Counting the Cost Again: the 2019 Numbers

The numbers for UK animal-research procedures completed during 2019 have now been published by the Home Office. The total was a little over 3.4 million, a fall of 3% on the previous year. That means that there has been a modest decline in the total every year since 2015, tending to correct the brisk upward tendency which began after the year 2001, when the number was 2.62 million. We’re now back at any rate to pre-2010 levels. In fact, 2019’s total is, as the Home Office text says, “the lowest number of procedures since 2007”.

Back in 2001, that 2.62 million number was the lowest total since the 1986 Act had been passed, the lowest in fact since the mid-1950s. The notable fact was modestly presented in that year’s report as the first of fourteen ‘main points’ in ordinary black type, after eleven pages of general introductory matter. By contrast, this year’s achievement appears in a special box of ‘key results’ on page one, a three-colour affair enriched with graphics of various kinds, the numbers being set in eye-catching 36 point type. Why not? It makes navigation of the essential information that much easier. But of course it also quite changes the reading experience. The feeling you get is that the Home Office, rather than merely allowing you to know all this, as in earlier days, actually wants you to know: wants you to know that the numbers have gone down, certainly, but also, it seems, that 57% of the procedures were made for the purpose of ‘basic research’ – not obviously a point to boast about, but getting the same vibrant treatment in that text-box. The remainder of the report is laid out in a similarly easy-read style.

No doubt it’s partly the ‘Concordat effect’ that we’re seeing, and have been seeing gradually over the last few years of these government reports: the fashion, that is, for a more bullish PR, which celebrates rather than apologises for animal research, cleverly extolling at the same time both its claimed great achievements and the promise to do as little of it as possible. It’s also, I suspect, a response to the two-yearly Ipsos Mori surveys of attitudes to animal research (the next one is or was due this year: see notes below for previous ones). These surveys habitually find that respondents consider themselves ill-informed about animal research and regard the institutions that practice or supervise it as secretive and untrustworthy. “Come See Our World!” is how the promotional organisation Americans for Medical Progress title their digital introduction to the wonders of animal research. It’s a slogan which the Home Office now seems to have adopted too.

Here, anyway, are a few points about that world, as it was in 2019.

Regulatory testing:

This is probably the most unsavoury class of procedure, conducted to satisfy national or international laws of one kind or another. It continues to make up about one quarter of all the experimental (as opposed to GA animal-breeding) work. It’s the industrial end of animal research, involving the mass through-put of animals in standard testing regimes. The products and devices being tested include medical therapies, but also pesticides and other lethal products, and the techniques used for testing them still include, astonishingly, the ruthless LD50 AND LC50 tests. Accordingly this category of research is consistently the worst for animal suffering. In other experimental work, about 4% of the procedures are usually counted as ‘severe’; in 2019, the rate for regulatory testing was 10.8%. We are told (on page 14) that ‘severe’ procedures are those which cause “a major departure from the animal’s usual state of health and well-being”. Since all sixteen of the “eye irritation/corrosion” procedures conducted on rabbits in 2019 were classified as ‘mild’, and there is an intermediate category ‘moderate’, we can form some idea of how major that departure has to be.

Moreover, it’s in regulatory testing that the largest numbers of specially protected animal species appear (“animal species appear”! you see how numbers and tables push the mind towards abstractions): for instance, 3002 dogs (85% of the year’s total) and 2426 monkeys (71% of the total). Not that mice aren’t the most numerous species here as elsewhere: 437,124 of them were used in 2019.

Protection of the natural environment:

Most classes of animal research have shared to a greater or lesser degree in the reduction of numbers last year; even the breeding of GA animals, which has been mainly responsible for the increase since 2001, shrank by that same 3%. One class which noticeably did not shrink was ‘Protection of the natural environment’. This accounted for 13,074 animals in 2018, but for 29,343 in 2019. The animals included 5821 horses and “other ungulates”, 898 birds, and 22,079 fishes. It’s a category of research distinct (at least for statistical purposes) from regulatory testing and from general toxicology. The primary purpose is to understand the health implications of pollutants in the environment, but a common associated aim is the conservation of species and ecosystems: looking after animals, then!

An article about this sort of research, published in the journal Disease Models and Mechanisms earlier this year, conceded that epidemiology, the comparative study of whole populations, “can provide strong statistical inference linking chemical exposure to disease.” But that’s not enough; to understand the ‘mechanism’ of the harm being done, it has to be animal research. In fact the article is titled ‘Casting a wide net: use of diverse model organisms to advance toxicology’. You’ll notice the ambivalence of that last word: what’s to be advanced is both our understanding of poisons, and the scientific discipline called toxicology. This latter aim is the real subject of the article, whose authors wish “to shift the perception of toxicology as an applied science to that of a basic science” and thereby to “enrich the field”. This, they believe, can best be done by relying less on mice and rats, and resolving instead to “utilize diverse model systems”, especially fish (so “casting a wide net” turns out to be a sort of pun; don’t forget that science can be fun!). After all, they say, “The tree of life is vast”; why confine ourselves to “a few distinct branches”?

It’s a classic instance of scientistic thinking: calling in big science to fix the effects, while leaving the causes untouched. (We see it happening also in the case of Covid-19.) More to the point here, the article reminds us that there are always strong professional interests bound up in the growth (and resistant to the contraction) of any branch of scientific research, including those that use animals. That jump to 29,343 will have been good for some.

Roadmaps to nowhere:

Yet in fact both these classes of animal research, regulatory testing and protection of the environment, as well as toxicology more generally, should be especially amenable to non-animal technologies. That was indeed the main aim of a project set up in 2015 with the publication of a document entitled A non-animal technologies roadmap for the UK: advancing predictive biology. It was a collaboration between various government-sponsored agencies (including the National Centre for the 3Rs), academia, and industry, whose purpose was to promote the development and marketing of ‘NATs’ (surely a good sign that it seemed worth abbreviating ‘non-animal technologies’). The thinking was frankly commercial (further evidence of real purposefulness, sad to say): “The market potential is huge”, said the ‘Executive summary’. Correspondingly, the stated objections to animal testing were pragmatic, not ethical: its failure to predict for humans had “far-reaching implications, from wasted resources . . . to large financial losses”. As the trendy word ‘roadmap’ implies, this was a programme for the future rather than a survey; it was described in business-speak as “stretching towards a 2030 vision”.

Since we’re a third of the way along that road now, you’d expect the toxicology numbers to be showing some effect; you’d at least expect them not to have grown. But then I can find no surviving trace of the NATs project among all the other ‘roadmap’ projects boosting themselves online. What’s happened to it? The NC3Rs makes the document available on its web-site, but seems to have said nothing further about it since the day of publication. Another of the collaborating agencies, the Medical Research Council, makes no mention of it at all. More eloquently, the Medical Research Council’s laboratories were second only to the huge Francis Crick Institute as users of animals in research in 2019 (Oxford University came third).

That abandoned roadmap was part of an official programme of reform devised during the period of the Coalition government in the UK – largely the effect of having, however briefly, a minister in charge of animals in science who really wished to get them out: Norman Baker. Two other substantial publications had set out the aims of the programme and the progress being made: in 2014 there was Working to reduce the use of animals in research, and this was followed a year later, as promised in its text, by a review of progress titled Delivery Report. In fact the promise had been to publish such reviews “regularly thereafter”. But no others have appeared.

The number of animals used in 2015 was 4.14 million, the largest number since the 1986 act came into force. There’s been a 17% reduction since that high point – a return, as mentioned, to 2007 levels. Perhaps we must regard that much progress as the finished legacy of the Coalition programme, and now we’re left again with optimistic reassurances and pious references to the 3Rs. God knows there was nothing radical about that programme, but it had serious intention. To have let it lapse is a shameful delinquency.


Notes and references:

The report for 2019 and the separate tables of numbers can be accessed from this page: .

The Ipsos Mori surveys for 2016 and 2018 are reviewed in this blog here:

The Americans for Medical Progress digital tour of animal research was briefly reviewed in this blog here:

Among the various responses to the statistics, the one from Fund for the Replacement of Animals in Medical Experiments is especially authoritative. It talks about the Coalition publications, which were also touched on in this blog when Norman Baker was the subject, here:

The article ‘Casting a wide net: use of diverse model organisms to advance toxicology’ was published in April of this year, and can be read here:

These are the three Coalition documents discussed above:

Working to reduce the use of animals in research (2014: quoted above from p.9):

Working to reduce the use of animals in scientific research: Delivery Report (2015):

A non-animal technologies roadmap for the UK: advancing predictive biology (2015: quoted at pp. 4 & 6):