Skip Navigation

Proceedings for Pain Management and Humane Endpoints

Animal Welfare Perspectives on Pain and Distress Management in Research and Testing

Andrew N. Rowan, Martin L. Stephens, Francine Dolins, Adrienne Gleason & Lori Donley
Humane Society of the United States


The public's perception of the levels of suffering endured by laboratory animals used in biomedical research and testing has fueled the controversy over animal experimentation. Public concern has been translated into laws and regulations that seek to limit laboratory animal suffering, in the form of pain and distress. Accordingly, there are now oversight (NIH/OPRR) and regulatory infrastructures (USDA) in place to tackle pain and distress issues. However, these issues are not currently being addressed with the urgency they deserve. Insufficient attention has been given to the definitions, recognition, measurement, and alleviation of pain and distress in animals used in research. The irony is that the scientists who use animals in research are in the best position to determine the biology of pain and distress and to develop the techniques that can diminish suffering. Scientists can be encouraged to accelerate the process of the alleviation and elimination of pain and distress experienced by laboratory animals in a number of ways: the publication and dissemination of their results; the publication of laboratory protocols detailing in-house 'best practices' (refinements); and, substantial increases in financial support for research to identify and minimize pain and distress.

Scientists and laboratory personnel support the idea of minimizing pain and distress in laboratory animals and this support is institutionalized in the Institutional Animal Care & Use Committee (IACUC). There exists, however, a surprisingly limited amount of published knowledge about how to determine when animals experience pain and distress, and how much suffering is caused by typical laboratory procedures. For instance, the use of pain-killers in laboratory rodents has only become accepted practice in the past decade. As a point of comparison, anesthesia for human neonatal surgery was more the exception than the rule twenty years ago. Like animals, human neonates are non-verbal and unable to express their subjective experience of pain and distress.

Animal distress that is not the result of pain (e.g. anxiety, depression or fear) is still largely ignored or overlooked by research institutions. Coherent, tractable measures to gauge levels of distress in the common laboratory animal species do not presently exist. For the most part, lab workers rely on ad hoc observations or on relatively insensitive measures such as weight loss of 20% or more, to ascertain whether animals are experiencing pain and distress. Given the difficulty in objectively documenting distress, it is also difficult to convince decision-makers to take action to mitigate the situation. It is therefore essential that we identify which procedures cause either pain and/or distress, and develop measures for assessing the relative severity of such pain and distress.

To address the gaps in our knowledge and to promote laboratory animal welfare, The Humane Society of the United States (HSUS) has launched a campaign to eliminate animal pain and distress in laboratory animals by the year 2020. The HSUS seeks to encourage scientists and IACUCs to take on this goal. While this is an ambitious target, it is certainly within the ingenuity and skills of those who use and care for laboratory animals, and within the scope and responsibility of the scientific community.

The Concepts of Pain, Distress, and Suffering

The terms "pain", "distress", "anxiety", "fear" and "suffering" describe experiences, and responses to experiences that are, in most cases, unpleasant and hence undesirable. Such terms are commonly used in everyday language to describe both human and animal experiences. However, the difficulty lies in understanding exactly what is meant when we actually use such terms. Dictionary definitions are often circular and unhelpful. For example, in the 1967 unabridged Random House Dictionary, pain is defined as both a sensation of acute physical hurt or discomfort and as emotional suffering and distress. Suffering is then defined as undergoing pain or distress. The Random House and other dictionaries appear to view pain, distress, and suffering as synonyms. However, a closer analysis reveals that this assumption is not supported (see Table 1 for a set of definitions of relevant terms.)

Table 1. Definitions of Pain and Distress Terms

NOCICEPTION- The process whereby potentially noxious and/or tissue damaging stimuli cause special receptors (nociceptors) to fire and send a nerve impulse along the nociceptive pathways. Pain perception may occur, but only when such nerve impulses are processed in the central nervous system. Pain perception is not a necessary part of nociception.

For example, if one applies a heat stimulus to the foot of a high-level paraplegic whose lower spine is still functional, the paraplegic will withdraw the foot. No pain is felt but the nociceptive reflex loop is still functional and will protect the foot from being burned. It is not easy to discriminate between a nociceptive reflex and an aversive withdrawal that is the result of pain perception.

PAIN - An unpleasant sensory and emotional experience associated with actual or potential tissue damage or described in terms of such damage (IASP, 1979).

Pain terms are very variable and people may talk of acute or chronic pain, or sharp or dull pain, for example. Pain is NOT solely physical NOR psychological, it is both.

ANXIETY- An emotional state involving increased arousal and alertness prompted by an unknown danger that may be present in the immediate environment (Kitchen et al, 1987).

Unlike pain, anxiety is a diffuse sensation that has no specific location in the body. Unlike pain investigators, scientists who study anxiety have not developed a code of conduct to limit the extent of anxiety an animal may experience.

FEAR - An emotional state involving increased arousal and alertness prompted by an experienced or known danger present in the immediate environment (Kitchen et al, 1987).

DISTRESS - A state in which the organism is unable to escape from acute stressors or adapt to an altered external or internal environment.

In acute distress, the organism will try to escape but in chronic distress, the organism will commonly engage in maladaptive (e.g. learned helplessness) behaviors (cf ILAR, 1992).

SUFFERING - A highly unpleasant emotional response usually associated with pain and/or distress. (Kitchen et al, 1987)

The adjective "emotional" stresses the affective nature of suffering. Suffering involves a threat to the "personhood" or self-concept of an individual rather than simply to the organic body and is a metaphysical concept. It cannot be reduced to "operational" terms and is, thus, not easily incorporated into "objective" sciences (see below).

Most of the discussion about animal pain and suffering concentrates on pain, not suffering. A report from the Netherlands, entitled the "Definitions of Pain, Stress and Suffering and the Use of These Concepts in Legislation on Animal Suffering," has almost no discussion of suffering itself although the term comes up frequently in the text (Voorzanger and de Cock Buning, 1988). Similarly, the report on animal pain and distress by the Institute for Laboratory Animal Resources (ILAR) (1992) defines and discusses pain, distress, anxiety, fear and discomfort but deliberately excludes any discussion of animal suffering. For the ILAR working party, suffering could not be defined operationally and therefore could not be reliably assessed. Like the term "obscenity", people are confident they can recognize suffering when they see it, but they cannot define what it is. Pain is also a very complex and private phenomenon, but is nevertheless considered easier to measure and to ground in the empirical world of biomedical research.

The Relationship between 'Pain' and 'Distress': Pain Without Distress; Distress Without Pain

In biomedical research, animals may experience pain, discomfort, anxiety, and fear in addition to functional deficits caused by experimental procedures. In most experimental protocols, an animal's pain may be treated with anesthesia and analgesics. These measures may relieve or even eliminate the experience of pain. To date, however, there are no similarly well-known methods to alleviate the distress, anxiety, and fear an animal is subject to before, during, or after experimental procedures. In some experimental protocols, anesthesia or analgesics are thought likely to interfere with the results and are therefore not used, leaving the animal with persistent and unrelieved pain. There is a need to understand animal distress and fear, and their relationship to pain, experienced by animals used in research.

Pain and its comparative phylogeny

While most humans can report whether or not they feel pain, animals cannot and this has led to problems in acknowledging animal pain (see Beynen et al, 1987 and Phillips, 1994 for examples). Non-verbal human infants were, until recently, also denied the capacity to fully experience pain confirming the importance of verbal report in legitimating pain perception (Anand et al, 1987). According to Daniel Tibboel (Personal communication - November, 1998, Zeist), in 1987 it was found that 85% of neonatal anesthesiologists agreed that human infants could experience pain but only 5% actually delivered pain relief. By 1996, 85% were giving pain relief. Thus, drawing attention to the issue of infant pain had a dramatic effect on the delivery of pain relief. It is also possible to study pain perception in animals using the same sort of techniques and reasoning by analogy as in human infants.

Typically, we resort to studies of non-verbal behavior, such as moaning and crying, writhing, wriggling and so on to infer the presence of pain perceptions in animals. Pain also has typical physiological and neurophysiological correlates which, unlike the phenomenological (or felt) occurrence that is pain, are subject to direct empirical investigation. For example, nociceptors (the nerve endings that, when stimulated, are associated with pain perceptions) have been found in all mammals and in other vertebrates. In addition, direct, percutaneous recordings in human subjects have demonstrated that feelings of pain are correlated with activity in the small myelinated (A-delta) and unmyelinated (C) nerves. Research on anesthetized mammals indicates that these same nerve fibers are activated exclusively (or most potently) by stimuli of noxious intensity. Such nerve fibers appear to be present in all vertebrates. Similarities between humans and animals have also been demonstrated in the central nervous system pathways involved in pain perception.

Thus, reasoning from analogy, from neuroanatomy, from neurophysiology, from neurochemistry, as well as from behavioral observations, most people conclude that animals, or at the very least the warm-blooded vertebrates, probably experience pain that is qualitatively and quantitatively similar to that experienced by humans. The USDA guideline on pain in animals states that if one has reason to believe that a stimulus would be painful to humans, it should also be regarded as painful to animals.

While most people assume that vertebrates perceive pain, the situation is not as clear for most invertebrates. However, the common octopus, with its large central nervous system and complex behaviors (e.g. Wells, 1978; Young, 1965, 1971) has been given the benefit of the doubt in Great Britain and is now protected under the Animals (Scientific Procedures) Act of 1986.

For insects, some argue that they do not perceive pain but that it is difficult to be certain. For example, Eisemann et al (1984) and Fiorito (1986) argue that insects do not perceive pain although they might still avoid some aversive stimuli. Others (e.g. Wigglesworth, 1980) are also uncertain about insect pain but believe that insects should be accorded the benefit of the doubt.

The conclusion that insects do not perceive pain is based on several lines of reasoning. First, although insects have complex nervous systems, they lack the well-developed central processing mechanisms found in mammals and other vertebrates (and the octopus) that appear to be necessary to feel (perceive) pain. Second, insects apparently have not been shown to have a nerve fiber system equivalent to the nociceptive fibers found in mammals. However, this does not mean that they do not have some (other) nerve fibers that carry nociceptive signals. Third, the behavior of insects when faced with noxious stimuli can usually be explained as a startle or nociceptive protective reflex. In some cases (e.g. locusts being eaten by fellow locusts), insects display no signs that the tissue damage that is occurring is aversive.

The conclusion that insects do not perceive pain appears to contradict the claim that pain confers important survival advantages. However, simple nociceptor reflex loops (producing the startle reflex) that involve no pain perception could confer sufficient evolutionary advantage in short-lived animals (like insects) that rely on a survival strategy involving the production of very large numbers of individuals.

If insects and most other invertebrates do not perceive pain, it would be relevant for ethical systems that relied on sentience as an important criterion of moral considerability. However, it would not necessarily indicate that insects should be accorded no moral consideration. Moral arguments that relied on reverence for life considerations (e.g. the Jain or Schweitzerian systems) or ecosystem values would still regard insects as deserving some moral consideration.

Distress and suffering

Distress is functionally and physiologically distinct from pain, although the two may often interact on a cognitive-emotional level. Distress involves the activation of neural pathways in the limbic system of the brain that process emotional response to pain, fear and anxiety. Distress itself is used as a qualifying, catch-all term for multiple negative states which precludes the ability to measure this subjective state of being. In humans, verbal answers to specific questions can, in most circumstances, be provided but the human subject may still be concealing his/her internal states (hence the uncertainty of lie detector tests). Animals are non-verbal and cannot similarly express and describe their feelings. Consequently, distress may only be measured by external standards. Given this limitation, and the relatively small degree of attention given to understanding the welfare implications of stress in animals, there are currently few methods that are applied in identifying and reducing the distress caused to animals in research.

Pain, fear, anxiety, discomfort and distress are all negative subjective states of being, and are typically described and grouped together under one larger heading of "suffering". "Suffering" is a widely used and abused colloquial term that has been subjected to very little careful analysis, even in the case of human suffering. Cassell (1982), one of the few to address the biological and psychological roots of human suffering, argues that suffering occurs when the integrity of a person (not the body) is perceived to be compromised or threatened in some way. (Personhood is defined in terms of an individual's mental life and is distinguished from the organic body). Damage to organic tissues can and often does lead to suffering but, for Cassell, it is the psychological reaction to such damage that is the key to understanding the idea of suffering.

The notion that suffering arises from a perceived threat to the integrity of a "person" has significant ramifications for any discussion of animal suffering. Animals would, according to the above definition, suffer only if they possess to some degree the qualities of personhood. In a later analysis, in which he specifically addresses the issue of animal suffering, Cassell (1989)and Byrne (1999) argue that only beings with a sense of the future (anticipation) and a sense of self are capable of experiencing suffering. Some animals do appear to have a sense of self (e.g. chimpanzees and other great apes) and a sense of the future or, at least, seem to be able to anticipate and reflect on future events. How far such abilities extend through the animal kingdom would necessitate a much more detailed analysis than is possible here. One could also argue that only animals that are capable of affective (e.g. emotional) responses might be included among the category of beings capable of suffering (Damasio, 1994).

It is quite clear that few, if any, people use suffering in the narrower sense articulated above - referring only to perceived threats to the "person" rather than simple vigilance to protect against threats to the non-reflective organism. Even scientists who object to using the term "suffering" when referring to animal distress will, nevertheless, still argue vehemently that animals (including invertebrates) are capable of suffering. However, the colloquial term "suffering" has such broad meaning (see dictionary definitions) that it cannot be used profitably (even after careful definition) when trying to assess the severity of aversive stimuli to animals, or even to discuss the level of distress experienced by animals. The point is not to dwell any further on whether or not animals experience suffering, but instead to examine one fairly specific element of affective machinery - namely, an animal's capacity to experience anxiety and fear.

Anxiety and distress

Considerable attention has been paid to the use of appropriate levels of painful stimuli by researchers who study pain in animals. For example, the International Association for the Study of Pain (IASP) has set up guidelines in which researchers are urged to design only projects in which animals are given the opportunity to terminate any painful stimulus and thus control the level of pain they experience (IASP, 1979). Some of the simpler pain research protocols (e.g. the tail flick and hot plate tests) involve systems where the animal makes the choice to end the pain by moving itself or its tail away from the stimulus. The tail flick and hot plate devices are also fitted with automatic cutoff switches so that, in the event the analgesia under study is very effective, the animal will still not suffer any tissue damage.

Similarly, one can develop systems that allow animals to "volunteer" for pain research by offering them a highly desired food or drink. Such animals are willing to accept some painful stimuli in order to gain the reward. However, at the pain tolerance threshold, they voluntarily choose not to participate any further. Primates appear to have very similar tolerance thresholds to humans. In all of the above cases, the research protocol allows the animal to control when the painful stimulus is terminated. There are some studies (e.g. of chronic pain) where such refinements are not possible but, even here, pain researchers have tried to ensure that the animals do not endure a significant level of pain (Casey and Dubner, 1989).

By contrast, researchers who study anxiety in animals, which is arguably just as, if not more distressing to animals, have not developed similar guidelines and approaches. Some may not have paid attention to animal anxiety because they do not believe that animals can be anxious. For example, Cassano (1983), a psychiatrist, has stated that: "fear is a primitive state of mind found throughout the animal kingdom, whereas anxiety is part of conscious experience and takes shape as a typically human function or attitude. Thus, the age of anxiety could be said to begin with the emergence of Homo sapiens."

It is not exactly clear what the difference might be between fear and anxiety. One might fear some definable danger whereas anxiety may refer to that state of uneasiness where the threat is undefined and elusive. Cassano does not provide any clear distinguishing characteristics between fear and anxiety. However, there is at least one relatively clearly defined neural substrate that appears to be involved in mediating anxious states and this substrate was, interestingly, found to be present in all vertebrates but in none of the invertebrates examined (Nielsen et al, 1978). This substrate has come to be known as the benzodiazepine receptor because it binds the anxiolytic benzodiazepine drugs such as valium with high affinity. It also binds alcohol and the barbiturate drugs which also diminish feelings of anxiety.

Building on investigations of drug binding to the benzodiazepine receptor and its subsequent behavioral effects, Gray (1982) has produced a comprehensive theory of anxiety in which he argues that "...'human anxiety', or something very like it, exists also in animals ...." Gray recognizes that many people may find this conclusion hard to accept because of the common belief that anxiety is an almost uniquely human state, dependent on such complex cognitive capacities as the ability to anticipate future events based on past experiences, to form a self image, or to imagine one's own mortality. Nevertheless, he argues that the observed effects of such anti-anxiety drugs as alcohol, the barbiturates, and the benzodiazepines (e.g. valium) in animals are so similar to the observed effects of these same drugs in humans that it seems more parsimonious to argue that these agents act upon a state in animals that is similar to the human state of anxiety.

Research has also identified anxiety-causing compounds that bind to the benzodiazepine receptor in the central nervous system. The best known of these are the beta-carbolines which, when administered to humans cause intense inner strain and excitation, increased blood pressure and pulse, restlessness, increased stress hormone levels in the blood and stereotyped rocking motions. One volunteer experienced such severe anxiety that he had to be physically restrained and injected with a benzodiazepine which provided relief within five minutes (Dorow et al, 1983). The administration of beta-carbolines to primates caused piloerection, struggling in the restraint chair, increased blood pressure and pulse, increased stress hormone levels in the blood and increased vocalization and urination (Ninan et al, 1983).

The similar reactions of human volunteers and primates to the beta-carbolines does not prove that both humans and primates experience the same sort of anxiety but it is hard to argue that animal "anxiety" is not a significant cause of animal distress and suffering. Gray (1982) has suggested that "anxiety" may have evolved from a biological behavioral system - the 'behavioral inhibition system' (BIS). BIS may confer an evolutionary advantage by stimulating a state of alertness to novel stimuli in an animal's environment, making the animal less likely to rush into danger. Excessive stimulation of the BIS can clearly cause animal distress and suffering, which can be seen in a strain of "nervous" pointer dogs (Reese, 1979). The distress in the nervous pointers caused by the presence of humans can be easily eliminated by appropriate drug therapy, suggesting that the problem might have been due to mutations in the pathways controlling the "anxiety/fear" response.

While the distribution of the benzodiazepine receptor in vertebrates appears to provide a relatively "clean" distinction between "sentient" vertebrates and "non-sentient" invertebrates, research over the past decade has produced a host of confounding factors. First, there are other benzodiazepine binding sites which have now been shown to be present in invertebrates (Lummis, 1990). These "receptors" are found in non-nervous tissue and they are different from those found in the central nervous system of vertebrates. Nevertheless, the presence of these different benzodiazepine binding sites confuses the speculative idea that the presence of benzodiazepine receptors might be an indicator of sentience.

Second, a variety of other receptors that mediate anxiety and other anxiolytic drugs have now been identified. For example, cholecystokinin peptides and their receptors appear to be involved in mediating anxiety and panic (Derrien et al, 1994). Handley and McBlane (1993) describe a number of drugs including the increasingly popular anxiolytic, buspirone, that act through 5HT(serotonin)-receptors to mediate "anxiety" in both humans and animals. Thus, anxiety cannot be attributed to a single neurochemical system in the central nervous system. Nevertheless, it is abundantly clear from the pharmacology of anxiety in both animals and humans that anxiety can be a significant cause of distress and suffering in animals.

The relationship between pain, fear, anxiety, distress and suffering

In order to understand the underlying reasons for animal suffering and to alleviate its occurrence in laboratory animals, we must first examine its components. In the model presented in Figure 1, pain, fear, anxiety and discomfort are all aspects of the external, behavioral manifestation of underlying processes. For instance, a painful stimulus applied during an experimental procedure, given its intensity, duration and frequency of application, may lead to anxiety and fear. The animal comes to expect (predict) the arrival of the painful stimuli and therefore develops anxious and fearful reactions to any prior stimuli that are linked in time and space to the onset of pain. The sight of the hypodermic needle approaching, causing an animal to cringe, is one example. This cascade of cognitive-emotional responses can be termed 'distress'.

Figure 1. Model of Pain, Distress, Anxiety, and Suffering

The cognitive-emotional filter through which an animal perceives its subjective experiences of the external world will in turn influence its internal states of being. If the animal perceives that the onset of pain is to be expected, perhaps on a daily or hourly schedule, it may suffer emotionally from the anticipation or expectation of the pain. In this case, the animal's emotional state and behavioral response may extend beyond its initial responses to the degree of pain inflicted by the original stimuli. Thus, the negative emotional states experienced by the animal may not only contribute but increase its sensitivity to the painful stimuli it anticipates.

The well-studied state of "learned helplessness", which occurs in both human and non-human animals, illustrates the point that cognitive-emotional suffering may be even more intolerable to an animal than the physical infliction of pain. Animals in states of severe suffering may display learned helplessness in which they typically show no response or attempts to withdraw or protect themselves from mildly painful or harmful stimuli. Humans who display learned helplessness are typically individuals who have been subjected to severe torture, physical, and mental/emotional abuse. This gives an idea of the experience that the animal has gone through to reach the state of learned helplessness. Animal models for learned helplessness do not respond to therapeutic attempts to rescue their severely abnormal behavior and non-survival tendencies (Seligman, 1975).

To lessen the potential slide downward of captive animals' behavioral and mental/emotional states into boredom, frustration, depression and finally severe apathy (learned helplessness), it is necessary to provide both social and physical environmental enrichment (see Wemelsfelder,1999, for a discussion of this topic). Some aspects of enrichment involve rewarding and reinforcing social, physical, and other environmental stimuli. Promoting well-being, however is not quite as simple as providing a "likable" experience. A dog may find chocolate very rewarding but it is not good for the animal. If an animal is provided with food and water in a safe environment, why do we not consider this necessarily sufficient to maintain a state of well-being (see Shepherdson, 1999, for a discussion of this issue)? Experiments indicate that captive animals will preferentially work for food rather than eat what is freely available, indicating that foraging activity is itself rewarding (contra freeloading: see Young, 1999). Play behavior is certainly associated in humans with well-being and pleasure, but how would we increase the incidence of such behavior in captive animals? In fact, while we may think we know play when we see it, behavioral scientists argue endlessly over how to define animal play and what such behavior might mean - see Mitchell, 1990.

Concern about P&D in Research

In the USA, the system of Institutional Animal Care and Use Committees (IACUCs), established in the mid-1980s is specifically charged with reducing the likely pain and distress that animals may experience when used in research. Thus, IACUCs are required (by the USDA regulators who oversee the conduct of animal research) to ensure that investigators have searched for alternatives if the research is likely to cause animal pain and distress even if anesthetics and analgesics are used to prevent any pain and distress. By contrast, investigators do not have to demonstrate that they have considered or looked for alternatives if the animal research project is placed in the non-painful category. The implicit message is that animal pain and distress is of greater public concern than animal death [euthanasia]. Despite this regulatory emphasis on alleviating pain and distress, the USDA provides little explicit guidance on the topic or on the potential impact of specific experimental procedures on animal well-being (e.g. dosing with chemical agents or infecting with pathogenic organisms).

It should be noted that the systematic reduction of animal pain and distress in research is not a trivial task. First, as noted earlier, there is much conceptual confusion in the use of such terms as pain, distress and suffering. Second, animal use in the laboratory and the classroom is very varied. Nonetheless, while the techniques used in biomedical research are certainly numerous, it is certainly not beyond our scope to determine underlying principles of pain and distress in animals which can then be applied to the varied models and methods. Third, animal pain, distress and suffering are not easy to recognize or measure unambiguously and there is considerable opportunity for legitimate disagreement among scientists.

Aversive or distressing stimuli can take a variety of forms. Some cause physiological stress (e.g. injury, surgery, disease, starvation and dehydration), some cause psychological stress (e.g. fear, anxiety, boredom and lack of social interaction), some cause environmental stress (e.g. restraint, excessive noise, the presence of people or other species and chemicals) and some cause a mixture of stressors (ILAR, 1992). There are difficulties in assessing the severity of such varied states. This, however, is a task that must be addressed.

Factors that Create Pain and Distress in Research

Pain and distress caused by specific research models and techniques raise serious concerns for those in the animal welfare community as well as in the scientific community. For example, it needs to be determined how an animal's pain and/or distress might affect experimental results. Good estimates of how much animal pain and/or animal distress is caused by particular techniques or methods (with empirical evidence to support the estimates) are not yet available. For this very reason, gathering data to discriminate amongst research models and specific techniques is essential. Additionally, pain and distress may be specific to a particular research model, species, or gender and may affect the extent of suffering caused in that particular animal model (e.g., tumor site and burden).

A preliminary list of research models/areas has been compiled (see Table 2) dividing the research models and areas into two categories depending on whether the resulting distress is pain-induced or non-pain induced. There are overlaps, yet the distinction serves to draw attention to the relatively neglected issue of anxiety and distress in research animals.

Table 2. Models and Areas of Research and Specific Techniques that Cause Pain-Induced and Non-Pain Induced Distress


Non-Pain-Induced Distress

  • aggression models
  • anxiety models (e.g., Vogel conflict-drinking model)
  • cancer (tumor burden, cachexia, therapy, carcinogenicity testing)
  • depression models (e.g., learned helplessness, forced swimming, infant separation)
  • diabetes models
  • drug addiction and withdrawal models
  • environmental stress models (e.g., hot, cold)
  • fear models
  • immunological research (e.g., vaccine potency testing)
  • infectious disease
  • motion sickness models
  • nutrition research
  • panic models
  • pharmacology (some) (e.g., Tumor Necrosis Factor, capsaicin research)
  • psychopathology (other than anxiety, fear, depression, etc., mentioned above)
  • radiation research
  • stress models (psychological)
  • toxicology (induced effects)
  • transgenic research

Pain-Induced Distress

  • arthritis models
  • burn research
  • cancer research (tumor pain)
  • chronic pain studies (acute pain should not be a problem if IASP* guidelines followed)
  • inflammation studies
  • experimental surgery
  • muricide as a model of aggression, neophobia, etc.
  • orthopedic studies
  • trauma research

Specific Techniques

  • anesthesia after-effects
  • antibody production (polyclonal and monoclonal)
  • aversive stimuli (e.g. electric shock)
  • bleeding techniques (including retro-orbital bleeding)
  • Complete Freund's Adjuvant
  • control animals denied experimental treatments
  • deprivation limits (e.g., water, food, sleep or social)
  • partners/experiences)
  • dosing techniques (e.g., gavage)
  • granuloma techniques
  • gut loop studies
  • knock-out technology
  • surgery sequelae

* IASP: Report of International Association for the Study of Pain; Subcommittee on taxonomy.

The HSUS Pain and Distress Campaign

For the most part, our ability to detect pain, and more importantly, distress, in laboratory animals is very limited. We lack good measures and methods for quantifying distress in the common laboratory animal species.

To address this lack of knowledge and our inability to generate objective measures of negative subjective states, and to promote laboratory animal welfare, the HSUS has launched a campaign to eliminate pain and distress in laboratory animals by the year 2020. It is apparent that those who use and care for laboratory animals are already concerned about animal pain and distress. In conjunction with IACUCs, they have played a significant role in addressing problems of animal pain and distress in the past ten to fifteen years. Nevertheless, The HSUS believes that a more systematic approach will hasten achievement of the campaign goal.

The HSUS campaign consists of the following four initiatives:

  1. Development of a detailed and referenced technical report on animal pain and distress.

    The HSUS has convened an international group of experts including laboratory animal veterinarians, animal behaviorists, physiologists, neurologists, veterinary anesthesiologists, philosophers and others to develop and author a comprehensive report on the subject. These experts include, amongst others:

    • Dr. Joy Mench, Chair of the working group, Animal Sciences, University of California, Davis
    • Dr. Kathryn Bayne, Veterinarian, AAALAC International
    • Dr. David DeGrazia, Philosophy, George Washington University
    • Dr. Gary Moberg, Animal Sciences, University of California, Davis
    • Dr. Paul Flecknell, Anesthesia, University of Newcastle-upon-Tyne
    • Dr. David Fraser, Animal Welfare, University of British Columbia
    • Dr. Gerald Gebhart, Pharmacology, Iowa State University
    • Dr. Lanny Kraus, Laboratory Animal Medicine, Rochester University
    • Dr. David Mellor, Nutrition and Physiology, Massey University
    • Dr. David Morton, Biomedical Sciences and Ethics, Birmingham University
    • Dr. Bernard Rollin, Philosophy, Colorado State University
    • Ms. Ann Wolfe, Philosophy, University of Wisconsin
    • HSUS Contact: Dr. Andrew Rowan, Humane Society of the U.S.

    The topics that this P&D working group will cover in the technical report are:

    Executive summaryDefinitions of animal pain, distress, discomfort, anxiety, fear and suffering The biology of pain, distress and suffering Recognition of animal pain and distress: current and potential approaches Alleviation of animal pain and distress Housing issues Pain and distress caused by specific techniques and research endpoints Conclusion and recommendations Appendices

    The report will be finalized and approved by the expert group and will then be published by The HSUS as a state-of-the-art review.

  2. Outreach to Institutional Animal Care and Use Committees (IACUCs).
    In the past, campaigns against animal research have usually selected individuals or studies as targets of protests. The HSUS proposes instead to invite the collaboration of those who will ultimately have to develop the techniques and implement the approaches that will make our goal possible. The HSUS has invited IACUCs to join in the campaign. The HSUS will assist in promoting an exchange of information and policies so that new ideas and initiatives can be disseminated in a timely fashion and so that "best practices" spread rapidly. The HSUS will also focus on specific practices and research techniques (such as infectious disease or toxicology research) where relatively little attention has been given to animal suffering and seek out approaches that would eliminate significant animal distress.
  3. Regulatory Aspects
    The United States Department of Agriculture (USDA) enforces the Animal Welfare Act which regulates animal pain and distress while the Office for Protection from Research Risks (OPRR) at the NIH enforces Public Health System (PHS) guidelines on animal research which include strictures against causing unnecessary animal pain and distress. The USDA annual reports provide the data for states' use of animals categorized under different column headings. The use of animals in painful or distressful experiments for which pain or distress relieving drugs were withheld is listed in the USDA report under the heading of 'Column E'. Facilities reporting animal use in Column E are required to describe these experiments and explain why drugs for pain and distress relief were withheld. See Stephens et al (1998) for an analysis of these descriptions.

    The HSUS supports a proposal to alter pain and distress reporting under the Animal Welfare Act that would discriminate between the levels: no/little pain/distress, moderate pain/distress and severe pain/distress. In the meantime, The HSUS will seek to develop some consistency in how pain and distress is currently reported.

    Table 3 identifies the percentage of total animal use reported in Column E in 1996 for the states that use, on average, more than 20,000 animals per year. For comparison, Table 4 provides a list of the states that reported zero or less than 1% in Column E from 1995-1997. It is possible that the numbers accurately reflect the way that animals are used. However, there are obvious discrepancies between the types of research and publications emerging from institutions in those states, and their lack of reporting of animal use in Column E. Presently, the variations in Column E reporting between states and within a particular state over a series of years are not being effectively dealt with by USDA.

Table 3. USDA data from 1996 on Column E (Unalleviated Pain or Distress) Use for States Using Greater than 20,000 Animals

State% of Animals in Column EState% of Animals in Column E
Delaware9.3New Jersey4.5
Georgia13.9New York7.9
Illinois3.2North Carolina8.0
Minnesota28.9Federal Agencies5.8

Table 4. USDA Data on States Reporting Zero or Less than 1% Column E 1995-1997 (Average Number of Animals in Thousands)

Alaska (0.3)
Arizona (5.0)
Hawaii (0.5)
Kentucky (5.3)
Louisiana (16.8)
Maine (0.8)
Mississippi (2.0)
Nevada (3.0)
Oklahoma (4.3)
Oregon (4.7)
Rhode Island (2.1)
South Carolina (6.1)
Tennessee (10.9)
Utah (4.6)
Vermont (1.1)
Virginia (19.2)
West Virginia (1.7)
Wyoming (0.3)

While it is possible the differences from state to state reflect real differences in animal use, we suggest that the range of percentages is most likely due to differences in reporting practices rather than differences in the types of research and experimental techniques that dominate in specific states (although Kansas and Iowa report a large number of animals being used in vaccine challenge tests, all of which are placed in Column E). From 1983 to 1991, Virginia reported an average of 10-30% of the animals used in Column E but for 1993, 1995 and 1996, the percentage in Column E was under 1%. Arkansas reported little or no use of animals in Column E for a number of years, and then in one year, 1993, there was a jump to 56.2% and to 21.3% the following year (see Figure 2). The following two years the numbers returned to 0% until 1997 when 35.5% were placed in Column E.

Figure 2. Column E Data from Arkansas, 1993 to 1997

Yearly fluctuations in Column E suggest that either some unusual experiments are being carried out in a particular state, or there are some inconsistencies in the way in which the data are reported by the institutions and the states. If these data are to be at all useful, then some reporting guidelines have to be developed. The task will not be easy, especially since there is room for considerable disagreement on what is likely to cause animal distress or how it should be measured.

At NIH (specifically OPRR), we will encourage the issuing of "best practice" guidelines covering specific techniques. An example of such an initiative is the OPRR letter stating that ascites production causes animal distress and should be used only if in vitro production of monoclonal antibodies is unsuccessful.

  1. Financial Support for Research on P&D
    One of the problems in the field of pain or distress measurement and elimination is that there is virtually no funding to support such studies. Clearly there are difficulties in encouraging agencies to provide funds for projects that deliberately cause animal distress but it may be possible to "piggy-back" such assessments onto ongoing studies that are investigating other topics that have already been approved. The HSUS plans to lobby both private and government entities to make available funds that might be used to develop more sensitive and accurate measures of animal distress that are practical in the laboratory and ways in which such distress can be alleviated.

Best Practices and Policies

Many institutions and animal facilities have developed policies and guidelines in which the pain and distress caused to animals is minimized or alleviated entirely. These documents are usually available only in-house and are not disseminated to other institutions and labs through professional publications. The HSUS would like to promote the dissemination of best practices by encouraging institutions to publicize their efforts on reducing pain and distress in animals used in research. Moreover, The HSUS would also like scientists to seek funding for piggy-back studies on refinement.

In the search for refinement and the development of "best practices", there are some factors that are not obvious that might be taken into account in suggesting refinements. For example, there appear to be strain and sex differences in an animal's experience of pain and distress (Mogil et al, 1997a, 1997b). These papers present data on nociception in two strains of mice, and the possible sex-specific mediation on pain sensitivity that might occur via the opioid receptors.

An analysis of policies covering specific techniques indicates that there may be considerable variation in what is permitted from one institution to another. The HSUS plans to summarize some of the policies on specific techniques and develop a discussion of the different practices. We have begun the process with an analysis of policies on the production of monoclonal antibodies gathered from the World Wide Web (see Table 5). The question is, which one of these policies cause less pain and distress to the animals, and what can be considered to be a 'best practice'? It is clear that more inter-institutional discussion and empirical studies are needed to assist scientists in making a determination on this and other policies.

Table 5: Examples of Policies on Monoclonal Antibody Production*

Penn StateStanfordU IowaU Minnesota
Monitoring subj. w/ solid tumorsNot specified3/wkNot specified3/wk
Primingas low as 0.1 ml pristaneNot specified0.2 ml max pristane0.5 ml max pristane
# of tapsmax 3 taps, last terminalNot specified2 taps, last after euthanasiaNot specified
Monitoring post inoculationdaily3/wk for 1st wk, then dailydailydaily
Replacement fluid after ascite harvestNot specifiedNot specified1-2 ml of saline subcutaneousNot specified
Anesthesia during tapanesthesia can be usedanesthesia used for new personnelNot specifiedNot specified

*gathered from the World Wide Web


The public's support for animal use in biomedical research has declined in recent years. The decrease in support is even more evident when the public is questioned about the experimental use of animals involving pain and/or distress. In this instance, the level of public support decreases significantly when harmful research is conducted on primates, dogs and cats (Plous, 1998). With the public's interest in the humane treatment of animals in laboratories and research, there should be greater attention provided to refining techniques, to publicizing best practices, and to eliminating animal pain and distress. The HSUS campaign seeks to encourage methods of refinement and replacement, with the goal to eliminate all animal pain and distress in research by the year 2020.

What does the research and speculation about animal pain, suffering and anxiety tell us about animal well-being? First, it is clear that we have to broaden our concerns about pain to include a number of other states, such as anxiety and fear, that are capable of producing considerable suffering. Second, as suffering is conceived in the discussion in this paper, it appears as though it may not be distributed as widely through the animal kingdom as our vernacular use of the term might suggest. Damasio (1994), for example, argues that suffering arose in creatures that possess sophisticated neurophysiology/ neuroanatomy capable of large-scale storage (memory) of a multitude of categories for objects and events. These memory capabilities are then available for manipulation and creation of novel solutions.

In the promotion of well-being we have some responsibility not simply to minimize animal pain, distress and suffering but also to enrich and enhance the existence of animals that we use and keep for human benefit. This is what may lie behind efforts to develop environmental enrichment programs for zoo and laboratory animals, and the pressure to change minimum standards of animal care into optimal standards. However, if our understanding of animal pain, distress and suffering is confused and incomplete, our knowledge of what might constitute animal well-being is even more insubstantial.

For the moment, we are left with far more questions than answers. Fortunately, there has been an increase in attention to pain and distress issues within science and academe. The result is a steady trickle of experimental data addressing animal distress and well-being and an increase in the debate about the conceptual issues. These activities will lead to improvements for both animals and the humans that rely on or appreciate them.


  • Bateson, P. (1991). The assessment of pain in animals. Animal Behavior 42: 827-839.
  • Beynen, A.C., Baumans, V., Bertens, A.P.M.G., Havenaar, R., Hesp, A.P.M. and van Zutphen, L.F.M. (1987). Assessment of discomfort in gallstone-bearing mice: a practical example of the problems encountered in an attempt to recognize discomfort in laboratory animals. Laboratory Animals 21: 35-44.
  • Brain, L. (1963). Animals and pain. New Scientist 18:380-381.
  • Byrne, R. (1999). Primate cognition: evidence for the ethical treatment of primates. In Attitudes to Animals: Views in Animal Welfare, ed. Francine L. Dolins, Cambridge University Press, Cambridge, England, pp. 114-125.
  • Carruthers, P. (1992). The Animals Issue. Cambridge; Cambridge University Press, 206 pages.
  • Casey, K.L. and Dubner, R. (1989). Animal models of chronic pain: scientific and ethical issues. Pain 38: 249-252.
  • Cassano, G.B. (1983). What is pathological anxiety and what is not? In The Benzodiazepines: from Molecular Biology to Clinical Practice, ed. E.Costa, pp 287-293. New York, Raven Press.
  • Cassell, E.J. (1982). The nature of suffering and the goals of medicine. New England Journal of Medicine 306: 639-645.
  • Cassell, E.J. (1989). What is suffering? In Science and Animals: Addressing Contemporary Issues, eds. H. N. Guttman, J.A. Mench, and R. C. Simmonds, pp. 13-16. Bethesda, MD; Scientists Center for Animal Welfare.
  • Damasio, A. R. (1994). Descartes' Error: Emotion, Reason and the Human Brain. New York: G. P. Putnam's Sons, 344 pages.
  • DeGrazia, D. and Rowan, A.N. (1991). Animal pain, suffering, and anxiety. Journal of Theoretical Medicine.
  • Derrien, M., McCort-Tranchepain, I., Ducos, B., Roques, B.P. and Durieux, C. (1994). Heterogeneity of CCK-B receptors involved in animal models of anxiety. Pharmacology Biochemistry and Behavior 49:133-141.
  • Dorow, R., Horowshi, R., Paschelke, G., Amin, M. and Braestrup, C. (1983). Severe anxiety induced by FG7142; a beta-carboline ligand for benzodiazepine receptors. Lancet ii:98-99.
  • Eisemann, C.H., Jorgensen, W.K., Merrit, D.J., Rice M.J., Cribb, B.W., Webb, P.D. and Zalucki, M.P. (1984). Do insects feel pain? - a biological view. Experientia 40:164-167.
  • File, S.E. (1987). The search for novel anxiolytics. Trends in Neurochemical Sciences 10: 461-463.
  • Fiorito, G. (1986). Is there pain in invertebrates? Behavioral Processes 12:383-386.
  • Gray, J.A. (1982). The Neuropsychology of Anxiety. New York, Oxford University Press.
  • Handley, S.L. and McBlane, J.W. (1993). 5HT drugs in animal models of anxiety. Psychopharmacology 112:13-20.
  • IASP (1979). Report of International Association for the Study of Pain; Subcommittee on taxonomy. Pain 6:249-252.
  • ILAR. (1992). Recognition and Alleviation of Pain and Distress in Laboratory Animals. Washington, DC; National Academy of Sciences, 137 pages.
  • Kitchen, H., Aronson, A.L., Bittle, J.L., McPherson, C.W., Morton, D.B., Pakes, S.P., Rollin, B.E., Rowan, A.N., Sechzer, J.A., Vanderlip, J.E., Will, J.A., Clark, A.S. and Gloyd, J.S. (1987). Panel report on the Colloquium on recognition and alleviation of animal pain and distress. Journal of the American Veterinary Medical Association 191: 1186-1191.
  • Lummis, S.C.R. (1990). GABA receptors in insects. Comparative Biochemistry and Physiology C 95: 1-8.
  • Mellor, D. J. & Reid, C.S.W. (1994). Concepts of animal well-being and predicting the impact of procedures on experimental animals. In Improving the Well-being of Animals in the Research Environment, editors R.M. Baker, D. Jenkin and D.J. Mellor, pp. 3-18. Glen Osmond, South Australia: ANZCCART.
  • Melzack, R. (1973). The Puzzle of Pain. London, Penguin.
  • Mitchell, R.W. (1990). A theory of play. In Interpretation and Explanation in the Study of Animal Behavior, volume I, editors M. Bekoff and D. Jamieson, pp. 197-227. Boulder, CO: Westview Press.
  • Mogil, J.S., Richards, S.P, O'Toole, L.A., Helms, M.L., Mitchell, S, R., and Belknap, J. K. (1997a). Genetic sensitivity to hot-plate nocioception in DBA/2J and C57BL/6J inbred mouse strains: possible sex-specific mediation by opioid receptors. Pain 70: 267-277.
  • Mogil, J.S., Richards, S.P, O'Toole, L.A., Helms, M.L., Mitchell, S, R., West, B., and Belknap, J. K. (1997b). Identification of a Sex-Specific Quantitative Trait Locus Mediating Nonopioid Stress-Induced Analgesia in Female Mice. J. Neuroscience 17(20):7995-8002.
  • Ninan, P.T., Insel, T.M., Cohen, R.M., Cook, J.M., Skolnick, P. and Paul, S.K. (1983). Benzodiazepine receptor-mediated experimental anxiety in primates. Science 218: 1332-1334.
  • Phillips, M. T. (1994). Savages, drunks and lab animals: the researcher's perception of pain. Society and Animals 1: 61-81.
  • Pitcher, G. (1970). The awfulness of pain. Journal of Philosophy 68: 481-492.
  • Random House Dictionary, unabridged (1967).
  • Reese, W.C. (1979). A dog model for human psychopathology. American Journal of Psychiatry 136:1168-1172.
  • Richet, C. (1908). The Pros and Cons of Vivisection. London, Duckworth, 136 pages.
  • Rowan, A.N., Loew, F.M. and Weer, J. (1995). The Animal Research Controversy: Protest, Process and Public Policy - An Analysis of Strategic Issues. North Grafton, MA: Tufts Center for Animals and Public Policy, 210 pages.
  • Seligman, M.E.P. (1975). Helplessness: on Depression, Development, and Death. W.H. Freeman, San Francisco.
  • Shepherdson, D.(1999). New perspectives on the design and management of captive animal environments. In Attitudes to Animals: Views in Animal Welfare, ed. Francine L. Dolins, Cambridge University Press, Cambridge, England, pp. 143-151.
  • Smith, J. A. and Boyd, K. M. (1991). Lives in the Balance. The Ethics of Using Animals in Biomedical Research. Oxford: Oxford University Press, 352 pages.
  • Stephens, M.L., Mendoza, P., Weaver, A. and Hamilton, T. (1998). Unrelieved Pain and Distress in Animals: An Analysis of USDA Data on Experimental Procedures. Journal of Applied Animal Welfare Science 1(1):15-26.
  • Thrush, D.C. (1973). Congenital insensitivity to pain: a clinical, genetic and neurophysiological study of four children from the same family. Brain 96:369-386.
  • Voorzanger, B. and de Cock Buning, T. (1988). The definitions of pain, stress, and suffering, and the use of these concepts in legislation on animal experiments. Proefdier en Wetenschap No. 1, Leiden University, Leiden, the Netherlands.
  • Wells, M. J. (1978). Octopus. London: Chapman and Hall.
  • Wigglesworth, V.B. (1980). Do insects feel pain? Antenna 4:8-9.
  • Young, J.Z. (1965). The organization of a memory system. Proceedings of the Royal Society, Series B 163:285-320.
  • Young, J.Z. (1971). The Anatomy of the Nervous System of Octopus vulgaris. Oxford: Clarendon Press.
  • Young, R.J. (1999). The behavioural requirements of farm animals for psychological well-being and survival. In Attitudes to Animals: Views in Animal Welfare, ed. Francine L. Dolins, Cambridge University Press, Cambridge, England, pp. 77-100.
  • Wemelsfelder, F. (1999). The problem of animal subjectivity and its consequences for the scientific measurement of animal suffering. In Attitudes to Animals: Views in Animal Welfare, ed. Francine L. Dolins, Cambridge University Press, Cambridge, England, pp. 37-53.

New ALTEX: 2/2018

ALTEX cover 51

Support ALTWEB, Make a Gift
Online Humane Science Course


Building a Better Epithelium: Breaking the Barrier to the Next Generation of Toxicity Testing
March 10, 2018
San Antonio, TX

SOT Satellite Meeting: Updates on Activities Related to 21st Century Toxicology and Related Efforts: Invited Presentations and Open Mic
March 15, 2018
San Antonio, TX

Social Housing Workshop
June 4-5, 2018
Beltsville, MD

2nd Pan-American Conference for Alternative Methods
August 23-24, 2018
Rio de Janeiro, Brazil 

Eurotox 2018
September 2-5, 2018
Brussels, Belgium

20th International Congress on In Vitro Toxicology (ESTIV2018)
October 15-18, 2018

More Meetings...