An Industrial Perspective
Lynn C. Anderson
Senior Director, Comparative Medicine
Merck Research Laboratories
Consideration of animal pain and distress is not a new issue. The utilitarian philosopher Jeremy Bentham addressed it almost 200 years ago when he wrote, “The question is not can they reason, nor can they talk, but can they suffer?”
For the purpose of today's workshop, however, I suggest that pain and distress be considered in the following framework:
The prevention and alleviation of pain and distress in laboratory animals is a legal and ethical imperative.
The quality of science depends on the absence of disease, injury, pain, or distress in our animal models.
The risk of causing pain or distress in a small number of laboratory animals may be justified if the research benefits society by identifying ways to prevent humans and other animals from suffering.
SOURCES OF PAIN
Pain can result from injury, surgery, or disease. It can be a potent source of stress and lead to distress or maladaptive behaviors; however, pain does not invariably lead to distress. Animals often adapt very effectively to pain or changes in their normal environment without becoming distressed.
As defined in Recognition and Alleviation of Pain and Distress in Laboratory Animals (NRC 1992), distress is an aversive state in which an animal is unable to adapt completely to stressors and shows maladaptive behaviors. This is the same definition of distress that Dr. Gebhart proposed this morning, and I recommend that USDA adopt it.
The assessment of pain and distress depends on animal observation and professional judgment. Qualified personnel must use subjective measures to determine the presence of pain or distress. Attempts to categorize the intensity of pain or distress as “mild,” “moderate,” or “severe” would require even more subjective measures.
The assessment of pain and distress does not depend on any single indicator. A very broad spectrum of behavioral changes can be related to distress. The Carstens and Moberg (2000) article recently published in ILAR Journal underscores the importance and the diversity of behavioral responses.
We know there are significant differences in pain response between species, between breeds, within strains, and among individual animals. Their response is further modified by the animals' gender, environment, age, nutritional status, and weight. In fact, an infinite combination of factors can influence an animal's pain response. Therefore, it would be impossible to reduce the assessment of pain and distress to a prescribed checklist that would allow us to identify with certainty what is painful or not painful, distressful or not distressful.
Assessment of pain and distress should be a collaborative effort and include the investigator, attending veterinarian, IACUC, and animal care staff. However, the concern for pain and distress should begin at the institutional level. The institutional leadership must establish a culture that demands respect and consideration for laboratory animals. In addition, the institution is responsible for ensuring that the people who work with the animals are qualified. The responsibility for assessing personnel qualifications and training may be delegated to the IACUC or the attending veterinarian, but it is ultimately the institution that must set high standards for personnel qualifications and provide the resources for appropriate training. The institution is also responsible for ensuring that there is adequate veterinary and animal care, that appropriate equipment and facilities are in place, and that a fully functional and qualified IACUC exists.
Investigators have the initial responsibility for assessing the potential for pain and distress in their laboratory animal models. They must consult with their veterinarians if there is the potential for pain and distress, but I advocate that investigators discuss their research protocol with a member of the veterinary staff to ensure that all aspects of the project support optimal animal welfare. Many of
our investigators who are trained at the molecular or cellular level have very little experience with whole animal research. Often, the veterinarian is able to enhance the overall design of a research protocol, which surpasses regulatory requirements but supports good scientific method.
In consultation with the veterinarian, each of our investigators estimates the number or the percentage of animals that may experience pain and distress in each study. This information is provided in the protocol submitted for IACUC review and approval. If any animals are likely to experience pain or distress, the investigator must also provide a scientific rationale for proceeding with the study. Because we understand that prospective estimations of pain and distress will not be 100% accurate for 100% of the studies, especially when working with novel compounds, our IACUC requires investigators to notify the committee immediately if, during the conduct of their research, there are any deviations from the estimated number of animals that experience pain or distress. In addition, at the end of each government fiscal year, every investigator is required to report the actual number of animals that experienced pain or distress.
As part of the protocol review process, the IACUC should evaluate the methods proposed to prevent or minimize pain and distress. The IACUC also has the responsibility to consider the scientific rationale for conducting a study that has the potential to cause pain or distress in laboratory animals and potential benefits to society. I agree with previous statements that anti-inflammatory agents and antibiotics, in addition to anesthetics, analgesics, tranquilizers and sedatives, are effective in relieving and preventing pain and distress. I support the USDA's proposed change that would include the use of all of these agents in Category D.
A qualified veterinarian or animal technician should be involved in assessing the animals during the course of the research and provide pain relief whenever possible. More than 30 years ago, the AVMA adopted an oath that requires veterinarians to protect animal health and relieve animal suffering, which is very pertinent to today's discussion. However, veterinarians are also expected to use their scientific knowledge and skills for the benefit of society and to promote public health and the advancement of medical knowledge. We cannot forget that there are still many disease entities that we have yet to conquer and that this will depend on animal research. Therefore, in some studies, the veterinarian may need to consider the need to treat animal pain and distress along with the ultimate goals of the research and the potential benefits to both humans and animals.
PAIN AND DISTRESS INITIATIVES
The HSUS pain and distress initiative suggests that we address the gaps in our knowledge of pain and distress and promote laboratory animal welfare by simply eliminating significant animal pain and distress. Although we understand many aspects of pain, more research is required to enable us to eliminate pain and distress further. We cannot address the gaps in knowledge by simply eliminating all research that results in significant animal pain and distress.
The USDA pain and distress initiative addresses the need to clarify the definitions of pain and distress and promotes expanding the categories for reporting pain and distress into mild, moderate, and severe. The biomedical community could benefit from additional guidance and clarification of the terms pain and distress by the USDA. However, there would be little, if any, added benefit to animal welfare by creating additional categories for reporting animal pain and distress.
I caution the USDA to avoid suggesting that certain procedures or animal models invariably induce pain or distress. For example, not all vascular procedures will cause pain or distress. Those that do will not be equal in intensity or duration of pain or distress. Likewise, not all animal models created by a specific technique, such as transgenic animal models, are subject to pain and distress. Any attempt to force certain procedures or models into specific pain or distress categories would preclude professional judgment. It would also fail to consider species-specific or individual animal variability. Finally, it would fail to consider the intensity or the duration of pain or distress.
Our investigators, veterinarians, and IACUC work diligently to evaluate prospectively whether a specific procedure in a specific research study may induce pain or distress. If we further expected them to estimate whether that pain would be mild, moderate, or severe, there would likely be differences of opinion. Although we may eventually come to a consensus regarding the categorization of pain or distress, similar professionals at another institution may have a different interpretation. This could lead to significant inconsistencies in the way animal pain and distress is reported and would not contribute to animal welfare.
More complex reporting requirements will undoubtedly increase regulatory burden. It will certainly require more time and effort to categorize animal
responses into different pain or distress categories that, in turn, will increase administrative costs. This regulatory burden does not benefit animal well-being nor does it benefit society; in fact, it may be detrimental to scientific and medical advances. It will decrease the scientific staff's available time for conducting their research. To put this issue in perspective, consider what would happen if 500 investigators each spent, conservatively, an extra 2 hours per year categorizing their animals' responses into mild, moderate, or severe pain or distress. That effort would require a total of 1000 hours. As a result, at least 6 months' worth of research work would not be done, which would likely delay or even impede important scientific discoveries. Animal welfare would not benefit; society may actually be denied improvements in health care or additional knowledge.
I would like to make several specific recommendations to the USDA. First, utilize the resources already available to assist investigators, attending veterinarians, and IACUC members to increase their understanding, recognition, and assessment of pain and distress. Specifically, I suggest informing them about credible references to use in this process. These include Recognition and Alleviation of Pain and Distress in Laboratory Animals (NRC 1992), the American College of Veterinary Anesthesiologists ' position paper on the treatment of pain in animals (ACVA 1998), and the recently published Carstens and Moberg (2000) paper. I also encourage the USDA to use and enforce the existing methods available to them to ensure humane animal care and use. This responsibility includes ensuring the qualifications of research, support personnel, and functional IACUCs.
If the USDA makes any changes in the way we report animal use, they should devise a system to ensure that each animal is accounted for only once. We maintain a number of animals that are used from year to year in very benign studies; yet, the same animals are reported each year in Category C. If the intent of the annual USDA report is to identify how many research animals are used in the United States, then reporting one animal four times over a 4-year period is misleading at best.
Finally, I think the USDA should use the annual reporting mechanism to help educate Congress and the public about the value of animal research. Currently, the USDA annual reports do not demonstrate how the number of animals used have increased our basic understanding of biology and medicine or prevented pain and distress to other animals and humans by providing safe and effective therapeutic agents, medical devices, or surgical interventions. I urge the USDA to correlate their reports of research animals used to the ultimate benefits to society.
ACVA [American College of Veterinary Anesthesiology]. 1998. Position paper on the treatment of pain in animals. JAVMA 213: 628-630.
Carstens E., and G.P. Moberg. 2000. Recognizing pain and distress in laboratory animals. ILAR J 41: 62-71.
NRC [National Research Council]. 1992. Recognition and Alleviation of Pain and Distress in Laboratory Animals . Washington, D.C.: National Academy Press.
QUESTIONS AND ANSWERS
DR. STEPHENS (Martin Stephens, Humane Society of the United States): Regarding your comments on the mild, moderate, and severe category system, you argued that there might be some inconsistencies in the reporting. However, the HSUS has documented fairly substantial inconsistencies in the current reporting scheme, and I believe that any inconsistencies in the new scheme should be weighed against that fact. I think over time, the inconsistencies in the new system would be ironed out after discussion at meetings like this, PRIM&R, and elsewhere. It could be that the same kinds of procedures should be classified differently depending on who is doing them and at what institution.
Finally, I would like to add that the current system seems to be as complex as any I have seen. In fact, throughout the 15 years I have been in this business, we have been trying to figure out what the current system means. I think it would be a little easier to tackle a system based on mild, moderate, and severe. Criticizing the new proposals and the direction we seem to be heading should be done in light of what we all agree is a very flawed current system.
DR. ANDERSON: I agree that there should be some clarification to ensure appropriate placement of animals in the existing C, D, or E categories. I think where we disagree is whether E needs to be further broken down into mild, moderate, and severe. I do not understand how those additional categories will benefit animals, how they will help us further prevent pain or distress.
Dr. Rowan said earlier that additional categories would help us focus on very severe cases of pain and distress. I maintain that we are already focusing on those cases, and we do not need to spend thousands of hours and hundreds of thousands of dollars in time and effort to get to those important issues. I do not see where mild, moderate, and severe pain categories are going to help us to do anything better for the animals.
MS. LISS (Cathy Liss, Animal Welfare Institute): I would like to add that I believe there is a serious problem of underreporting because of those who do not want to say there is any pain. If there is the opportunity to say it is mild, then it is more likely to be reported and, as a result, addressed by the IACUCs, the veterinarians, and others.
DR. HAYWOOD (J.R. Haywood, University of Texas Health Science Center): I am not sure animals that belong in Category E are underreported. I think the
question is what these additional categories would do to help the animals. As Dr. Anderson just pointed out, animals that need attention are receiving attention, regardless of their classification.
MRS. STEVENS (Christine Stevens, Animal Welfare Institute): We need reporting, but some people are not reporting correctly. Some people do not care anything about whether animals suffer. A prime example currently is Dr. Frederick Coulston who, in his work with chimpanzees (obviously the most advanced animals that have to be watched) is unable to keep capable veterinarians because one after another have left. He has had young veterinarians directly out of school who have never had any experience with chimpanzees. As a result, 13 chimpanzees have died in the last year or 2, from horrible complications of not being watched.
I also want to draw your attention to the fact that veterinarians have different opinions. To say simply that veterinarians' judgment must be followed is flawed because you cannot proceed on the supposition that every veterinarian is equally able and equally humane. There are enormous differences.
DR. ANDERSON: There is a mechanism, through the USDA, to address animal welfare concerns at the Coulston Foundation. However, I would ask whether reporting animal pain and distress according to mild, moderate, or severe would make any difference in the quality of those animals' lives. Would checking a box to differentiate between mild, moderate, or severe make any difference in animal welfare?
MRS. STEVENS: I think it could make a difference because the National Institutes of Health are keeping Coulston from going bankrupt. This is taxpayers' money, and we are supporting him by the tens of millions of dollars. I object to that money very personally.
DR. SCHULTHEISS (Peter Schultheiss, USDA): As a laboratory animal veterinarian who has spent the past year working with the USDA in animal care, I would like to address Dr. Gebhart's concerns regarding the value of reporting (not of breaking down “E” into mild, moderate, and severe). In reviewing the annual USDA reports, I learned something of which I was not aware via any other mechanism: that according to the statistics, about 80% of all animals in Category E were involved in testing, and more than half of those animals were used specifically for biologics testing for vaccine potency, purity, and safety.
While I was with the USDA, I received a call from Dr. Stokes, who asked me for a relative breakdown of testing procedures so that we might be able to concentrate our efforts on developing alternatives where they are most needed. Given the way we collect information now, I was not really able to pinpoint it, but I was able to bracket animal numbers as to what effect a certain alternative would have. That use of data would be an example, I think, of the value of collecting information for targeting where alternative work is needed.
I have a separate question for Dr. Anderson concerning her statement that the present system fails to consider intensity or duration. Other than pain or distress
that is not more than slight or momentary, you are correct, there is not really any duration or intensity consideration. I would ask you for a proposal to include that issue in a way that would not complicate the system so much that reporting really would become a burden. I think it would be a great idea to incorporate it according to agreed breakdowns such as 5 minutes, 10 minutes, 2 days, 2 weeks, or whatever is appropriate.
DR. ANDERSON: My point was that the USDA should not attempt to place certain procedures or animal models into specific pain and distress categories because animals do not respond to the same procedures with the same intensity or duration of pain and distress. For example, ovariohysterectomy may induce a very different pain response in a mouse than it would in a dog or other species. In this example, the pain or distress associated with the procedure would depend on many factors, including the interventions used to prevent pain. If, for example, the USDA were to suggest that all abdominal procedures cause severe pain and distress, the use of professional judgment would be precluded.
DR. GLUCK (John Gluck, Kennedy Institute): With respect to the point about how is it going to benefit animals to categorize pain, I think the answer is in your own talk. You started with Jeremy Bentham and the ethical system that he promulgated, which of course requires a judgment of right behavior based on consequences. I think if I, as a researcher, am faced with having to confront both my own and my expert colleague's judgment about the consequences of a particular intervention, whether it produces extreme, moderate, or mild pain, this ethical system will permit me to make explicit judgments about the ethical basis of my own experimental work. I believe it is valuable in that previous speakers have talked about our incredible ability as human beings to distance ourselves from some of the consequences of what we do, especially when we know our heart is really in it and we are really interested in those kinds of issues. I think the animals potentially can benefit, and I think the science can also benefit.