12 Monkeys. No… 8. Wait, sorry, I meant 14.

January 23rd, 2010 by Ben Goldacre in bad science, methods, trial registers | 77 Comments »

Ben Goldacre, The Guardian, Saturday 23 January 2010

Like many people, you’re possibly afraid to share your views on animal experiments, because you don’t want anyone digging up your grandmother’s grave, or setting fire to your house, or stuff like that. Animal experiments are necessary, they need to be properly regulated, and we have some of the tightest regulation in the world.

But it’s easy to assess whether animals are treated well, or to assess whether an experiment was necessary. In the nerd corner there is another issue: is the research well conducted, and are the results properly communicated? If it’s not, then animals have suffered – whatever you believe that might mean for an animal – partly in vain.

The National Centre for the Replacement, Refinement and Reduction of Animals in Research was set up by the government in 2004. It has published, in the academic journal PLoS One, a systematic survey of the quality of reporting, experimental design and statistical analysis of recently published biomedical research using laboratory animals. It’s so not good.

The study is pretty solid. They describe the strategy they used to search for papers, which is important, because you don’t want to be like a homeopath, and only quote the papers that support your conclusions: you want to have a representative sample of all the literature. And the papers they found covered a huge range of publicly funded research, behavioural and diet studies, drug and chemical testing, immunological experiments, and more.

Some of the flaws they discovered were bizarre. Four per cent of papers didn’t mention how many animals were used in the experiment, anywhere. The researchers looked in detail at 48 studies that did say how many they used: not one explained why they had chosen their particular number of animals. Thirty-five per cent of the papers gave one figure for the number of animals used in the methods, and then a different number of animals appeared in the results. That’s pretty disorganised. They looked at how many studies used basic strategies to reduce bias in their results, like randomisation and blinding.

If you’re comparing one intervention against another, for example, and you don’t randomly assign animals to each group, then it’s possible you might unconsciously put the stronger animals in the group getting a potentially beneficial experimental intervention, or vice versa, thus distorting your results.

If you don’t “blind”, then you know, as the experimenter, which animals had which intervention. So you might allow that knowledge, even unconsciously, to affect close calls on measurements you take. Or maybe you’ll accept a high blood pressure reading when you expected it to be high, knowing what you do about your own experiment, but then double check a high blood pressure measurement in an animal where you expected it to be low.

Only 12% of the animal studies used randomisation. Only 14% used blinding. And the reporting was often poor. Only 8% gave the raw data, allowing you to go back and do your own analysis. About half the studies left the numbers of animals in each group out of their tables.

I grew up friends with the daughters of Colin Blakemore, a neuroscientist in Oxford who spoke out to defend animal research at great personal risk. My first kiss – not one of these sisters, I should say – was outside a teenage party in a church hall, in front of two special branch officers sitting in a car with their lights off.

People who threaten the lives of 15-year-old girls, to shut their father up, are beneath contempt. People who fail to damn these threats are similarly contemptible. That’s why it sticks in the throat to say that the reporting and conduct of animal research is often poor; but we have to be more grown up.


++++++++++++++++++++++++++++++++++++++++++
If you like what I do, and you want me to do more, you can: buy my books Bad Science and Bad Pharma, give them to your friends, put them on your reading list, employ me to do a talk, or tweet this article to your friends. Thanks! ++++++++++++++++++++++++++++++++++++++++++

77 Responses



  1. ChrisP said,

    January 23, 2010 at 7:53 am

    Important article – count me gobsmacked.

    Does anyone know if the Home Office check the publications arising out of animal research to make sure the methodology and reporting were sound? A poor output would be a good reason to revoke a licence.

    I’m assuming the methodology has to be clearly specified before a licence is granted, or perhaps this is left to individual ethics committees?

  2. ayupmeduck said,

    January 23, 2010 at 8:48 am

    Let’s see what PETA supporters and their like make of this. This, and the similar referenced papers, could be used to support their case. Or they could attack the messenger. That The Guardian is not allowing comments on it’s site, suggests that they suspect the later. Hope I’m wrong.

  3. nickuk72 said,

    January 23, 2010 at 9:46 am

    Nothing like jumping to conclusions without having all the facts ayupmeduck – ironic given the subject of the article

  4. md1364 said,

    January 23, 2010 at 9:52 am

    Really Ben, your campaign against the selective data use of certain groups to reinforce their results (or lack of) is documented ad nauseam, and I find the use of ‘…like a homeopath’ to be unnecessary and by now just childish. Reminiscent of high school piss taking, when I read one of those throw-away references it blinds me the often superb wit and intelligence of your blog/column to the point where I am likely to stop reading altogether in a state of disillusionment.
    I hope this will not elicit a similarly childish response from you because this site provides me with so much excellent and nutritiously healthy food for thought that I really don’t want to have to abstain for such a small percentage of junk content.
    Keep up the crusade but get off the soap box sometimes.

  5. ayupmeduck said,

    January 23, 2010 at 10:09 am

    @nickuk72 – What conclusion did I jump to?

  6. David Colquhoun said,

    January 23, 2010 at 10:32 am

    @md1364
    I think you need to do your homework. The “European Network for Homeopathy Researchers” actually had the nerve to produce a document “An Overview of Positive Homeopathy Research and Surveys”/ It shamelessly and quite explicitly selected the positive results and omitted the negative ones. No wonder homeopathy is regarded as a sick joke. www.homeopathy.org/research/research_reviews/ENHR.pdf

  7. Keith19 said,

    January 23, 2010 at 10:50 am

    In my opinion this is one of the areas that the pharmaceutical industry is leading academia in basic science animal research. Randomisation and assessing statistical power before an experiment is carried out is the norm. Blinding is harder, but again is starting to become routine and is actively encouraged by management and statisticians. This is presumably because mistakes at this point in the research, either false positives or negatives, can end up costing the companies millions.

  8. osipacmeist said,

    January 23, 2010 at 10:51 am

    I hope we can expect more like this from BG. Bad Science needs to be publicised and criticised, it might do some good. Plain ole crazy wooh just likes the publicity.

    and

    OOh Md1364. Almost as pompous as a nutritionist…. You must be so mature.

  9. md1364 said,

    January 23, 2010 at 11:05 am

    @osipacmeist

    dare I say typical? There was no attack in my comments just a critique of style, why would you feel it necessary to compare me to someone you obviously hold in such low regard?
    Whatever

  10. gimpyblog said,

    January 23, 2010 at 11:10 am

    This is a very tricky area to debate and I do have several criticisms of the report. Attention should perhaps be drawn to the fact that the authors only looked at published research. There are many animal experiments where the results will never be published. Either because the project failed, someone else beat the researchers to it or the findings weren’t sufficiently interesting to be worth publishing. I expect an analysis of these numbers might make the figures on animal research look even worse. Also I feel the reviewers exclusion criteria were too strict as it appears to ignore the huge numbers of animals used to create gene knockouts, knock ins, and various other mutations that are then used to provide tissue for later experiments. It also appears to exclude the use of animals in the production of reagents such as antibodies. Additionally it is restricted to studies of higher vertebrates (mice, rats, primates) and primates themselves were included because they were’ethically sensitive, rather than being widely used, despite other species such as fish, birds, rabbits, etc being excluded because of “small numbers or [use] in more specialised areas of research”. It is possible that focussing on primates where (anecdotally), it seems media scrutiny has resulted in particularly high standards compared to other species possibly distorting the analysis.

    This criticism would be to miss the point though. I think we should move away from arguments that animal research should be replaced or reduced, and focus on refining existing protocols and recognise that animal research is here to stay, it is far too useful and informative to ever abolish and there are entire fields of inquiry where the use of animals is a necessity (developmental biology for example).

  11. Sili said,

    January 23, 2010 at 11:14 am

    As ChrisP said, this should be checked by the Home Office.

    Denmark has an Council for the Ethical Treatment of Animals completely analogous to the one for humans.

    I believe it’s not possible to do experiments on animals without first getting permission from them. This includes demonstrating necessity, among other things providing a comprehensive literature review to show that the research either hasn’t been done before or that replication is paramount.

    Presumably experimental design is then scrutinised as well to ensure that the results are meaningful, but I do not know if this is the case. But animal welfare is certainly ensured.

    My alma mater hired a professor of animal research some years ago, who was to deal with exactly these issues to help reduce the use of animals. She emphasised the need to be completely open about this kind of research.

  12. adzcliff said,

    January 23, 2010 at 11:19 am

    Md1364. Personally I’m here for BG’s soap-box. Surely it’s up to blogs’ readers to decide whose soap-box they find interesting and whose they don’t? Blog, soap-box, all the same thing to me.

  13. thepoisongarden said,

    January 23, 2010 at 11:30 am

    2.ayupmeduck

    The Guardian is, now, accepting comments. I just posted a fairly inane one just to check.

    Technical oversight, error or change of mind under pressure?

  14. Reggie Dixon said,

    January 23, 2010 at 3:46 pm

    @Md1364
    You aren’t a homeopath by any chance ? Just dip Ben’s article in water then dilute it trillions of times, you know the score.

  15. md1364 said,

    January 23, 2010 at 9:17 pm

    @David Colquhoun. Is it the way I write or the way you read? How did you interpret what I said as disputing the fact that certain groups do use data selectively? If something can’t be proven empirically then so be it, if it doesn’t work at all that’s ok too, I don’t understand why homeopaths would want to get into a discussion with people whose mindset is not open to what they believe or why doctors would feel it was necessary to argue endlessly with people whose methods they believe to be totally ineffectual. The point I was making was simply that the comment about homeopaths was a jibe that did not need to be included to make the point, I feel the whole context was sidelined for a pointless snide remark, which only served (in my opinion…. most importantly) to reduce the impact of the piece.
    @adzcliff Point taken and understood, if you like this then that is fine, I was only offering a point of view that, having re-read the article, I still feel is valid.
    @Reggie Dixon So thick I can’t even be bothered to address your ….. point?

  16. chris lawson said,

    January 24, 2010 at 2:53 am

    @md1364: You may find that the statement “There was no attack in my comments just a critique of style” somewhat incompatible with “childish”, “reminiscent of high school piss taking”, and, uh, “childish” again.

  17. md1364 said,

    January 24, 2010 at 9:24 am

    @ chris lawson Yes point taken and I am sorry for having contributed towards taking this so far away from the serious cause for concern that Ben Goldacre intended to be the focus of his blog.
    My intention was to address a point to the author personally and instead I seem to have mobilised his online protection squad, when I stated that I find this column and blog to be intelligent and written with wit it was totally ignored as was the part about how it was one small area that I objected to, so my last contribution here is to ask where is the balance in your appraisal people?
    I was guided to this site after reading the Bad Science book which I greatly enjoyed but I am finding the one-eyed attitudes that I encounter so often amongst the comments here to be a serious turn off, almost as biased as a homeopaths results!

  18. Indy said,

    January 24, 2010 at 9:36 am

    @md1364: Sorry to have to point it out, but you seem to have fallen into a common trap of calling attitudes biased when they happen to disagree with you.

    Fantastic that you like Ben’s work, and appreciate its importance. However, you may have missed the point of some of it! This is that judgements in science must be based on the evidence, and therefore it is not biased just considered to judge homeopaths and their ilk by the standards to which we endevour to hold science and science reporting.

    If you are going to make a criticism, you must expect it to recieve comments when it seems to boil down to you having a soft-spot for homeopaths. Live by the sword…..

  19. raphazlab said,

    January 24, 2010 at 10:10 am

    The PLoS study starts by:
    “For scientific, ethical and economic reasons, experiments involving animals should be appropriately designed, correctly analysed and transparently reported”
    Although there are additional ethical concerns relating to the use of animals, surely, the sentence is still valid without the reference to animals:
    “For scientific, ethical and economic reasons, experiments should be appropriately designed, correctly analysed and transparently reported”
    This then raises the question of whether experiments involving animals are just average, better or worse, in terms of design, analysis and transparency or reporting, than experiments not involving animals.

  20. simoncardnell said,

    January 24, 2010 at 12:12 pm

    Obviously the focus this, is on the process of the experiments rather than the rights or wrongs of the issue, however there is thrown in prejudice-

    Characterising those that oppose the ill treatment of animals as likely to ‘dig up your grandmother’s grave’ or ‘set fire to your house’ is an irrational ad hominem.
    Despite the medias constant portrayal of ‘animal rights extremism‘ or the buzz word of the last decade ‘terrorists’, rationally you’d expect those that oppose the killing/ maiming of animals to be generally peaceful, non-violent types.

    Of course you can always find the lunatic fringe to demonize a group, as I’m sure the ‘mad scientist torturing vivisectionist’ image is scoffed at by those in favor of animal experimentation, but those in favor shouldn’t fall into the same trap, regarding animal rights campaigners.

    Its often those that support the research that have a mawkish sentimentality about animal, as experiments on primates and other ‘fluffy’ animals, being reduced and regulated whist the ‘ugly’ rodents suffering, does not hit the radar screen (80% if all experiments).
    www.guardian.co.uk/science/2008/aug/13/controversiesinscience.ethicsofscience

    The truth is that most animal research is both pain-free and unnecessary, however we shouldn’t carry out experiments that we wouldn’t ethically approve of when the subject was a human animal.

    I’ve yet to see any explanation of how animal pain is morally different, from human pain?

  21. md1364 said,

    January 24, 2010 at 2:39 pm

    Not so much biased as an attempt to be balanced, I do understand the point of basing beliefs on the results of correctly carried out experiements that are blinded or double blinded and correctly randomised and with the largest sample possible drawn from an accurately representative group, with the intention published before the experiment is carried out. My opening statement was that the examples of homeopaths and other groups, who are attempting (inexplicably) to use the same methods to prove their beliefs and then skewing the results and making a mockery of the whole exercise by being selective in choosing results and not being rigourous in their methods, ARE NOT IN DISPUTE BY ME, just that using the label in this context is unnecessary and feels like preaching to the converted, and the intention of talking to ones selective interest group in these terms is open to question. It is a tactic of rhetoric which, while it will reinforce the beliefs of those who believe already could well alienate any who might be looking in from the outside, therefore doing more harm than good.
    From a detached position it seems to me that this is more than just taking issue with homeopaths methods, because why should they concern you all when they do no good surely they’re just irrelevant, looks more like you guys may have some issues of your own.

  22. ferguskane said,

    January 24, 2010 at 3:48 pm

    Thanks for bringing our attention to this article. I, while not opposing the use of animals in research, am deeply uncomfortable with it. To an extent, I’d say that this is how things should be, and indeed how things are with many animal researchers. However, the article suggests that much animal research is of poor quality. Such research should, in so far as is possible, be abolished.

    The title ‘National Centre for the Replacement, Refinement and Reduction of Animals in Research’ fits with this view and I hope that we do see a reduction in the use of animals in research. The alleged prominence of poor quality research provides a potential mechanism for this aim.

    I’m not sure about the aim of refinement of animals in research though. Is this finishing school, or genetic modification?

  23. reprehensible said,

    January 24, 2010 at 4:15 pm

    V. good article.

    Whilst as Ben say’s it “might be easy to assess whether an experiment was necessary” I would argue that all to often the issue of opportunity cost is also ignored, in animal studies and in general. Even if pharma is doing their studies properly, ultimately it must be questioned what are they for. For example, how many more me-too drugs do we need as a society? or new interventions giving marginal benefit? There is also an issue here regarding who’s opportunity.

  24. Reggie Dixon said,

    January 24, 2010 at 7:52 pm

    @md1364
    I’m sorry to hear about your low opinion of your intelligence but I’ll take that as a “Yes”

  25. md1364 said,

    January 25, 2010 at 6:31 am

    @Reggie Dixon………..if I were to dilute something that many times would it then be as weak as your wit?
    But anyway, I’ve enjoyed the experience here despite my original point being largely misunderstood and being personally vilified by being compared with the lowest forms of life possible (homeopaths and nutritionists}, so really people, can we now allow the subject of the original article to reassert itself in this thread or do we have to keep arguing off subject trying to settle personal scores like a poorly guided high school debating society?

  26. adzcliff said,

    January 25, 2010 at 9:25 am

    Md1364, perhaps we do need to get back to the article in question, but just on your criticism of BG’s ‘throw away’ and ‘by now childish’ comments about homeopaths. There may be some truth in the fact that this point has been stressed to the point of ‘ad nauseum’ for regular readers of this blog. But, I think, you need to remember that this is also a column in The Guardian, where it can’t be assumed that readers are already savvy to the issues that BG campaigns on (likewise, a percentage of readers here will be new). If this comment sent a handful of readers of on a little internet paper chase to work out quite what he was on about, that can’t be a bad thing. I suspect it’s all part and parcel of good popular science writing, that you have to labour your points many times over, on the assumption that a proportion of readers haven’t read your back catalogue. Don’t know if that’s food for thought?

  27. jazwalker said,

    January 25, 2010 at 9:34 am

    @adzcliff

    yes, that’s what i was thinking, although the counterargument, is that by insulting homeopaths without having time in this article to justify it, some people may just feel that these articles are baseless criticism of all alternative therapies, in order to sound controversial.
    though i too enjoy these articles and turn up here every week to read them, i agree that throwing homeopaths in here, felt a little, well, unprofessional. at the risk of inciting godwin’s law (which probably by now wouldn’t be a bad thing), i felt that BG was comparing researchers to homeopaths in much the same way that people compare x to the Nazis…especially when not justifying the comparison.

    i think people are mistaking md’s constructive criticism on the *style* of the article, with the common trolling that can be found round here…and i don’t know if there’s any editors in the audience, but they surely must agree that any author worth their salt, can take a little criticism over style.

    Thanks.

  28. nekomatic said,

    January 25, 2010 at 10:17 am

    Gimpy,

    “it appears to ignore the huge numbers of animals used to create gene knockouts, knock ins, and various other mutations that are then used to provide tissue for later experiments. It also appears to exclude the use of animals in the production of reagents such as antibodies”

    The methodological issues highlighted by this study don’t really apply to that type of research though, do they? If you’re doing stuff to animals to test a hypothesis about what will happen to or in the animals, you need to make sure your test is statistically valid by doing things like randomisation, using an adequate sample size and so on. If the animal is being used to produce a tissue, cell line, reagent etc that’s then used for something else then the animal step doesn’t need statistical validation per se – you would need to verify using a valid test that the product you obtained did what you thought it did, but that would be a separate bit of the experiment.

  29. Trez75 said,

    January 25, 2010 at 12:48 pm

    Sadly I think that this information will only be used as ammunition by those groups against animal testing.

    Rather than it being an indication that the methodology in conducting tests with animal subjects be tightened so that as few animals suffer as is necessary, I suspect it will become “The results of all of the tests aren’t accurate, rendering them useless, so there’s no justification for using animals at all”

  30. gimpyblog said,

    January 25, 2010 at 12:59 pm

    @25.

    nekomatic, yeah you’re right. however, by excluding these uses of animals you are not getting a ‘true’ picture of all animal usage.

  31. outeast said,

    January 25, 2010 at 2:02 pm

    Amimal testing, like nuclear power, is one of those unfortunately essential things I strongly but very unhappily support. This research is deeply depressing.

    @Sili

    I believe it’s not possible to do experiments on animals without first getting permission from them.

    This line had me re-reading several times – I read it (and the later sentences) as suggesting that rersearchers should be carefully pitching animals for the right to experiment on them.

    Your actual point was good, but the unintended one gave me a real belly-laugh. :)

  32. skyesteve said,

    January 25, 2010 at 5:17 pm

    I presume most of us are agreed that any animal experimentation in this day and age must be rigorously justified. By the same token the way such experiments are carried out must be to the highest ethical and scientific standards. Anything less is just shoddy and shoddy science is no better than “bad” science in my view.

  33. md1364 said,

    January 25, 2010 at 5:58 pm

    @adzcliff Point taken, conjurs up images of apoplectic naturopath type clicking on links to be made even more angry by more of BGs columns. I doubt many firm believers will be swayed by what he has to say as the ideological chasm is way too wide….. My real question is why do homeopaths et al try to use scientific methods to explain what they do? I think most people who feel they want to use ‘alternative’ medicine don’t really care about such proving anyway. Ben makes a good point in the Bad Science book when he says that the reality of what happens is far more interesting than the crap they come out with

  34. adzcliff said,

    January 25, 2010 at 8:56 pm

    Cheers for this Md1364. I assume you mean ‘proving’ in the evidencing sense (rather than the ritualistic connotations it has with homeopaths). I think it is a fascinating one, how science has something to say when results are positive, but is narrow-minded and arrogant when negative. It always intrigues me how the likes of Prince Charles can be such an erudite campaigner on environmental science – using scientific data well to inform his argument – but shelves this worldview when promoting his preferential spiritual theories of health, mind & body. Cognitive dissonance, it’s a funny old thing?? Anyway, I’ve got this ten pence piece that can land heads 10 times in a row; 999 of my mates reckon they’ve proved me wrong, but this one mate….

  35. Reggie Dixon said,

    January 25, 2010 at 10:24 pm

    @md1364
    Interestingly you claim that I’m “thick” and weak of wit after slagging off Ben. Enlighten me as to whether attacking people you don’t know in this way works just as well away from the internet.

  36. Niha said,

    January 26, 2010 at 12:53 am

    I was teached about randomisation and blinding at my first year of Psychology degree…

  37. md1364 said,

    January 26, 2010 at 6:34 am

    @Reggie Dixon. OK this really is the last time i will address any comments re my original statement. Your selective reading does rather prove my point about why I would not rate your intelligence very highly, and I did not slag off Ben I merely criticised his use of certain references that I felt detracted from his point by sidetracking. I have read and evaluated and responded to the comments from others here and have taken on board the points which I consider valid and well argued. For your benefit I will explain, I said that IT was childish not that HE was childish, I said that his column/blog displays wit and intelligence and that I greatly enjoy it, I said that there was ONE SMALL criticism.
    Finally, you would appear to know more about attacking people on the internet than I.
    AND NOW PLEASE CAN WE LET IT GO AND MOVE ON SO I CAN GET BACK TO READING THE INTELLIGENT COMMENTS ENTERED HERE?

  38. Camp Freddie said,

    January 26, 2010 at 12:31 pm

    I review the results of other peoples animal tests for classification purposes (according to EU/government requirements for chemical testing). The data is almost all unpublished.
    Since this is seen as mysterious, I thought I’d share some general comments.
    The studies I review are all on ‘lower’ vertebrates (rat/rabbit/mouse/guinea pig). No primates.
    They are ‘standard’ studies on acute, chronic toxicity or carcinogen/mutagen/reproductive toxicity. Chemical safety requirements, not cutting-edge pharma research.

    Unpublished data must usually comply with an international guideline on reporting (OECD). In general, the unpublished studies are far more transparent than published ones. Mainly because an unpublished study takes the form of a large report (30 pages for even a simple study) and doesn’t have to be condensed into a few pages for some journal article.

    All the studies I’ve seen have used randomised animal selection into treated or untreated groups.
    Reporting of the number of animals used is always there, though sometimes you have to check the appendices.
    Blind assessment not 100%, since the studies occasionally label the animal cages with treatment or dose information. Blind assessment is highly encouraged but not a requirement.

    In some cases this isn’t a problem (“is the animal dead?” is not subjective), but for skin reactions or sub-lethal signs of toxicity there is potential for subjective error. The biases will generally be for over-classification, since people might not see a mild reaction in a control group, but will in a treatment.

    The EU leads the way with prevention of unnecessary vertebrate testing, but other countries often permit (or require) tests that I think are unessessary (e.g. testing of two sources of a chemical which are analytically equivalent).

    The ‘standard’ chemical tests are designed to minimise animal use at the expense of accurate/precise data. Generally this works well and people don’t mess around killing 20 rats to find out if the lethal dose is 900 or 1000 ppm when it really doesn’t matter. But sometimes it can be unclear if an adverse reaction is an unlucky result in an N=1 experiment or whether N=1 was sensible in order to prevent unnecessary suffering in a more robust N=3 experiment.
    Also, limit dose measurements can make it hard to assess the safety of mixtures without additional testing (a pure substance is ‘toxic’, but what about a 30% dilution?).

    I find it shocking that higher primate tests could be reported so vaguely. I expected some vagueness in something like lab environment acclimatisation conditions, but not in ‘number of animals’.

  39. iamjohn said,

    January 26, 2010 at 12:57 pm

    Thanks Ben, to steal Dawkins’ claim, this is another oasis of clear thinking. I’m glad that you did make this point (though I would expect no less) and it’s a shame that someone would have to worry about making a criticism of something like animal testing. It seems that this fear, combined with a subconscious bias in evaluating evidence, is what leads to such polarised viewpoints on a range of subjects.

  40. HelenC said,

    January 26, 2010 at 2:22 pm

    Ayupmeduck said ‘let’s see what PETA supporters and their like think of this’. I’m a PETA supporter. We’re all individuals though, so I don’t know whether I’m typical.

    Firstly, thanks to Camp Freddie for shedding some light on unpublished animal studies using rodents. First some questions, Freddie: (1a) are so many experiments done on rodents because scientists are really interested in rodents, or are they really interested in the potential effects of toxic substances in humans? (1b)If it’s the latter, then – even if studies are correctly carried out and reported – how useful is that data in understanding how these substances affect humans? I assume that whilst there are some similarities between us and rodents, our immune systems etc will not be identical? (2) Do unpublished reports contain details of the conditions in which animals are kept and what is actually done to them? Ben states in his article that ‘it’s easy to assess if animals are treated well’. I guess it is if you have access to the laboratories, but otherwise we’re all a bit in the dark.

    One reason why many people distrust animal experimentation is that, even when experiments are properly reported, the language is so sanitised that it is difficult to tell exactly what has been done physically to an animal to produce the results.

    I know that I can’t live a full and healthy life without some degree of medical intervention. I also realise that some degree of secrecy is needed to protect scientists from criminal activity. But I need to be able to make informed decisions. I’ve never bought cosmetics or cleaning products that have been tested on animals because I don’t think our desire for smooth skin and shiny pans is worth causing suffering and death to an animal. Similarly, I would like weigh up the symptoms of an illness against the animal suffering caused by developing a potential treatment. At the moment I don’t have enough information to make that decision.

    Please note – however rude the responses are, I promise I will not burn down Ben’s house or dig up his Grandma.

    Cheers me dears,
    HelenC

  41. skyesteve said,

    January 26, 2010 at 5:37 pm

    completely off topic but this is priceless!

    www.mydavidcameron.com/

    My favourite so far being “too many tweaks make a twat”…

  42. heavens said,

    January 26, 2010 at 8:03 pm

    I’m with Camp Freddie: What is *wrong* with these people that they aren’t randomizing their test subjects? It’s EASY! It’s FREE! It’s NORMAL! It’s NECESSARY!

    I can’t even imagine what the techs would say if I told them not to “bother” randomizing the animals (mice and rats) for even the least important little study. They might well call the local hospital to have me evaluated for dementia.

    We choose mice randomly even for the rare n=1 pilot study (usually, “If you feed this to a mouse, does it promptly fall over dead? No? It lived through the night and looked okay this morning? What a relief. Okay, the real study is going to by n=20, and…” — because we don’t want to accidentally poison 20 mice if there’s any way to avoid it). “The first one you catch” might be bigger, slower, or sicker than average.

    It’s true that we don’t always manage perfect blindness, because the tech who is doing the injections is required to report any concerns that he or she sees to the vet — IMO every person in the building, including the cleaning staff, is subject to ‘mandatory reporting’ — and we’re too small to always have separate employees preparing and giving injections, but I’ve never sent an unblinded sample in for histopathology or even blood chemistry work.

    But, still — not randomizing? Not even at the level of drawing cards or little slips of paper out of a pile? That’s pretty pathetic.

    (I think I’m going to cling to the hope that randomization is so ingrained and so patently obvious that it didn’t seem necessary to mention it in the paper, rather than it not actually being done.)

  43. Reggie Dixon said,

    January 26, 2010 at 10:24 pm

    @md1364
    No actually, I think homeopathy when claimed to be more than placebo is worthy of nothing other than ridicule and wonder why anyone would object to this. Again you make another unsupported assertion about me. Do you find this works well in person ?

  44. PsyPro said,

    January 27, 2010 at 7:51 am

    There are two issues here: (1) the actual use of non-human animals in the research, both with respect to numbers and what was actually done to them, and (2) whether it matters. In the absence of a compelling argument for (2), (1) hardly matters. So, the real issue is (2), not (1), at least as I see it. And, unlike some popular philosophers (Singer), I do NOT think (2) has been settled in anyway that makes the debate about (1) the real issue.

    To forestall the usual nonsense, yes, I m an animal lover: I have numerous pets (usually birds and dogs) who populate my house and cottage. I do not without reason kill non-human animals, including mice in the aforesaid cottage, despite their ubiquity. But, none of that has any bearing on the issues here. Descartes routinely dissected live dogs and cats (as he believed their screams were merely “mechanical”, not true suffering). Although I cannot do as Descartes for my own state of mind (whether suffering or not: I cannot tear up the favourite stuffed toy of my son, even though I know the toy does not suffer), I am not convinced he wasn’t right (for all the wrong wacko religious reasons). I can see why we might want to avoid causing pain to some non-human animals where we reasonably can, I am less convinced, as with Descartes, that suffering is ever a non-human attribute.

    So, t get back on topic: why do we care? Not sure we do. So, 14, uh, 12, uh, 8 aside from being stupidly sloppy as per the science may not be any real moral issue.

  45. HelenC said,

    January 27, 2010 at 10:25 am

    Well, PsyPro, if an animal has a central nervous system, then I think we can infer that they feel pain and are capable of suffering. I interpret their yelps/howls/whines etc as communication of this internally felt pain – in the same way that I would interpret children’s cries, howls, screams etc. Perhaps you don’t think children feel pain, either, as they are not as fully developed as adult humans?

    Why do you offer the fact that you keep pets as evidence that you are an animal lover? I don’t keep any animals as pets, because it strikes me that arrangements are always stacked in favour of the owner rather than the pet. I offer my lack of pets as evidence of my ‘love’ of animals, i.e. my concern that they should be free to exhibit the full range of their innate behaviours, unrestricted by human intervention.

    Returning to the main points of my first post, I would very much like Camp Freddie or heavens (or anyone directly involved in animal experimentation) to let me know how useful these experiments are in terms of understanding human patholody, and what are the conditions in which animals are kept.

    Cheers,
    HelenC

  46. HelenC said,

    January 27, 2010 at 10:27 am

    Sorry – the word in the last sentence should, of course, be ‘pathology’.

  47. adzcliff said,

    January 27, 2010 at 11:36 am

    Hi PsyPro.

    Descartes may have been ahead of his time in many aspects of science, philosophy and mathematics, but he was also capable of some howlers (as it turns out). Don’t forget it was he who opined – on what evidence, I’m not sure – that the pineal gland was where the physical body and spiritual mind ‘all came together’. I think I’m with HelenC when she suggests that a nervous system might be a good indicator of an organism’s suffering capabilities (for good evolutionary reasons).

    Cheers

    Adzcliff

  48. Camp Freddie said,

    January 27, 2010 at 11:54 am

    A few answers to HelenC:
    I should point out that I don’t do the studies or work with people that do, I just read the reports from testing companies, commissioned by some manufacturer. Therefore I can’t say much about what actually happens day-to-day in a lab.

    (1a) are so many experiments done on rodents because scientists are really interested in rodents, or are they really interested in the potential effects of toxic substances in humans?

    I deal (mostly) with the registration of pesticides, but occasionally other chemicals. I don’t do drugs or biological research.

    The studies I see are primarily interested in potential effects on humans, but the data are also used for ecotoxicological risk assessment (e.g. if we spray this insecticide will it kill all the field voles). Rodents are used because they are very well understood biologically and they have suitable lifespans and breeding.

    (1b)If it’s the latter, then – even if studies are correctly carried out and reported – how useful is that data in understanding how these substances affect humans? I assume that whilst there are some similarities between us and rodents, our immune systems etc will not be identical?

    The basic tests are for lethality and irritancy. Eye irritancy is generally the same for humans and animals since our eyes really aren’t very different. This is a 1-3 animal experiment. If the first animal shows a serious reaction or pain, the test is stopped. If it’s OK, 2 other animals are used to confirm.
    Skin irritancy is similar. Furry animals have thinner skin than us, so animals give a worst-case result.

    Lethality can be roughly extrapolated between species, scaling the result according to body weight or volume. However, the purpose of the tests (these days, not always for old studies) is to fit the chemical into a rough hazard category (i.e. if I swallow it, should I phone an ambulance) rather than a specific ‘X grams will kill a human’. Basically, you treat a small number of animals at a high dose, if they survive it’s non-hazardous. If they die, you treat at pre-set lower doses until they survive. You end up with something like “Rat oral toxicity = 500-2000 mg/kg, therefore the chemical is harmful by ingestion”. This classification triggers various safety requirements depending on the use for the chemical (see COSHH regulations in UK).
    These tests decide the labels on chemcials that you’ve probably seen, the jolly rodger or black cross on an orange background.

    For pesticides, there are extra tests for chronic toxicity, which is used for determining safe exposure levels for workers or consumers. This looks at a number of issues, like potential mutagens, carcinogens, teratogens and more obscure things like endocrine disruption. I’m not an expert on chronic toxicity, and you need to be one to be taken seriously. No one wants to be responsible for a thalidomide-type disaster. Tests normally use mice, but rats and sometimes even dogs are used. Primate tests would be very unusual (I’ve never seen one).

    (2) Do unpublished reports contain details of the conditions in which animals are kept and what is actually done to them? Ben states in his article that ‘it’s easy to assess if animals are treated well’. I guess it is if you have access to the laboratories, but otherwise we’re all a bit in the dark.

    Yes. Housing and environment conditions are reported. Size, materials, numbers in a cage, temperature, humidity. Treatment and assessment data are reported for each examination. Behavioural abnormalities are assessed and are a very important part of judging if further specialist testing is needed.
    Generally a lot of effort goes into keeping the animals healthy (other than the obvious dosing of them with a potential poison!). A company wouldn’t be happy if they were told their chemical is a suspected neurotoxin due to odd rat behaviours, if it was actually just stressed because of noisy building work next door. I’ve only visited one UK lab and it seemed very hygenic and modern. I really can’t say if this is normal or not.

    Lastly, I should point out that testing is generally avoided where possible in the case of general industrial chemicals. Don’t assume that every bottle of chemical has been tested on animals. Old data or general historical experience is often used and mixtures of chemcials are normally extrapolated from their ingredient toxicity rather than given a whole new set of animal tests. People are more wary about pesticides, so limited testing is normally done on formulations, especially if a mixture is in a different form to its ingredients (e.g. where a powder is formulated as an emulsion).

    If you want some info on the simple tests, the OECD guidelines are here:
    oberon.sourceoecd.org/vl=1624555/cl=31/nw=1/rpsv/cw/vhosts/oecdjournals/1607310x/v1n4/contp1-1.htm

    Please excuse the inappropriate smiley face symbols that OECD’s library uses to denote that a guideline is available for download!

  49. Camp Freddie said,

    January 27, 2010 at 12:17 pm

    Shorter version:
    Basic tests are required by government to decide basic safety requirements for chemicals, where no data currently exists. They are very useful for humans for things like “Is it safe to handle?”, “What if my kid swallowed some?” or “What happens if I get it in my eye?”. They can’t be used to say, “Luckily you only swallowed 10 grams. If you’d swallowed 15 grams you’d be in trouble!”.
    They are also useful for ecotoxicological risk assessment, “Is it safe to spray on a field or does it harm the wildlife”.

    I can’t honestly say how useful the more advanced tests are for extrapolation to humans, as I am not an expert in them.

    I can also mention that for environmental risk assessments, studies are done on fish (and bees, birds, insects for pesticides) purely for fish and bird protection (i.e. we don’t extrapolate to humans).

    The dividing line for whether a study is protected by animal rights legislation is generally “does the animal have a spine”, so fish studies are only allowed ‘where necessary’, algae and bees can be tested at will.

  50. gimpyblog said,

    January 27, 2010 at 12:33 pm

    HelenC, why do you think that understanding human pathology is the only aim of biology? What about our uses of animals for food, clothing, entertainment, companionship, etc?

    But, evolution is why we study animals to gain understandings of ourselves. Many biochemical pathways and gene interactions are conserved across species.

    And why would you allow animals “to exhibit the full range of their innate behaviours, unrestricted by human intervention” but not extend that courtesy to fellow members of your species who want to use their innate curiosity to understand life?

  51. Camp Freddie said,

    January 27, 2010 at 12:55 pm

    I just noticed that the OECD link has all their toxicity guidance. Only Methods 402, 403, 404, 405, 429 and 425 are basic tests.
    Method 425 probably gives the best insight into how and why chemical safety tests are done.
    Tests 401 and 406 are rarely used anymore.
    401 was replaced by 423 and 425, so as to reduce numbers of animals (rats) used.
    406 (guinea pigs) was replaced by 429 (mice) due to 429 being more reliable/accurate.

    No one has brought up computer modelling yet. Basically, this is allowed in theory but is generally very unreliable or even impossible.

    As a way of avoiding testing, it is quite common to get an expert toxicologist to argue that if butyl-chemicalidine and heptyl-chemicalidine were tested and found to be irritant/toxic, then our new compound hexyl-chemicalidine will also be irritant/toxic due to its shared/similar funtional groups. Therefore no additional tests are necessary and are considered to be contrary to Directive xx/yy/zz Article 19b(iv) on the use of vertebrate animals.
    (The actual arguments are a bit more long winded but this is the gist)

  52. HelenC said,

    January 27, 2010 at 2:14 pm

    Thanks for your replies, Camp Freddie. Very helpful, and I will have a look at the OECD link.

    Hi gimpyblog, I focussed my questions around human pathology because that is the greyest area in terms of understanding the actual benefits of animal research.

    As far as food, clothing, entertainment and companionship are concerned, there are some very good alternatives to animals!

    Of course we must exercise our human curiosity, providing that it happens within ethical limits and – where animal suffering is involved – with demonstrable benefits. It’s these limits and benefits that need to be properly debated in a dispassionate and informed way. That is unlikely to happen given the media’s aversion to facts, the strength of the industrial lobby and the apathy of the majority of our citizenry.

    In the meantime, when it comes to decisions about whether to use or buy something, I have to resort to my own makeshift cost/benefit analyses based on incomplete information. Because the PETA bunny logo doesn’t appear on tubes of topical steriod cream, for example, I have to assume that its testing and manufacture has caused extreme animal suffering. I therefore have to ask myself, ‘is the alleviation of eczema symptoms worth the suffering and death of numerous animals?’ From my perspective, the answer is ‘no’. If I had a more serious medical condition would I think differently? It would be a fairly difficult call even with all the facts, but without them it’s impossible to say.

    Cheers,
    HelenC

  53. merrick said,

    January 27, 2010 at 3:37 pm

    Glad to see we can take the actions of a tiny minority of violent extremists to discredit the entire body of anti-vivisection thought.

    I’m off to my local mosque to blame them for 9-11.

  54. ferguskane said,

    January 27, 2010 at 4:18 pm

    @HelenC.

    The question of the costs and benefits of animal experimentation are is a very good one, and one that I wish I could answer.

    My gut feeling is that certain animal experiments have been invaluable to the development of medical science and have presumably saved millions of lives. At the same time, I have the impression that if the ratio of good research to poor research is similar across scientific fields, there will be a lot of animals experimented on without good reason.

    Animal experimentation appears to have been important in the development of penicillin and our treatment of diabetes, to name a couple of examples. Equally, can we say how our knowledge of the workings of the nervous system would be different/delayed if Galvani had not experimented with frogs? And further, without experimenting in the past, would vets have the same ability to do good as they do now?

    So a challenge: Can anyone point us to a good, balanced exploration of how animal experiments have contributed to our present day scientific knowledge?

    As for PETA. At the moment, the thing that upsets me more than animal experimentation is the way we treat non free range animals. I wish PETA would focus their energy on that. (I’m not a vegetarian, but I do care)

  55. Moeche said,

    January 28, 2010 at 11:25 am

    I would like to add my point of view to Ben´s comment on this study published in PLoS ONE.
    First thing is that Ben´s conclusion from the study is that it reports on bad experimental design and interpretation of the data. But this is not totally correct. The study, as it was set up, was analyzing the reporting of animal testing. If you check the actual paper you will realize that the effort was put on analyzing “reporting”, even though they talk about “quality of experimental design”.
    Does it make any difference? Well, I think so. As stated by Ben “Some of the flaws they discovered were bizarre”. These “bizarre flaws” might reflect a wrong approach to the papers scrutinized in the study. And the “bizarre flaws” should ring a bell. Most of the studies did not use randomisation and blindness? Or they did not report it? Or the journal asked the authors to take all that away from the final published version? Or they are obviously there but the authors failed to recognize them? There are a number of articles that did not even stated clearly what was the aim of the study in the introduction? Honestly, I have never in my life found an article not stating the aim of the research in the introduction. Might be this a case of bias interpreting the results by the authors of the study? But of course the authors did not explain how they assessed this point. It is nowhere in the paper and it is an obvious example of subjective interpretation.
    I am a scientist working on biomedicine using animal experimentation (mice) and in my experience, bad reporting of methods used in a study is a general flaw on published research of any kind. This is a problem that is even increasing and for which the authors are not the only ones to be blamed. Academic journals have a lot of responsibility on this as well for not reviewing carefully the accurate and complete reporting of the methods used, and actually most of the times, discouraging it.
    Having said that, it is the editors and the reviewers task to assess quality of experimental design and interpretation of the results. Basically the same “flaws” can be observed in loads of articles published even in the big journals (Nature & Science) but not just on those ones dealing with animal experimentation. This means that there might be a problem with accuracy and completeness of methodological reporting, not that animal experimentation is particularly sloppy and inconsistent. Actually, the authors of the study could (should?) have compared reporting of a different experimental methodology to say something about how does animal experimentation compare to other methodologies. And they could have examined if there is a link between flaws and quality of the journals in which the papers were published. Are we talking here about science published in obscure very specialized journals?
    Finally, Ben states that the study is pretty solid but for one thing I missed the list of the actual papers under scrutiny. I guess there are pretty obvious reasons not to give details of what papers did the whole thing wrong but yet … it is a crucial point in the Methods section of this paper that should be included if anyone wants to check and reproduce their results. Does this make it a solid study? Well, in my opinion this is at least as bad as what the authors are trying to criticize. But we should not forget that these very same authors might have a clear conflict of interests, since (at least some of them) make a living of teaching how to design and assess animal experiments.
    This study, and the way it has been covered by Ben, IMO gives the false idea that animal experimentation is poorly designed, ridiculously conducted, badly interpreted and consequently, useless. Providing, in case they needed it, an extra (false) argument for people opposing animal experimentation. All this by means of a poorly conducted and reported study.
    Sorry for the looong comment.

  56. HelenC said,

    January 28, 2010 at 12:07 pm

    Yes, good point, Ferguskane. Because PETA are a global organisation, they focus on the worst abuses of animals globally and – despite the poor conditions of a lot of UK non-free range animals, I guess they are not the worst. Having said that, I agree that getting Selfridges to stop stocking fois-gras (is that how you spell it?) is probably not the highest priority, although it is a good headline-grabber! How many people can afford to buy it anyway?

    Several factors seem to affect people’s willingness to look for other sources of protein. Highest is probably financial constraints, and the water is further muddied by terminology such as ‘high quality’ and ‘low quality’. Most people don’t realise that ‘high quality protein’ is simply that which contains all of the essential amino acids, rather than a comment on the quality of the food itself. Even when I’ve explained protein combining to confirmed meat eaters, they remain unconvinced.

    Anyway – yes, let’s see if anyone is able and willing to take you up on your challenge!

    Cheers,
    HelenC

  57. gwenhwyfaer said,

    January 28, 2010 at 11:04 pm

    @md1364:
    Sorry to come back to your comments, but it occurs to me that there might be an additional dimension than just taking puerile digs at the homeopaths. The thing is, they’re also rather litigious, and that litigation is very damaging to science writing as a whole. It may be that having won against the notoriously litigious Matthias Rath, Dr Goldacre quite fancies his chances with the homeopaths too – and such digging is an open invitiation to “come and have a go if they think they’re hard enough”, with an eye at some point to getting the whole field of scientific discussion declared off-limits for libel. If he’s doing that, presumably it’s with the backing of the Scott Trust…

    Just a random thought. I don’t know anything about anything, of course. But I thought you might appreciate a response from someone else who disagreed with you in a mature, (hopefully semi-)intelligent way.

  58. heavens said,

    January 29, 2010 at 12:27 am

    HelenC,

    In rejecting the eczema cream, you seem to have missed the fundamental cause-effect issue: The same number of animals are treated the same way regardless of whether one thousand, or one million, people have their symptoms relieved.

    A more reasonable question to ask yourself is “Given that these animals are already dead, am I really helping them by avoiding this product?”

    Drug testing doesn’t scale like leather shoes or eating meat: Using less doesn’t mean any animals are protected.

  59. DavidGSFarmer said,

    January 29, 2010 at 11:21 am

    @Mosche
    Excellent points. The distinction between ‘not reported’ and ‘not carried out isn’t really made as clear as I would like here.

    Your question on the quality of the journals is addressed by the authors:

    “In fact, the search retrieved papers from a range of publication
    years (1999 – 2005), covering a wide variety of research areas, and an extensive range of journals across the impact factor spectrum, including Nature and Science. Whilst it would be useful to know if there is a relationship between the quality of the papers surveyed and the impact factors of the journals they were published in, this analysis was not in the remit of this survey.”

    Damn straight it would be useful. I think the importance in the distinction between quality of reporting and quality of research has been adequately made here. The authors justify the link between the two as follows:

    “…this kind of reporting omission can easily be rectified. But if not, incomplete reporting masks potentially flawed experimental methods.”

    Potentially, but I think I’d agree with Mosche in that I’m unsure how valid the link between what the authors consider to be ‘bad reporting’ and the validity of the experiment is, particularly when the authors state that, for example, if the assessors did not understand the aim of a study then it was reported as not having an aim listed. They justify this by saying that this means that non non-specialist would understand the aim of the study but, given that this is not a text book, why should that reduce the relative validity of a study? Communication of science is great and all, and should absolutely be encouraged and celebrated (avid reader by the way, Ben) but specialists are the ones who will read the research, assess it in context and conduct further research accordingly. The fact that it can’t necessarily be understood by professors of political science doesn’t mean that it wasn’t appropriate and necessary to use animals in the study.

    By Mosche
    “…I missed the list of the actual papers under scrutiny. I guess there are pretty obvious reasons not to give details of what papers did the whole thing wrong but yet … it is a crucial point in the Methods section of this paper that should be included if anyone wants to check and reproduce their results. Does this make it a solid study? Well, in my opinion this is at least as bad as what the authors are trying to criticize.”

    Boom.

  60. HelenC said,

    January 29, 2010 at 12:25 pm

    Hi Heavens,

    No – I’m not missing the fundamental point. The same number of animals suffered during animal testing of cosmetics, irrespective of how many people bought them. The groundswell of opinion against animal testing of cosmetics resulted in the development of animal-friendly products by companies who shared those ethical values and saw a business opportunity. Once a sufficient number of consumers moved their custom to the animal-friendly options, some of the mainstream cosmetics companies revised their practices. This shift in buyer behaviour, coupled with legislation, has helped us get to the point where there is no need for anyone to buy cosmetics or household products that have been tested on animals at any point during their manufacture.

    If I use drugs that are tested on animals, then I am helping perpetuate animal-testing – however small my contribution. The reason I think it is important that we understand the cost of particular products in terms of animal suffering is so that we can make a balanced choice.

    The argument you put forward is the same as the one many people use for buying air-freighted, out of season fruit and veg from supermarkets. ‘Well, if I don’t buy it, it will go to waste, and the same number of airmiles will have been used anyway’. Yes they will. But if supermarkets are left with rotting fruit and veg because consumers refuse to place their desire for choice above concern for the environment, then supermarkets will eventually take notice and stop.

    Have a good, animal-friendly weekend.
    HelenC

  61. Arthur Dent said,

    January 30, 2010 at 10:59 am

    HelenC said “there is no need for anyone to buy cosmetics or household products that have been tested on animals at any point during their manufacture”.

    It is true that in the EU the Cosmetics Directive prohibits the testing of cosmetics on animals. However most of the ingredients currently used in cosmetics have already been tested at some point on animals. Secondly, you will find very few ‘new’ ingredients being used in cosmetics other than those that have been tested on animals for other non-cosmetic purposes.

    Finally the REACH regulation in effect requires all substances manufactured or imported into the EU for any purpose (including cosmetics) to have relevant data which in many cases can only be obtained by animal experiments. The European Parliament has not yet resolved the inconsistency between these two legal instruments but then a high level of cognitive dissonance is a requirement for all MEPs.

  62. HelenC said,

    February 1, 2010 at 9:20 am

    Hi ArthurDent,
    Yes – I’m aware of the inconsistencies in EU regulations. That’s why PETA only approve of products where the ingredients themselves haven’t been tested on animals, as well as the final product. The PETA logo is the only guarantee of cruelty-free status. I still stand by my comments that there is no need to use cosmetic and household products that have been tested, but you need to go further afield than the local supermarket. Ecover don’t pass muster, for example, but Bio-D do.
    Cheers,
    H

  63. Arthur Dent said,

    February 1, 2010 at 9:45 am

    Hi HelenC I am fascinated that PETA is able to recommend substances that have not been tested on animals, do they provide a list of such substances?

  64. CaptainKirkham said,

    February 1, 2010 at 3:03 pm

    It’s just a shame that PETA can’t do that sterling work without being misogynistic and racist at the same time.

    There are other animal rights organisations that (a) aren’t violent and (b) aren’t PETA. I wish they could get more media attention, but unfortunately the media aren’t interested in sober and useful debate, they’re more interested in PETA’s obsession with naked women.

  65. Arthur Dent said,

    February 1, 2010 at 4:20 pm

    Further to my question to HelenC I have tried to follow up what PETA actually say. As far as I can see, and I would be happy to be corrected, PETA do not list iny ingredients that they consider to be untested on animals but instead list companies that have signed their agreement:

    Companies listed either have signed PETA’s statement of assurance or provided a statement verifying that they do not conduct or commission any animal tests on ingredients, formulations, or finished products and that they pledge not to do so in the future

    This may be a step forward for those who wish to use ‘cruelty free’ products but doesn’t come anywhere near a statement that ingredients have not been tested on animals. The assurance merely means that the company you buy from hasn’t tested anything on animals not that the ingredients in the things you buy from them haven’t been tested on animals by other people.

    I would be very surprised if many of the ingredients in cruelty free cosmetics and household products have not, at some point, been tested on animals.

  66. HelenC said,

    February 2, 2010 at 1:42 pm

    Hi ArthurDent,
    To be approved by the PETA ‘leaping bunny’ scheme, companies and their supply chain must not conduct or commission animal testing on any of their products or the consituent ingredients. They must also agree to submit to independent audit.
    More information at www.gocrueltyfree.org/e_retailers.php.
    You can also email PETA direct at info@peta.org.uk. You’ll get an auto response saying it may take up to two weeks for them to reply, but in practice I have found them to be much quicker. I did look into this before signing up. I think everyone appreciates that we’re reliant on companies being honest, but that holds for most things.
    I also sometimes email companies direct for info re their company policy, products, ingredients etc if they’re not on the PETA list.
    H

  67. lizzi_fish said,

    February 2, 2010 at 1:56 pm

    I might be a bit late joining the comments here, but to answer the question right at the top of the page from ChrisP; a laboratory has to complete a Home Office return at specific intervals throughout the year to report how many animals were ordered from breeding facilities and how many were used – I believe this is part of the (3Rs) reduce, refine, reuse strategy for using less animals in testing.

    I used to be an animal technician, and did the job because I love animals (surprisingly enough, a lot of people do it for that reason). I worked in a lab that was regulated by GLP – e.g. not a research facility, and I completely agree with Ben when he says we have some of the strictest regulation in the world. I also took some quite serious abuse from animal-rights activists in my time, and found them to be ill-informed, reactionary and attention-seeking. But maybe that was just me. They weren’t even from PETA.

    I can go into more detail on the subject if anyone is interested; I can type at length about animal testing, animal rights and the pharmaceutical industry from personal experience.

  68. proveyance said,

    February 4, 2010 at 2:53 am

    First response/post here.

    This site seems like a great resource. I just wanted to chime in after reading the first few comments to this post.

    Since I can remember (and I do eat meat and support hunting and use of animal materials), the notion of man having dominion over every other animal was disturbing and clearly a product of the ego.

    My suspicion (I prefer not to believe things) is that it’s doing more harm than good to experiment (which is light torture imo) on animals. Although I don’t know much about this, I don’t think you have to be in the “expert class” to know that it’s karma of the worst kind.

    I know, all the great advancements and the lives saved. But that implies that a human life is some how important, while animals are relegated to serve us. Take a dog, for instance. A dog is very willing to help a human, as are humans willing to help dogs. But, don’t they deserve to be willing participants. If an animal could comprehend the situation, I’m sure some would offer to assist. That’s the real tragedy.

    What’s with zoos, while I’m at it? I hear that, despite how great they are some are failing in this economy. Are you wondering for even a moment what they are doing with the “stars of the show”, now that the show has been canceled. I digress, and apologize for leaning towards ideology.

    What does science have to say about animal rights? I think it is not their expertise. Yet, they are charging ahead. But hey, that’s society as the collective intellect expands and then shrinks.

    How about this? I encourage anyone in favor of animal experimentation to attend a native American ritual vision-quest. If, afterward they are still in favor then I have more to learn. But, we are all free to do as we will.

    That wasn’t bad. I look forward to more peeking around here. I fear what I will find in the areas of false-flag terror, and terrorism for that matter. And why do suicide bombers blow themselves up? Is it narcissism?

  69. ocswimwear said,

    February 4, 2010 at 8:19 am

    I often wonder if there would be as much of an issue with animal testing if the animals used were ugly bugs; as long as cute fury animals are used a large number of people will never accept it. If anything the issue just highlights how fickle and shallow the human trait known as sympathy actually is.

  70. mickjames said,

    February 6, 2010 at 2:27 pm

    Is it possible that once researchers have broken free of the moral and sentimental constraints that would prevent most of us carrying out animal research that other values begin to erode as well? That their view of what is “necessary” allows them to push the boundaries of what is “acceptable”? Nowadays (pace Wakefield et al) people are much more restrained in the experiments they are allowed to do on children and live human subjects but this was not the case in the past. My (pretty cursory I admit) reading on the subject suggests that in these cases as well that the exhilaration of the transgression leads to some pretty cavalier behaviour and that these studies, for all the pain and suffering involved, end up as virtually useless. Worth investigating?

  71. mangie said,

    February 7, 2010 at 1:52 pm

    Ben’s article is important. It is about how good science shoots itself in the foot.

    @Keith19 said

    “In my opinion this is one of the areas that the pharmaceutical industry is leading academia in basic science animal research. Randomisation and assessing statistical power before an experiment is carried out is the norm……This is presumably because mistakes at this point in the research, either false positives or negatives, can end up costing the companies millions.”

    That is correct. To take that forward, there are three comments I would like to make:

    a) The pharmaceutical industry behaves in a few instances in a terrible way once those millions are committed, and allows human clinical trials to be corrupted. Exactly the same as the animal scenario where poor practices allow a whole mode of research to be attacked.

    b) As the in animal welfare problem the deviants are not properly punished in the interests of the non deviants.

    c) Many of these pharmaceutical animal studies are not published. The TGN1412 disaster is an example of where that leads. No matter how good these studies might be in informing industry where to risk their billions, they are therefore not part of science.

  72. anticarrot said,

    February 11, 2010 at 4:49 pm

    Animal Testing is a lot like Abortion. One side prefers the term ‘experimentation’, while the other prefers ‘torture and slaughter’. Each side thinks their term is more acurate than the opposition. Both siddes also forget the terms are not mutually exclusive.

    >Animal experiments are necessary
    >But it’s easy to assess whether animals are treated well
    >Only 12% of the animal studies used randomisation.
    >Only 14% used blinding.
    >People who threaten the lives of 15-year-old girls

    The problem starts when the pro-testers don’t live up to their claims. By the figures above, hasn’t PETA has been *entirely* *correct* 86%-88% of the time when they claim testing is evil and pointless? Pity Mr Goldacre can’t bring themselves to considder this.

    There is also the recent horror story about fox hunting. No, not the hunting, no not the banning. But the serious questioning that foxes might suffer if they were running for their lives. While I agree that you can easily measure suffering, it is also blatantly obvious that when large sums of money are at stake, a large percentage of people chose to lie. Strange that Mr Goldacre is seemingly unaware of this very recent and well publicised ‘contraversy’.

    But not really surprising. If you gave them a choice between conducting a truely necessary but harmful experiment on either a group of chimps, or a group of severely mentally ill orphans, they’d all choose the chimps. They would maintain this choice even if you gave them ccompletely solid evidence that these specific humans were no more intelligent than the chimps. (This is what is known as a ‘special case’.) Even if you pointed out the human testing would be much more accurate. And the problem with the reasoning they’d give (humans are special) is that it is basically alternative medicine – that which cannot be proven to be true, or that which has actually been proven to be not true.

    As to 15 year old girls… I’m shocked. Really I am. Aren’t there several important queestions we shoudl be asking before we make succh sweeping generalisations? Have we reduced the number of participants to the bare minimum? Can we reduce the suffering by refinement? Is it possible to replace this with another method? Well, the answers are no, no, and no. It’s down to one and nothing else has worked. Surely therefore by your own standards…

    Incidentally, the 4th R really should be ‘redirection’. Are you flogging a dead horse so hard that you are draining funds from more worthy research? For example many cancer treatments are not chemical (drugs) but rather physical (radiation) and mechanical (microsurgery). Given that cancer can happen in a thousand different ways (or more) then killing thousands of animals each year for each specific variation is a very questionable practice, when the alternatives offer solutions that are potentially far more generic, quicker, and cheaper.

    When scientists reject science before they even start experimenting on animals (which they do, as demonstrated above) then the results are going to be screwed up and unscientific. Which again brings us back to PETA being right the vast majority of the time.

  73. dankat123 said,

    February 14, 2010 at 12:34 am

    Serious moral issues deserve serious moral consideration, not the superficial treatment contained in the article, informed, apparently, by the author’s personal history. How disappointing.
    Making use of the research evidenced considered in the article, one might reason as follows (taking as a starting point that many will consider contentious, but that is a separate discussion):
    1. Animals differ from humans in a morally relevant way. From this one can justify the claim that experiments on animals are morally permissible;
    2. As a matter of contingent fact our present state of knowledge and technical ability means that experimentation on animals offers the best prospect for understanding, treating and curing a number of illnesses etc. From this, and given 1 above, one can assert that experiments are practically necessary.
    3. Although animals are not morally equivalent to humans, they have a capacity to feel pain. From this one can claim that animals are morally relevant agents.
    4. From the above it follows that as moral agents humans have a moral duty to minimise that suffering that animals suffer in experiments.
    5. From the above it follows that as moral agents humans have a moral duty to find alternative methods to aid us to understand, treat and cure illnesses i.e. to make 2 false.
    6. From all of the above it follows that if it is true that the “reporting and conduct of animal research is often poor” then those scientists responsible for the poor reporting and conduct are failing in their moral duty.
    7. It is obviously true that “people who threaten the lives of 15 year old girls are beneath contempt” and that “people who fail to damn these threats are similarly contemptible”
    8. It is equally obviously true from the above that people who fail in their moral duty to minimise the suffering of animals in experiments are beneath contempt and that people who fail to damn these failings are similarly contemptible.
    9. It is indeed true that some growing up is needed.

  74. toni49 said,

    February 19, 2010 at 11:28 am

    I think that what is being essentially missed by a large number of people here is that the article isn’t actually about the ethics of animal testing. What I believe the author is trying to say effectively follows the adage “get your own house in order”. In order to have stand up to anyone who doesn’t believe in animal testing, researchers need to ensure that their work is of the highest possible scientific standard. You can’t possibly have a reasoned debate stating that the work is necessary if there is a question over whether it properly follows basic scientific principals.

    In the same way that nuclear power stations are expected to have better health and safety records than any other industry in th UK simply because of public perception, so too should animal testing have squeeky-clean methods.

  75. Ben Goldacre said,

    February 19, 2010 at 1:42 pm

    something the antivivisection people seem not to have noticed is that if this study has found experiments are being done with too few animals to be useful, this may mean that more animals are needed to do decent research.

  76. moonbug said,

    March 6, 2010 at 1:01 am

    @Ben Goldacre

    I sat on a university Animal Experimentation Ethics Committee (AEEC) as a Animal Welfare representative about 7 years ago in Australia. I know at the time several university AEECs had statisticians either sitting on the committee or advising them because of several of the points you make regarding using a statistically valid sample. As an animal welfare rep I was one of the “antivivisection people” and it is a concern that many people (both “pro” and “anti”) have.

    Of course the papers do like to pick up on the extremist vibe, there are alot of people who are involved in the movement against animal experimentation who have strong scientific backgrounds who do argue against animal experimentation on grounds of poor science, which has been highlighted by the research in the post.

  77. Ben Goldacre said,

    March 6, 2010 at 1:24 am

    i don’t doubt that there are soundminded people who have valid concerns about the extent of animal experiments, or even a consistent moral standpoint that it is never acceptable, who would never resort to violence. what does concern me very deeply is that when i asked, for example, www.buav.org to point me at where they had come out clearly and condemned violence committed in the name of their cause, they were unable to do so.

    if there were armed guerrilla’s violently attacking people in the name of evidence based medicine, i would be the first to stand up and condemn them, distancing myself from their activities, and perhaps even trying to open a public and transparent dialogue, arguing with them to change their methods. this seems not to be the case with mainstream antivivisection campaigners, and to that extent, i find them very unpleasant.

You must be logged in to post a comment.