The Humane Society of the United States Pain and Distress Initiative
Andrew N. Rowan
Senior Vice President for Research, Education, and International Issues
Humane Society of the United States
Our pain and distress initiative consists of four main segments (HSUS 2000). First, we have commissioned the preparation of an expert report. Some may be beginning to wonder if it will ever arrive, and we are as anxious to see it as the rest of you are. Dr. Joy Mench, the chair of the panel, assures me that it is in process and close to completion. The report is meant to be a detailed analysis, somewhat akin to the NRC (1992) publication. The charge to the expert panel was given by providing an outline of the material we wanted addressed. However, the actual content will be the exclusive work of the panel.
Second, we have sent out five letters over the past 2 years to IACUCs as part of our outreach to research institutions. We are focusing on the IACUC as the institutional entity because we believe it has the appropriate oversight responsibility for this topic. We plan to send out more letters in the future. Some IACUCs have threatened to return our letters attached to bricks and less desirable objects that the US Postal Service might prefer not to handle, but by and large the reaction that we have had to date has been positive.
I know when I was chair of an IACUC at Tufts, getting the type of material that we have been distributing would have been useful to me and the other IACUC members. I am hoping that slowly people will begin to look past the source and find it useful to their IACUCs as well.
Third, we have also focused on regulatory issues and on some of what we perceive to be the shortcomings of the current regulatory system. Finally, we want to encourage more research and funding on how to assess and limit animal pain and distress.
CONTEXT OF THE INITIATIVE
Several of you in the audience have suggested that the message implicit in our initiative is too critical, especially of the laboratory animal personnel who are on the front line of laboratory animal care and welfare. We cannot escape the fact that the initiative does criticize what is currently being done, and I am not going to try to sugarcoat the issue.
However, I want to state very strongly that such criticism is not directed at people's motives or character. We recognize that there is an enormous amount of concern for animals in laboratories. Pain and distress is something that concerns all of us whether it is experienced by humans or animals. A concern for other beings' suffering is a fundamental element of any humane, empathic, individual.
What we think is the main issue is that we disagree on definitions and on the interpretation of what few empirical data are currently available to guide us in deciding what causes animal distress. That is where I see the disagreement, and if there are friction and unhappiness with the HSUS pain and distress initiative, we would like to see the debate focus on how we differ as regards definitions and the like.
DEFINITIONS OF TERMS
People commonly use pain, distress, and suffering as synonyms. I want to argue very strongly that these words do not refer to the same states. There are overlaps but they are absolutely not synonyms. There are other terms as well that describe parts of responses involved in suffering.
Twenty years ago, nociception was described to me as a scientific euphemism for pain. However, I now know that this is not a scientific euphemism at all. It is a very precise term that is not equivalent to the sensation of pain at all. It simply refers to the “stimuli ” passing from the nociceptors up through the nerve fibers that conduct those stimuli. The sensation of pain may or may not be an outcome of such nociception. In fact, what may be viewed as behavior signaling pain may simply be a nociceptive reflex.
One example that illustrates this concept is the experience of a human being with a high spinal cord break. Such an individual could still have the nociceptive reflex loop for the lower limbs and will withdraw his or her foot from a hot iron or a painful stimulus, but there will be no pain perception because the central nervous system is not involved. The behavior is the result of a nociceptive reflex.
With regard to pain, the HSUS is not greatly concerned about pain research in the context of our pain and distress initiative. By and large, the International Association for the Study of Pain (IASP) has established good guidelines for ensuring that pain research protocols minimize pain. In quite a bit of acute pain research, the animal controls the level of intensity of the stimulus, and the animal says, “This is too much; I am going to stop now.” In a sense, the animal is a
volunteer as part of that program. There are concerns about chronic pain studies because the animal is no longer a “volunteer.” In addition, pain caused in other research programs can be addressed using some of the newer long-acting analgesics.
Anxiety and fear are not the same as pain and, unlike pain, there is no obvious organic location associated with anxiety. These states can be treated with anxiolytics and sedatives, but I doubt that many programs do so. One example of the impact of fear in the laboratory was a dog model of anxiety developed by W. G. Reese (1979) out in the Midwest 20 years ago using pointers. In the presence of humans, the pointers would urinate, defecate, and display tonic immobility. If you turned your back on them, you would get bitten. However, in the absence of humans or under anxiolytic therapy, you could not tell which pointers came from the anxious strain and which ones did not.
Discomfort is different from pain and anxiety/fear. It may have an organic location, but we really do not have any sort of good treatment other than palliative care. Disease and malaise may involve discomfort but also include states that are different from those we might characterize typically as discomfort. Distress occurs when any of the states listed above (pain, anxiety, fear, disease) become too great. Distress may be prevented by appropriate psychosocial initiatives. For example, an animal that trusts its handler is far less likely to be distressed than an animal that is fearful.
Finally, we have the overarching concept of suffering. Suffering is a state that requires a certain level of pain, anxiety, or distress before it is experienced. As I discuss later, suffering and distress are not the same concepts. The model that may be used to describe the relation between some of the baseline states like pain, fear, discomfort and disease, and distress and suffering is as follows. The primary baseline states occur at different intensity levels and for different time deviations. Whether or not these baseline states cause distress is a function of intensity and duration. A low stimulus for a long time may cause distress whereas an intense stimulus for a moment may not. Once one has a state of distress, the level of suffering experienced is a function of the cognitive capacity of the being experiencing the distress.
Let us examine the concept of suffering in a little more detail. Some interesting anecdotal data in the literature describe a few humans (not many, it is a very rare syndrome) who cannot feel pain. When you read some of these case studies, you find that some of these individuals report that they are very fearful of surgery or that they just cannot stand going to the hospital. They fear the hospital and report that they suffer in hospitals, and yet they experience no pain whatsoever. This is one of those examples which proves the point that pain and suffering are not the same.
Another side of the suffering question (related to cognition) is the question of whether insects can suffer. A fairly persuasive argument can be made that they do not. Insects have short life spans and there is not a compelling evolutionary reason why pain-induced learning would make much survival sense from an
evolutionary context for an insect. Also, they do not have the central nervous system machinery that appears to be necessary to develop a sense of suffering (lobotomized humans have very low affect and do not appear to suffer as we know it). By and large, one can explain most insect aversive behavior in terms of nociceptive reflex loops, rather than requiring a sense of distress or suffering.
In summary, distress may be caused by pain; however, in the laboratory, it is probably much more often caused by illness, fear, or anxiety. When does stress become distress, and what specific measures could help us determine when stress becomes distress? This is an area where we are still completely in the dark in our opinion. There are various types of empirical measurements that one can make at a distance, such as behavior or appearance. Other measurements can be made on handling, such as physical condition, weight, pulse rate, temperature, clinical signs, provoked behavior, and biochemical measures such as the blood level of the corticosteroids.
A whole raft of different physical and biochemical parameters can be measured that produce differences from one situation to the next. However, none of these measurements let us know where to draw the cut-off points between mild, moderate, and severe distress. Mouse blood corticosteroid levels have been measured at between 4 and 500/ng/mL. Where between 4 and 500 do we draw a line and characterize the animal as experiencing distress?
Currently, the common approach is to say, “I looked at the animals, and they looked fine.” However, when we look at the animals, we usually look at them during the day. Rodents are nocturnal animals. The level of activity during the day is about 5% of the activity at night. Therefore, if you are looking for differences, you are looking at a very low level of activity in the first place. They are all asleep in the back of the cage, perhaps. It is not easy under these circumstances to see whether they really are fine or whether they have some mild or moderate disturbance. At night, or under red light (as is used at the University of Birmingham in England [D. Morton, 1997, personal communication]), it will be much easier to tell whether the animal is not behaving normally, but I doubt that many programs inspect the animals at night or under red light.
Norm Peterson (personal communication, 1999), now at Johns Hopkins University, acquired four cages with activity wheels attached and looked at nocturnal activity of the animals based very simply on the number of times the activity wheel turned every night. In a normal cage of four mice, the activity wheel turned 1600 revolutions per night. When the mice were given intraperitoneal injections of incomplete Freund's adjuvant before the end of the day, the activity wheel turned 900 times per night. What does this difference mean in terms of distress? Do we have any reason to be concerned by such a difference? We at the HSUS would argue that we do, but perhaps others would argue differently.
Nonetheless, the argument is currently moot because we simply do not have much behavioral data on which to even start a debate.
Weight loss is a measure that we increasingly believe has a lot of promise as a screen for stress and distress. It is a relatively simple measure, and it is easy to obtain. Most stresses induce a transient decrease in feeding, an increase in the metabolic rate, and a negative energy balance due apparently to a biomechanism involving the actions of corticotrophin releasing factor and local cytokines (Dallman 2000). We have few data on weight loss in different experimental situations, and each strain of rat or mouse is likely to produce a different standard growth curve. In the HSUS white paper, we state that weight loss is a relatively insensitive measure. However, we retract that claim and now recognize that it can be very sensitive indeed. We will be pushing weight measurements as a simple way to promote both better science and less animal distress.
In terms of changes to the system, we absolutely must have adequate definitions. Dr. DeHaven provided a definition of distress this morning. Whether that definition is the right one is something we all need to discuss, as I am sure Dr. DeHaven agrees.
We must establish consistent reporting practices and report pain and distress when it occurs. The HSUS is concerned that if we are not reporting pain and distress when it occurs, then we are not paying attention to it. Of course, simply reporting such pain and distress does not guarantee that it will be attended to, but at least we would know that it was not being overlooked due to lack of attention.
We would like to see three grades of pain and distress—minor, moderate, and severe. The reason we would like to see reporting of all three grades is because we think we can focus our attention on severe distress and start studying how we can refine the technology to reduce that grade to moderate and then, eventually, to mild. Reporting severe pain and distress permits us to set priorities for research and action.
We need to support research on distress identification and alleviation. Such research might include more studies of weight loss and behavior in different research situations. The generation of more data then needs to be followed by broad-based discussions to determine cut-offs for the different categories. The fact that establishing such cut-offs will be difficult and involve some uncertainty is no reason why we should not start the process.
We need to support claims of action with real data. Such data is not something that the HSUS will be able to develop; such data will only be developed in laboratories and research facilities. Most of you already have access to such data anyway. We simply need to identify the personnel time to collect and analyze it.
We are pleased that attention to the pain and distress issue is growing. We
believe that greater attention to these issues will benefit science, will benefit scientists, and of course will benefit the animals.
Dallman M.F. 2000. Coping with challenge: Welfare in animals including humans. Dahlem conference paper. Berlin: Dahlem Foundation. In press.
HSUS [Humane Society of the United States]. 2000. HSUS Pain and Distress Initiative. URL: <http://www.hsus.org/programs/research/pain_distress.html>.
NRC [National Research Council]. 1992. Recognition and Alleviation of Pain and Distress in Laboratory Animals . Washington, D.C.: National Academy Press.
Reese W.G. 1979. A dog model for human psychopathology. Am J Psychiatry 136: 1168-1172.
QUESTIONS AND ANSWERS
DR. GEBHART (Gerald Gebhart, University of Iowa): Dr. Rowan, regarding your discussion about developing categories of mild, moderate, and severe pain and distress, the issue is going to be who will provide the category. Is it going to be the investigator, the veterinary technician, or the veterinarian? I guess these are issues that have yet to be resolved, but I would like your opinion regarding who should be doing that and how you would define the different categories between mild, moderate, and severe.
DR. ROWAN: I would not even begin to develop the definition. This area is where we believe we need data and a detailed discussion between researchers, laboratory animal care staff, and animal protectionists to develop appropriate guidelines. Although IASP developed guidelines, it did not produce guidelines for categories of mild, moderate, and severe pain. Perhaps it is too ambitious a challenge to expect to produce reasonably clear categories. Nonetheless, some other countries are attempting to establish several categories of pain and distress, and they appear to be able to come to some level of agreement on the systems they use.
I would argue that leaving the development of appropriate categories up to individual institutions would be one way to start, but that there should be some sort of overarching data gathering group that develops some standard measures, once again based on data.
Part of the impetus behind the HSUS initiative is to stimulate a data-based dialogue and argument. I do not want to stand here and say 10% weight loss should classify a procedure under severe pain or distress because 10% weight loss occurs very quickly in a rodent. Perhaps it should be characterized as moderate, or even as mild, distress. Who knows?
We need to engage the issues and learn what people think and why they think it. What is needed is a lot more traditional scientific dialogue, which has underpinned so much of the biomedical progress of the past century.