Do numbers matter? Key questions for the public are: how is animal research conducted, what are the scientific and medical benefits and how well is it regulated? If these questions can be answered satisfactorily, I’m not convinced that they are too interested in the scale of animal research, ie the numbers of animals used.
However, a small item in today’s Guardian newspaper reports on an estimate for annual worldwide laboratory animal use that totals 115 million. The figure comes from a paper by antivivisectionists in the current issue of the journal ATLA published by the Fund for the Replacement of Animals in Medical Research. Unfortunately it is not freely available online, but I found a few minutes to read it.
There are fundamental questions about the way the antivivisectionists have collected and analysed their statistics. They include animals that are bred, but not used in research – so called ‘surplus’ animals – and animals that are bred and humanely killed so that their tissues and organs may be used in research. There is likely to be considerable overlap between the two, and both are excluded from the statistics published by the UK government, which regulates the use of living vertebrate animals in research. The authors also include in their ‘missing animals’ category GM animals used for breeding. They claim that except for two countries, these are not included in official estimates.
The authors arrive at a figure of 57 million worldwide for these ‘missing’ animals, extrapolated from extremely limited, very variable and out-of-date estimates. Yet this figure makes up more half of their estimated total of 115 million animals in research worldwide, a figure they know the media will pick up as the topline result.
For the ‘surplus’ they use old estimates produced by Great Britain (80%) and Norway (38%). The authors themselves admit ‘These data are very incomplete and variable, and may not justify a formal extrapolation to worldwide figures’ and that they ‘do not command such high levels of confidence’. Nevertheless, they use this arguably invalid extrapolation to add 59% to the total.
Another assumption which may not be valid is that most countries do not include breeding of GM animals in their official estimates. The authors allude to two countries that do include them; they comprise a significant minority (33%) and a very tiny minority (0.7%) of procedures in Britain and the Netherlands respectively. Based on these wildly disparate and limited figures, they produce an average of 17% and apply this to other countries, ie they add 17% to official estimates.
While examining the ATLA paper, I came across an interesting statistic that relates to rodents bred for research regarded as ‘surplus’. While many are used as breeding stock or humanely killed so that their tissues and organs may be used in research, some are killed and sold as pet food – for raptors and reptiles. According to the Wall Street Journal in 1999 (reproduced by the UK’s Animal Procedures Committee in 2003), some 180 million rodents are killed for this purpose every year in the USA. At least 10 times greater than the number of rodents used in research in that country.
The antivivisectionists must feel that overall numbers are important, as they have clearly spent a long time collecting a mass of data. But without context they are rather meaningless. That context may be ‘social’ as above, or related to the bigger medical research picture. For instance, spending on UK biomedical research has increased by at least 50% since 1995 while animal procedures have increased by only 18%.
The strict regulatory regime in the UK does provide, on an annual basis, a mass of statistics about animal research. We know that in 2007, 3.2 million procedures were conducted using just over 3 million animals, and that 83% of these procedures used rodents. We also know the numbers of GM mice used in breeding (850,000 in 2007). Of course, this is not news – we covered the latest UK statistics in a recent blog entry. Such bald figures may not be of much public interest either.
