gr8tt

Bernard E. Rollin, Ph.D.
University Distinguished Professor, Professor of Philosophy, Professor of Animal Sciences
Professor of Biomedical Sciences, University Bioethicist
Department of Philosophy, Colorado State University

In part as a result of an ideology affirming that science has nothing to do with ethics, 20th century science has enjoyed an abysmal track record in engaging issues that are of great importance to society. This is true regarding every ethical issue occasioned by scientific activity. Even in the area of research on human beings, the research community has been extremely cavalier about ethical treatment of research subjects, despite the fact that every citizen should have, by the time they reach adulthood, developed a reasonable grasp of moral obligations towards other human beings. Despite this evident truism, there was never a day during the 20th century, when some human person was not being grossly mistreated while serving as a research subject. From Walter Reed’s questionable use of soldiers to study Yellow Fever; to the 40+ years of the Tuskegee syphilis experiments, wherein African-American prisoners in Alabama were told that their syphilis was being treated, when in fact researchers were using them to study the course taken by syphilis, and no treatment was tendered to them at all; to the recent tragic death of 16-year-old Jesse Gelsinger in a gene therapy trial run in flagrant violation of the researcher’s own protocol; to thousands of experiments conducted in total disregard of the principle of informed consent.

A similarly surrealistic disregard of both common sense and common decency may be found in the scientific community’s response to revelations of data falsification and other misdeeds in research—the president of the American Association for the Advancement of Science assured the public that any scientist behaving in that manner must be viewed as suffering from “temporary insanity.”

If human beings and, indeed, moral and methodological principles presuppositional to the very nature of scientific activity could be so cavalierly abused, it is not difficult to guess the fate of moral questions arising in the course of research on non-human animals. After all, if blatantly obvious moral constraints on the treatment of people, clearly codified in consensus societal ethics could be ignored, how much more so the treatment of animals, where no moral principles appeared in social consensus ethics, other than the prohibition of deliberate, purposeless, unnecessary, sadistic cruelty. And tellingly, a literature search conducted for me by the Library of Congress on “analgesia for laboratory animals” in 1982 as part of my effort to convince Congress of the need for legislation protecting animals in research, yielded not a single reference.

In the same vein, between 1976 and 1985, I and two colleagues worked on drafting the laws currently governing the use of animals in research. We searched assiduously but vainly in the scientific literature for a reasoned discussion of the ethical issues occasioned by such use. What we found was epitomized by a video entitled “Will I Be All Right, Doctor?”, a phrase uttered by a frightened child before an operation. The physician assures the child that he will be, as long as doctors are left alone to do as they wish with animals. So mawkish and irrelevant to ethics was the film, that when it was premiered at a meeting of laboratory animal veterinarians, assumed to be sympathetic to the message, and comments were solicited from the audience, the only response was “I am ashamed to be associated with a film pitched as low as the worst anti-vivisectionist clap-trap.” Advertisements defending unconstrained animal research appeared in Time Magazine affirming that “95% of the animals used in research are rats and mice, and you kill them in your kitchens anyway.” Needless to say, that argument did not play well with the public.

In the end, the ineffectual and clumsy way in which scientists respond to ethical issues comes from the ideology inculcated into them in the course of their education, affirming that “ethics is just emotion.” The notion of rational ethical argument is seen as oxymoronic. Clearly, any activity that harms another being, who is in some way negatively impacted by the harm and is aware or sentient, i.e., to whom the harm, be it pain, fear, or some other form of negative mattering matters, raises an ethical question: what right do we have to inflict such an insult? The answer is far from clear regarding invasive research on animals, but the question is surely legitimate, particularly since our most carefully articulated and thought out ethical notions, i.e. those used in assessing our treatment of other human beings, strongly limits harming a minority for the benefit of the majority.

One can distinguish three layers of ethical questions regarding the use of animals in invasive research for the benefit of humans. First, what justifies such use of animals when we would not allow the use of humans for similar purposes, even “marginal” or “defective” humans? It is sometimes said that such use is justified by virtue of the fact that we are more powerful than animals. That, of course, is not an ethically relevant argument, because it presupposes that “might makes right,” a notion that ethics exists in large measure to oppose! It is also argued that we are “superior” to animals, or evolutionarily “higher”; therefore we can use them as we wish. This claim, too, has obvious responses. In what morally relevant way are we superior? There are animals that are stronger, swifter, more adaptable (e.g., cockroaches) than we are. Ironically, it is sometimes affirmed that we are superior because we can judge our actions in moral terms. If this is indeed the case, then we should be more, not less, morally attentive to other beings, and certainly not transgress against their interests in a cavalier way.

Let us suppose that we have answered the question of what entitles us to use animals in terms of cost-benefit emerging from animal research: that is, that the benefit to humans outweighs the cost to animal subjects. This is, in fact, a common argument. If that is the case, then the only morally permissible research would be found in those cases where the benefits to humans clearly exceed the costs in animal suffering. Certainly, some animal research meets that criterion, as when a small group of animals are used in a way that helps vast numbers of humans. But, a great deal of research, perhaps the vast majority, does not meet that standard. Toxicological testing of new cosmetics, and a whole host of similarly trivial products, which inflicts significant harm on test animals, does not seem to produce a positive cost-benefit ratio. Neither does a good deal of psychological research, such as studies of learned helplessness. Neither does weapons research. Neither do studies of animal aggression. One could proliferate a long list of research not meeting the cost-benefit test, which is nonetheless regularly performed. So, once again, we fail to adequately respond to a clear ethical challenge emerging from animal research.

Thus far, a conscientious person should be feeling uneasy about animal researcher failure to provide first of all a sound moral justification for hurting animals in research. Second, such a person who responds by appealing to cost-benefit as a justification should feel equally uneasy about the fact that much research does not pass the cost-benefit test. Third, there is one more layer of ethical concern about animal research that we do not adequately address. That is, that we do the utmost to minimize animal suffering arising in the course of animal research, and maximize the animals’ ability to live a life where the interests and needs flowing from its biological and psychological nature are respected.

Certainly, the control of pain was a fundamental moral obligation to research animals that the research community failed to meet until compelled to do so by federal law. From the zero papers I found in the literature search I conducted on laboratory animal analgesia in 1982, the literature has grown to over 11,000 articles I found a year ago when I redid the search, with of course a correlative increase in use of pain control. Equally gratifying is the fact that far fewer researchers would claim today that animals do not feel pain. But let us recall that, at least by the research community’s estimate, only 15% of research protocols involve pain. Even if one mistrusts that source, we can double that estimate and still believe that the majority of research does not involve pain. Furthermore, in the majority of cases, the pain can be attenuated or controlled pharmacologically, leaving a small percentage of protocols requiring that animals suffer pain. What is often ignored, however, is the fact that almost 100% of research protocols fail to keep and house the animals they use under conditions congenial to the animals’ biological and psychological natures. If that is the case, then virtually all animals suffer deprivation that is probably as onerous as uncontrolled physical pain. Such deprivation includes full-time light for nocturnal animals, no opportunity to burrow for burrowing animals, lack of companionship (e.g. for dogs housed in solitary cages), food alien to what they would consume in nature, failure to respect how they are built to acquire food, and myriad other assaults upon their natures. Much of the reason for a thrust for enrichment is to rectify that wrong. It is far easier to create enriched environments in which the animals spend the majority of their time than it is to respond to the moral imperative that only research with a positive cost-benefit ratio should be performed. Respecting research animals’ biological and psychological natures is more like the moral requirement of providing food than it is like providing treats.

The difficulty of achieving moral goals is often directly proportional to the profundity of the moral imperative underlying the goal in question. Enunciating the moral ideal of human equality was presuppositional to creating the concept of American democracy, and occurred early in our history. Yet, as the cases of African-Americans and women vividly illustrate, realizing that goal has taken hundreds of years, and is far from finished. The three moral questions we have formulated regarding animal research are clear, yet we have made virtually no progress in responding to them. The first two questions strike at the heart of animal research; consequently it is not difficult to see why they have been ignored. But the third question and what it entails are currently achievable. Even if we lack full knowledge of animals’ needs and natures constitutive of their telos, we certainly know enough to come much closer to satisfying those needs than we currently do, where design of housing and husbandry is based almost exclusively on the convenience of those who keep the animals.

“Enrichment” is therefore not the bighearted largess that much of the research community seems to believe it is. Rather, it represents an obvious step towards creating a decent life for research animals and minimizing avoidable suffering. The fact that providing proper environments that suit the animals’ needs and natures has not been done historically, may be inconvenient or more expensive for research facilities, may skew baseline data, or otherwise lead to new practical issues in research management, does not mean that creating such animal management is not a moral imperative. If moral behavior was the easiest and most convenient option, everyone would be moral. The sense of moral discomfort that should arise when one reflects upon the morality of animal research should at least translate into a moral imperative to create living conditions for the animals we use that do not assure a life of misery.

Volume 10, January 2012

Responses to the article noted below:
Environmental Enrichment of Laboratory Rodents: The Answer Depends on the Question
Toth, Linda A.; Kregel, Kevin; Leon, Lisa; Musch, Timothy I., Comparative Medicine, Volume 61, Number 4 August • Pages 314-321

Commentary
By Karen Froberg-Fejko, VMD

I was disappointed by “Environmental Enrichment of Laboratory Rodents: The Answer Depends on the Question.” I expected a much-needed article championing reasons why it is essential to provide environmental enrichment for rodents. To my dismay, the authors chose to weigh in on the side of those who see little value in providing EE for rodents. I offer my opinion as a lab animal professional who sees the glass as half full as opposed to half empty.

First, we must ask: what is enrichment? Definitions vary widely; however, we must accept that the intention of EE is to allow the expression of normal species behavior. Although there are wide variations in rodent species and strains in the response to EE, the dominant behaviors in rodents such as sheltering, nesting, foraging, and gnawing are hard-wired. Attempts to allow what is normal should be offered and noted, specifically in the materials and methods sections of publications to provide evidence supporting introduction of EE.

Mice respond differently to different environments, and the difficulty for us is to identify individual needs. The number of rodents used in research can present an overwhelming, but potentially rewarding challenge to animal caretakers. Instead of adopting a herd mentality, we need to carefully evaluate what is going on at cage level. We must recognize that “one size does not fits all,” and I would argue that the responsibility of monitoring the effects of EE must be placed upon the human caregivers. I have confidence in the honest evaluation by daily caregivers because they have hands-on experience and know what is effective. I realize human emotion has a finite role in a study design; however, because we are discussing the needs of rodents, there is no black and white, so we must rely upon observations of what appears appropriate until more research in this area has been conducted.

In my travels, investigators frequently state that their mice “LOVE their shelters.” I interpret this to mean that the sheltering option they have provided to their mice is having a positive effect. I believe that it is essential to validate the effects of different types of EE upon rodents, but due to fiscal challenges and other priorities, it is not happening. If we have to wait to validate EE, then the rodents will suffer. In the interim, I believe we should rely on our caregivers and investigators to accurately monitor and note changes on different types of EE offered to rodents.

Second, who should determine whether enrichment interventions benefit or harm the animal or the science? The authors of the article maintain that EE should not be imposed unilaterally or arbitrarily by any stakeholders. I submit we have a responsibility to impose best practice stipulations since rodents are not covered by the Animal Welfare Act. Rodents are the most utilized animal species in research, and there are no federal regulations to protect them. This makes the GUIDE an essential document, promulgating best practices that address the needs of rodents as opposed to settling for minimum standards of care.

The authors state that complex and unexpected effects of EE on research variables are possible. No one would argue that research variables must be minimized in order to collect valid data, and this concern is magnified in a tox or GLP environment due to the nature of this type of research. There is concern in a tox and GLP environment that providing environmental enrichment could potentially affect the outcome of a study. However, not offering the opportunity for animals to express species-specific behavior through enrichment can lead to the most important variable of all, stress. A barren environment is stressful. Stress affects every physiologic function of the body, and we must strive to minimize it.

There are many uncontrolled variables within a research environment including personnel rotations, differences in HVAC and caging systems, monthly fire alarms, vibrations, and changes in light levels. We must recognize them as an inherent part of lab animal research. I have had dams cannibalize their young because of building renovations. This behavior dissuaded with the addition of EE. My point is that EE can be extraordinarily beneficial to achieve a desired outcome. There are numerous published research articles and plentiful testimonials supporting the positive effects of EE. But it is important to recognize that institutions must invest in EE programs and staff training. A successful EE program requires careful planning, controlled implementation, close observation and frequent re-assessment. Performance goals should be planned and close evaluation must be documented. A well-managed EE program will undergo continual assessment. It requires time, money and commitment.

There is no doubt that the article has generated discussion surrounding the implementation of EE for rodents. It is our responsibility to stay close to ongoing research and support more studies in the future. Providing EE is good animal welfare, and we must strive for the delicate balance of good data collection conducted in the least stressful environment. I wanted to share a comment by an anonymous author which emphasizes the importance of EE: “For those who pledge to take responsibility for the welfare of animals and vow to use scientific knowledge and skills for the advancement of medical knowledge, the wise composer of this oath sees no conflict between relieving animal suffering and advancing science. Indeed there is none!”

The Standardized Environment Must Be Enriched
By Emily G. Patterson-Kane, Ph.D.

Toth et al (2011) musters a number of arguments against willy-nilly environmental enrichment. And my issues with them, spelled out below, are more a matter of attitude than fact. But when it comes to environmental enrichment, attitude may be more important than many people realize….

1) Environmental enrichment is not well-defined

No abstract concept is uniformly defined. Environmental enrichment aims to create environments for animals that don’t suck. Each person proposing a definition has their own idea about what sucks the most about the captive environments they are working with, and honestly, there is a lot to choose from. Thus, the diversity in definitions reflects just how badly environmental enrichment is needed. (I and quite a few others are not, as it happens, a fan of the term “environmental enrichment”. I am, however, a huge fan of designing captive environments that don’t suck.)

2) Environmental enrichment can take many forms

Because there are many deficits, there are many solutions; we don’t always know what they are, so we have to try different methods before we strike the right one. Even a solid floor can be enriching to an animal that previously lived on a wire floor. Hanging wire-floored cages were the very model of scientific standardization, as long as you didn’t care that your rats were miserable, had sore feet, underdeveloped brains and compromised immune systems. It may seem like a small gain now, but environmental enrichment as a concept and rallying cry played a large part in achieving it, for the benefit of animals and science. (Actually, there are still a substantial number of labs that cling to their hanging cages, out of a sheer unwillingness to adapt to the new standard of care and shift their baseline.)

3) Varied housing impairs standardization

“Enrichment” is essentially an argument for a transitional period during which we determine the optimal environment, moving us from housing that causes suffering and impairs scientific validity to housing that supports good welfare and good validity. I have to say that most areas of research suck at standardization, so it is entirely a valid area of concern. I once read pretty much every study that used an open field with rats, and the only ones that used a field of the same size, shape and color seemed to be the ones actually using the same piece of equipment. The number of studies that even report home cage parameters are negligible to this day. I also read almost every study that ran a rat through a Hebb-Williams maze, an admirably standardized piece of equipment. So standardized, in fact, that we haven’t the slightest idea what it actually measures—and even more dangerously, we think we do. (If it measures intelligence-or some- euphemism-for-intelligence, how does extreme hunger or foot shock make animals more “intelligent”?)

4) Evidence that enrichment is beneficial is mixed

Evidence of gravity is mixed if you stand near the edge of the Grand Canyon and there is a good updraft. Serious and learned reviews since at least the 1970s have shown that barren housing impairs nearly all of the structures and functions of an animal (yes, even a domesticated one) and thus its ability to serve as a model of normal function. The only real question is: how do we fix this?

5) Housing type effects research outcomes

Well, duh. Housing is important and housing is part of the research model. The role housing plays is 1) it makes the animal normal except in an area where deviations from normality are an explicit part of the model, 2) it occurs in a manner that is well-defined and reported to the point of allowing replication, 3) housing is standardized to the extent that the need for unnecessary replication is minimized. If housing is not a large part of the model, minor to moderate changes should not have major outcomes unless the model is poorly understood, the model has poor validity, or standard conditions introduce major abnormalities. (I am not suggesting making changes arbitrarily. But I am rejecting any notion that they should not be made at all just because they can affect your data.)

Conclusion
Standardization cannot be considered in isolation from validity. And, I would argue that in most cases, animal research has disappeared up its own uniformity and is now rather unwilling to come out. Normal animals make valid models, and barren environments make abnormal animals. Thus, the standardized housing for any species must be free from any factor that causes deviations from normal function. The standardized environment must be enriched.

And, if we do not yet know exactly how to properly house every kind of animal we use, perhaps it is damn well time we figured it out. The authors state that developing enrichment that prevents stereotype is “a complex task”. And giving enrichment to large colonies of animals can be challenging. Well… tough. Using these animals carries with it the absolute duty to use the most effective and humane methods. If we admit ourselves incapable of knowing what these methods are, then maybe that is the very first research question we should answer. Do not pass go, do not collect two hundred dollars. If we are merely unwilling to ensure the task is carried out… well, expect no sympathy.

That said, we need to admit that this is a difficult time for many researchers. We do not yet have an array of accepted, proven enrichment cages suitable for standardization. Baselines need to be shifted, which is not an easy task and can’t be done every time someone wants to toss a new toy in the cage. And yes, standardization has become harder to achieve. It is reasonable for researchers to require that enrichment of proven worth be implemented according to a scientifically responsible protocol that is respectful of the primary research goal—that being the reason everyone is there in the first place.

Why is this not always happening? Perhaps it is because a lot of researchers have dug their heels in and are essentially not allowing enrichment, even when it does meet these criteria. Thus technicians who are with the animals all day want to sneak in any small enrichment they can. Administrators make rules that you must enrich, because enrichment is good and they want to make sure that you  do it. You see, if you want a careful and measured—a scientific—approach, when you do finally make a change, it needs to be large, proactive and effective. Not just a toy, but a whole new cage type, or room, or building. “Adequate” is not enough. You have to be willing to embrace and create a new standard. Only then will you be accepted as the leader, the boss, the rule-maker when it comes to working out how to get there.

Volume 10, January 2012