and DEFRA are:

1) Why have you not implemented a controlled trial on the efficacy of different testing strategies for Bovine TB?

2) Why will you not take note of the human TB data? This clearly shows that identification of the live bacteria is the critical key to controlling this disease.

3) When will you recognise that all the data from both bovine and human TB indicates that there is no justifiable scientific rationale for the involvement of an external host to maintain persistent infection in the herd?

DEFRA called out over flawed bovine TB claims at international vet conference

Posted on 13/08/2022 by The Badger Crowd

The UK’s Animal and Plant Agency statistician Colin Birch was roundly criticized for his presentation yesterday (12/08/22) at the 16th International Symposium of Veterinary Epidemiology and Economics (ISVEE 16) held at Halifax Convention Centre, Halifax Nova Scotia, Canada.

Birch presented data from badger killing zones in England in recent years, with no proper explanation as to why he had not also used data from unculled areas to compare. He claimed that a reported 50% reduction in bovine TB herd incidence in culled areas was due to badger culling.

The audience seemed less than convinced. At the end one question pointed out that it is not possible to attribute the reduction in bTB incidence to badger culling as the reduction in the unculled area had a similar trajectory. Cattle measures (Testing and movement controls) that were introduced prior to and over the same period (in both culled and unculled areas) would reduce incidence in the manner observed.

A further point was made from the audience that it looked like Birch and APHA were trying to make and promote ‘policy driven evidence’ to satisfy the ministry (Defra). Birch had no coherent response to this but said that he did not agree.

The unpublished manuscript by Birch and others is yet to be fully disclosed, but comes at a highly sensitive time for Defra and Minister George Eustice and Natural England Chairman, Tony Juniper and his scientific staff. They want to sign off the killing of another 40,000 largely healthy badgers from September of this year, despite the science suggesting that complete failure is the most likely outcome.

In March of this year, Defra issued flawed data (see here) in response to a detailed peer reviewed paper (see here) published in Veterinary Record which showed that badger culling in England since 2013 has failed. In a response to the paper, Defra produced a media outburst designed to undermine it, that claimed badger culling had little or no effect in the first two years, and therefore the analysis used was flawed. Observers have been left baffled and talking about government competence, since all the Defra data presented shows large drops in herd incidence over the first two years, suggesting that it is cattle measures that are responsible for these declines, and not culling.

Despite high public interest in this most controversial of policies, Defra have become tight-lipped on their home-made dilemma since March 2022, and defiantly issued more cull licences in June. But despite well and truly losing the science argument they still appear desperate to try to show some reason to prop up their policy and to enable them to keep killing badgers. This fell flat at today’s conference as the science community strongly questioned Defra’s handling of data.