... and more

In our posting below we explored in a little more detail, Defra's headline grabbing claim:

A key finding of the field study, conducted over four years in a naturally infected population of more than 800 wild badgers in Gloucestershire, was that vaccination resulted in a 74 per cent reduction in the proportion of wild badgers testing positive to the antibody blood test for TB in badgers.

Further explanations have been passed to us which confirm that of the 844 badgers trapped in this project, after the prescreening with three different tests, just 262 were negative to at least one test. We questioned that background level of TB in the posting below, but now we have the words of the author himself which confirm:

"We worked in a high density population of naturally infected badgers, in what was the largest clinical trial in wildlife of its kind. Studies of vaccination are always focused on evaluating the prevention of new incident cases, so our analysis is based on 262 animals in 64 social groups that were test negative when they "entered" the study and which we caught a second time so that we could see how the vaccine had affected them. However, the total population size we report of 844 badgers is important, since it correctly includes all the animals that were already infected and gives an indication of the typical force of infection present in badgers in a TB problem area."

It most certainly does. 69 per cent of the badgers captured presented a result which would have guaranteed the death of a similarly tested bovine, and gives an indication why 34.9 per cent of herds in Glos had TB breakdowns in 2009. And 844 trapped over 55 sqkm gives a captured population density of 15.3 badgers per sq km. which is higher on both counts than the Consultation assumptions and which the Defra press release omitted to mention. But we digress...

In the posting, we quoted work done by Chambers et al, on the sensitivity (ability to detect disease) of the Statpak rapid blood assay. This work was published in 2008, and involved the postmorteming of almost 1500 badgers to more accurately validate this diagnostic test.

The Statpak achieved a very variable sensitivity ranging from 33 - 78 percent, the latter in grossly infected, super excreter badgers. Its average was a published 49.2 per cent only, which we compared with the much rubbished Brock test. If you remember, this is one Prof Bourne described this as 'poorer than hoped' as it 'only' detected about 40 per cent of infected badgers.
And Statpak starts life at 33 per cent? And averages 49 per cent? Do the maths.

But we are now even more puzzled by the Defra headline and inevitable media fest on this '74 per cent reduction' in the same breath as 'tuberculosis' and '800 badgers'.

The lead author on the Statpak validation we quoted in the paper below as 'Chambers et al., 2008, and is none other than Dr. Mark A.Chambers, who was lead author on the Vaccine project running at the same time. As were a number of the et als.

So let's get this right.
* While validating the Statpak blood test at an extremely low sensitivity in 2008, in the paper just published, the same authors attempt to morph their project to assess whether BCG is safe for badgers, into an efficacy test of BCG - even though they say it should not be taken as such?
* From an original trapping of 844 badgers, they then shake out the positives (582) leaving 262 testing negative of which around 160 are vaccinated. (60 percent of the 262, leaving 40 percent as controls ?) So it was not 844 trapped badgers which formed part of this 'efficacy' bit, (that should not be counted as efficacy) but 160?
* Having turned the badgers loose after their annual vaccination, the authors have no knowledge of what (or even if) any challenge from m.bovis has been faced. But they assess the results of the BCG vaccination on approximately 160 naive candidates, with gammaIFN (sensitivity 80.9 percent) and achieve a benefit of 19% to this unknown challenge ?
* Similarly with Statpak, which their own validation procedures give only a 49 percent sensitivity to, and they achieve 74 per cent benefit, again on an unknown, unquantified challenge ?

Having been questioned as more of this comes to light, the authors are keen to stress that :

"It's important to realise that the 74% (73.8%) figure represents a reduction in incidence of positive antibody tests brought about by vaccination and should not be equated to a vaccine efficacy of 74% "

And they do say that in the paper. But it is a pity Defra (or whoever wrote the press release) didn't realise that. Or the media or anyone else similarly taken in mislead by the headlines which followed the press release.

Dr. Chambers is also keen to stress that Statpak is very sensitive when faced with grossly infected badgers. Sure it is, but in this project the badgers were not post mortemed. So was it the 33 percent end of Stakpak's very variable sensitivity which it was flagging up? Or the 'more sensitive' 78 per cent? They don't know because they didn't look.

And then the inevitable wriggle. That annexes may be overlooked,(nope, read those too) and possibly not immediately explicit to 'lay-persons'. ( A wild assumption there.) And the fact that the main paper was prepared for submission to a regulatory authority as a 'health and welfare' issue for badgers. But presumably not submission to the VMD as an indication of efficacy, as the previous Statpak validation paper would have already covered that bit?
Nothing like telling you, you're stooopid is there? And of course cannot be expected to understand papers of this ilk.

And then the nitty gritty (that's not a very 'scientific' term, but hey, we'll live with it):

"...the decision was only taken subsequently by Defra to make the data widely available as part of the public consultation."

Well, well well. Impeccable timing with a startling, if misleading headline. So it was Defra who decided to publish. So that's all right then. And the media fest, headline grabbing, 800 badgers, 74 per cent reduction in TB? Which most of Defra's 'lay readership' swallowed hook, line and sinker? Is that all right too ?

Of course it is. And not a hair of a single badger will be harmed. Even the hairs on the heads of the 43% blood assay positive ones, roaming the Gloucestershire countryside and still available for more research.

Things are never what they seem.

(We have updated the headcount of blood assay positive rejects to this research, after contact from the authors. They offer around '43 per cent' of the 844, a figure which they describe as a 'typical force of infection present in badgers in problem areas'. The remaining 26 per cent comprised badgers which were only trapped once and those which expired during the research.)