Stress and distress: a discussion by the Refinement and Enrichment Forum

SHEILA ROBERTS, Duke University Medical Center, Durham, USA
PASCALLE VAN LOO, Utrecht University, The Netherlands
POLLY SCHULTZ, Wisconsin Regional Primate Research Center, Madison, USA
AMY KERWIN, Primates Incorporated., Madison, USA
EMILY PATTESON-KANE, Purdue University, Wast Lafayett, USA
CHRIS SHERWIN, University of Bristol, Bristol, UK
ANITA CONTE, College of Staten Island, Staten Island, USA
JONATHAN BALCOMBE, Physicians Committee for Responsible Medicine, Washington, USA
DAVID MORTON, University of Birmingham, Birmingham, UK
VIKTOR REINHARDT (Moderator), Animal Welfare Institute, Washington, USA

"The terms 'tress' and 'distress' are often used in the scientific literature but usually without a definition. If you use these terms, how do you define them? What are the signs that tell you that an animal is stressed or distressed?" (Reinhardt)

"I use the term 'dystress' for stressors that an animal cannot cope with physiologically.1 This situation is indicated primarily by persistently raised corticosteroid levels and, more importantly, an increased number of corticosteroid receptors. Secondary clinical signs such as increased blood pressure, and pathological signs such as gastric ulceration often appear in association with dystress. The term 'distress' I reserve for a stressor that an animal cannot cope with mentally.1 The 'psychological well-being' of a distressed animal is impaired. Distress typically manifests in conspicuous changes in the subject's behavioural expressions. The species-typical repertoire becomes less diverse, certain behaviour patterns occur more frequently while others are performed in a stereotypical manner. Species-atypical, so-called abnormal behaviours develop in many species." (Morton) "In the context of distress, `abnormal behavior' is also referred to as `maladaptive behavior'2 to distinguish it as a direct response to a stressor to which the subject is unable to adapt." (Reinhardt).

"I don't work with these terms very much but I would suggests that `stress' (eustress) is the arousal above the resting or baseline state, while `distress' implies subjective feelings of anxiety, frustration or fear." (Patterson-Kane) "Depending on the species, these feelings are expressed in maladaptive behaviors such as stereotypical movements and gestures, self-injurious feather picking or self-injurious biting (Figure 1), and hyperaggression." (Conte)

Figure 1 Self-injurious biting occurs in about 10% of macaques who are caged permanently without companionship.3,4 This behavioural pathology is a sign that the subject is distressed,2 i.e., unable to adapt to the stressor of social deprivation.

"Since both terms are often used in regulatory texts, I have also adopted them and use them in the following scheme: An external situation or internal factor (stressor) leads to stress (eustress in contrast to distress) which implies an alteration of the subject's physiological equilibrium (e.g., increased heart rate, increased corticosterone output) and behavioral equilibrium (e.g., fear reaction, struggling, temporary diarrhea, defensive aggression (Figure 2), etc.). Eustress is not harmful to the subject, BUT it disturbs the subject's equilibrium, hence has the effect of a potentially data-biasing variable that needs to be accounted for in the research context.

  Figure 2 Defensive aggression is a common stress response of confined animals who are cornered by approaching personnel and hence experience simultaneously intense fear and aggressive arousal.5

If the subject cannot adapt to the stressor, i.e., return to physiological and behavioural equilibrium, stress becomes distress. Distress is the inability to adapt to a stressor (e.g., permanent social deprivation of social animals (Figure 1); repeated enforced immobilization; repeated exposure to pain; repeated exposure to a fear-inducing situation). Distress harms the subject. Pathophysiological processes (e.g., chronic diseases, generalized alopecia or depression) and maladaptive behaviors (e.g., self-injurious biting, hair-pulling, stereotypical movements) typically develop as a result of distress." (Reinhardt).

"Stress does not need to compromise animal welfare. Distress, however, may disrupt biological functions critical to the subject's well-being." (Balcombe)

"I see an animal as being stressed when it engages in non-injurious stereotypies, is restless and conspicuously alert and shows little interest in food treats that it may accept but will not eat. A distressed animal gives the impression of being in panic. Its eyes are wide open, it stays in the back of the cage and persistently refuses to take, let alone eat otherwise preferred food treats. Stress is not a serious state while distress puts the animal at a high risk of physiological imbalance that compromises its physical and psychological well-being." (Anonymous)

"If you know your animals through direct experience you also `know' when they are stressed and when they are distressed, even though you may not find the words to describe exactly the signs of stress or distress. What really matters is the first-hand experience with the animals. Second-hand knowledge derived from books is of little use here." (Reinhardt)

"Referring to my experience with macaques I assume that an animal experiences stress when being introduced to an another unfamiliar macaque, similar to how a human might feel the first day on a new job, or going on a first date. Stress can be positive OR negative, but either way can cause physiological responses. A certain amount of stress is part of life for any creature. This kind of stress is probably best termed `eustress' as opposed to `distress'. A macaque would probably experience distress when being confronted with a chimpanzee, similar to how a human would feel when being confronted with a lion. The total lack of control over a life-threatening situation is distressing for all animals. Unlike eustress, distress is always negative and is more likely to cause serious physiological disturbances." (Schultz)

"When I worked with rhesus macaques in a laboratory setting, I noticed that the monkeys encountered acute stressors on a frequent basis. The acute stressors included unfamiliar personnel entering the room, personnel using a net, pole or glove to coerce monkeys to enter transport boxes, and personnel using enforced restraint for blood draws and injections. Struggling (during restraint), fear grimacing, pacing, open-mouth threatening, and/or alarm vocalizations were the reactions to such stressful situations. Occasionally, an acute stressor triggered a brief gesture, such as abruptly biting a limb, that the monkey didn't typically display.

Confinement in a species-inadequate environment was a chronic stressor to which many of the rhesus macaques could not adapt. They developed maladaptive behaviour patterns, such as stereotypical pacing, back-flipping, self-biting (Figure 1) and fur-plucking, all signs I interpreted as psychological distress. The frequent occurrence of acute stressors presumably contributed to the distress that the animals were experiencing." (Kerwin)

"While stress is a state in which the animal is highly alert and ready to respond to a disturbing, potentially threatening situation, distress is an aversive state in which the animal is pushed beyond his/her limit of adjustment and starts to show clinical and behavioural signs of dis-ease. Stress may hurt, but it does not harm as distress does. Being approached by unfamiliar personnel is a typical stress situation, being forcefully immobilized for several hours is a typical distress situation for animals and humans alike." (Weilenmann)

"I think of a stressed state as one in which an animal's emotional and physical states are elevated above baseline, but the elevation is usually minor and/or brief. I think of distress as a word used to describe an animal who is experiencing excessive stress or continuous stress. Although both `stress' and `distress' have negative connotations, I think distress is always bad, but stress can be both good or bad. As already pointed out, a certain amount of stress is part of life for any creature, and I think some types of stress are exciting and good for animals since mild stressors can make life a little more interesting. To use an example mentioned several times already (regarding NHP's), introducing a new cage-mate probably causes some stress, but, assuming the NHPs are compatible, this is a good stress, as it breaks up the monotony and probably leads to good, albeit elevated, emotions. However, when stress gets out of hand, either because of its intensity, frequency, or harmful nature, that is when I say the animal is distressed. In practice, I think distress requires action to alleviate, but stress usually does not. I even consider some level of stress as normal, and, depending on the study, research conducted on animals experiencing normal levels of stress may be more biologically relevant than research conducted on animals shielded from all stressors (if it were even possible to do so). However, it is important to be aware when stress is present, since it could affect data, and also, some stress can develop into distress.

How do I recognize stress and distress? Not having the option of assessing physiological stress-sensitive parameters such as blood pressure and glucocorticoids, it is difficult for me to tell when an animal is stressed based on my definition. I generally have to rely on intuition and anthropomorphism. In most cases, I make a note of when I think an animal has reason to be stressed and keep an eye on him/her, and I try to mitigate stressors that I think could lead to distress. There are some physical manifestations that I regard as indicative of stress rather than distress. These comprise ordinary reactions and coping responses to brief stressors, such as alarm-vocalizing and hiding in the back of the cage when personnel enters the room. When these responses are very frequent and persistent, severe (e.g., depression), or detrimental to the animal's health (e.g., self-mutilation, chronic diarrhea, anorexia) then, to me, the animal is distressed.

There is a gray zone in which I'm not sure if an animal is exhibiting stress or distress. For example, an animal who bites himself only once or twice a month in response to a particular stressor is of great concern, but this may not be something that needs drastic action since the response is fleeting rather than habitual. Of course, in many cases, animals in the `in-between' category are really on the verge of becoming distressed, so I definitely watch them very closely and try to prevent this from happening." (Roberts)

"It is commonly argued that good scientific methodology is characterized by a low variance of the test response. Is anybody aware of articles demonstrating that data variability is higher in animals who show physiological signs of stress than in `control' animals who show no signs of stress?" (Reinhardt)

"There is a whole area of study (i.e., fluctuating asymmetry) which is based on the hypothesis that environmental stressors cause deviations from perfect bilateral symmetry. Wouldn't this mean that stressed animals would show greater variance in symmetry?" (Sherwin) "Yes that is logical, but nobody has yet taken the trouble to support this hypothesis with actual data. To my knowledge there is only one article that implicitly concludes from scientific data that stress can increase the variation of a biological test response: Chance (1956) studied rats and noticed that the variation in ovary weight (test response) was greater if the animals' cages were small, if there was frequent disturbance by changing cages and cage-mates, and if the rats were caged either singly or in groups larger than two.6"(Reinhardt)

"It would be great if indeed we could show that stress increases variation, but I think this is wishful thinking. Several investigators have examined if enrichment increases variation or not. However, as far as we can see now, whether variation increases or decreases depends very much on the parameter we are measuring. With the danger of playing the devil's advocate, I would argue that for some stress parameters, also a low variance of test response might mean that you do work with stressed research subjects. Acute stress in mice caused, for example, by weighing or injection increases the heart rate to about 750-800 BPM. However, the heart rate cannot increase much above 800, since this would lead to a cardiac arrest. So any stressor that prompts a very high heart rate is likely to result in a relatively low variation compared to a stressor that leads only to a moderate increase in heart rate but a normal, i.e., unrestricted variation of data points. This `ceiling effect' of extreme values probably applies to most stress sensitive parameters, including respiration rate, core temperature, corticosterone, prolactin and glucose. The question now arises, what circumstance is 'best science':

1. moderately stressed animals, higher variation of data points hence more animals needed, or

2. highly stressed animals, but lower variation and hence fewer animals needed to achieve statistical significance of the results?" (van Loo)

"Variation of the test response per se is certainly not `bad', as it is a natural phenomenon. A stressful situation is a temporary life-given challenge to the subject. Each individual tends to respond to a challenging, i.e., stressful situation such as injection or venipunture differently depending on past experiences, age, gender and personal disposition. So stress-related variation of test responses are biologically normal.

If we can create `perfectly standardized' housing and handling conditions for our animals, we may be able to eliminate stress-related variation of the test response altogether, but are we now dealing with little test machines that yield results that are so `clean' that they no longer reflect biological processes? Where is the boundary?" (Reinhardt)


1 Morton, D.B. (1997). Ethical and Refinement Aspects of Animal Experimentation. In: Veterinary Vaccinology. (Pastoret, P.P., Blancou, J., Vannier, P., and Verscheuren, C.). Elsevier, Amsterdam, 763-785

2 Institute for Laboratory Animal Research (1992). Recognition and Alleviation of Pain and Distress in Laboratory Animals. National Academy Press, Washington, DC

3 Novak, M.A., Kinsey, J.H., Jorgensen, M.J., and Hazen, T.J. (1998). Effects of puzzle feeders on pathological behavior in individually housed rhesus monkeys. American Journal of Primatology 46, 213-227

4Alexander, S., and Fontenot, M.B. (2003). Isosexual social group formation for environmental enrichment in adult male Macaca mulatta. AALAS [American Association for Laboratory Animal Science] 54th National Meeting Official Program, 141

5 Reinhardt, V., and Reinhardt, A. (2001). Environmental Enrichment for Caged Rhesus Macaques (Macaca mulatta) - Photographic Documentation and Literature Review (Second Edition). Animal Welfare Institute, Washington, DC

6 Chance, M.R.A. (1956). Environmental factors influencing gonadotrophin assay in the rat. Nature 177, 228-229

Reproduced with permission of the Institute of Animal Technology.
Published in Animal Technology and Welfare 5 (2), 99- 102 (August 2006).