Online Issues

<< All Back-issues

<< This Issue's Table of Contents

ILAR Journal V33(3) 1991


Prolonged Water Deprivation: A Case Study in Decision Making by an IACUC
F. Barbara Orlans, Ph.D.
F. Barbara Orlans, Ph.D. is a research associate at the Kennedy Institute of Ethics, Georgetown University, in Washington, D.C. This article is based on a chapter of a forthcoming book entitled In the Name of Science: Issues in Responsible Animal Experimentation by F. Barbara Orlans, to be published by Oxford University Press, New York.

INTRODUCTION

The purpose of this case study is to illustrate how a conscientious Institutional Animal Care and Use Committee (IACUC) tackles a controversial procedure and arrives at a decision. It concludes with a note describing how case studies can be used for training purposes, in order to maintain anonymity, only details relating to humane considerations are described.

A protocol involving prolonged and repeated water deprivation of monkeys recently came before an 1ACUC. The protocol involved adult rhesus monkeys, Macaca mulatta, performing visual discrimination tasks as part of a project investigating the development and alleviation of strabismus in children. The monkeys were deprived of water for 22 out of every 24 hours for each of the 5 test days in any 1 week. After the 22-hour water deprivation period, each animal was placed in a restraint chair and tested for a period of 1 to 1% hours. During the test period, one drop of water was given as a reward each time the monkey performed a required task. During the remaining 45 to 60 minutes of the 24-hour period, the animals were allowed as much water as they needed. On the 6th and 7th day of the week, the animals were also allowed free access to water. The cycle was then repeated.

The IACUC was uncomfortable about the water deprivation aspects of the study but did not question other aspects of the protocol. Committee members thought that the monkeys were being subjected to undue suffering. They questioned whether such a long period as 22 hours water deprivation was either necessary for the protocol or tolerable for the animal. The committee required the work to be stopped temporarily while additional information was sought.

The investigator argued that a 22-hour period of water deprivation was standard practice among investigators, and that any change in protocol would interfere with his baseline data collected for more than 6 years. According to the investigator, the animals seemed to be in satisfactory condition; they maintained satisfactory body weight and none had ever died from water deprivation. He said that other investigators who currently use such periods of deprivation had encountered no problems.

The committee's responses were threefold: (1) The fact that the procedure was standard practice did not necessarily indicate that it was desirable or justified; (2) The claimed interference with baseline data was an untested assumption since alternatives had not, at that point, been tested; and (3) the animals' failure to die was an unacceptable criteria because life can be sustained under intolerable conditions.

The committee conscientiously reviewed this protocol at three meetings over a 2-month period. During this time, it obtained expert opinions, conducted literature searches, and sought out relevant policy statements. Eventually, notwithstanding the investigator's objections, the committee finally disapproved the water deprivation aspect of the study.

The IACUC concluded that prolonged water deprivation was not justified for the following reasons:
(1) Although the animals appeared to be in satisfactory physical condition, they were subject to unjustifiable mental suffering;
(2) The animals' thirst had no relevance to the subject of scientific investigation. This was not research on thirst--thirst was being used merely as an inducement to perform a task. (There was a consensus among committee members that justification for using water deprivation would be enhanced if this were thirst research); and
(3) Alternative methods exist to induce animals to perform tasks.

The committee learned that given free access to water and food, animals will work well for rewards of high preference foods or drinks that are not part of their normal laboratory fare. Rhesus monkeys will work well for rewards such as mandarin oranges, yogurt, Gatorade (a commercially available fruit drink), or malted milk. (For the animals' general health, these nutritionally balanced rewards are better than others such as sucrose pellets or marshmallows). The investigator was advised to redesign his protocol along these lines.

An analysis of the information that helped the committee arrive at this decision is described below. The sources are expert opinions, a literature search, and statements of national policy.

EXPERT OPINION

The committee solicited expert opinions from a number of primatologists, psychologists, physiologists, other scientists, and relevant professional scientific associations. It described the protocol and sought advice. A wide range of opinions emerged. Many who objected to this protocol indicated that there is no rationale for this type of water deprivation. Several scientists wondered whether human beings could tolerate 22 hours a day without water for 5 days a week. Primatologists as a group tended to show particular sensitivity to animal welfare issues involved. They, and others, concluded that the monkeys would suffer severe distress from the long period of thirst followed by drop-by-drop water administration and said they would disapprove this procedure. Several experts questioned whether the daily dehydration would affect the monkeys' vision and, therefore, the researcher's results. Many offered constructive advice regarding alternative procedures involving rewards that could be substituted for extreme thirst (see above).

On the other hand, a number of opinions were voiced in support of the procedure. For instance, a spokesperson from a professional association, after having sampled opinions from association members, reported, "There was no sense (among these scientists) that there was anything wrong with this regimen, not at all." One expert who had used a similar procedure said, "Animals adapt to drinking only once per day, and there is no cause for concern over this protocol."

Among those experts who were tolerant of the 22-hour deprivation, several suggested evaluating the animals at the start of the experiment and then continually monitoring them to be certain they remain in good health. Several experts stressed the importance of monitoring animals individually -- averaging the figures for water intake or urine output of a group of animals would not be satisfactory, because individuals vary greatly. Indeed, care must be used to establish baseline data for an individual animal. For instance, if the relative humidity of the environment is low, or if a particular animal is prone to sneezing (thereby losing water droplets), the animal's water intake would tend to rise. Another factor is the housing arrangement. If the animals are group-housed, the social rank of the animal will influence how much water a particular animal may obtain, because the animals themselves prioritize access to both water and food.

Each individual animal should be checked for continued good health judged by stability in weight; stablility of performance in the experimental protocol; development of signs of dehydration, (e.g. daily total water intake, urine output and specific gravity, and sequential multiple analysis of blood); and the development of signs of stress. Disturbances of normal activities can signal stress, including changes in normal sleep/wake cycles; abnormal levels of social adjustment when placed in a cage with other animals of the same species; and emergence of abnormal behaviors such as foot/thumb sucking, cage chewing, abnormal vocalization, and aggression. These experts recommended that the monkeys should be removed from the study if their health begins to deteriorate.

After listening to the experts, the committee was faced with a difficult choice: (1) Should the procedure be abandoned completely and another more humane method substituted? or (2) Can the water deprivation be made humane enough (with health monitoring) to allow it to continue?

The health monitoring so far conducted by the investigator had been superficial and not the rigorous monitoring suggested by the experts. The committee gave some thought to requiring that the investigator conduct a pilot study of two animals with rigorous monitoring. With this additional information, the committee perhaps would have been convinced that it would be all right for the investigator to continue with additional animals. But this pilot study idea was put aside and, in the end, never acted on because other influential information emerged.

In discussions with the experts, the committee debated whether the investigator should be asked to substitute food deprivation for water deprivation because animals can withstand food deprivation with less stress. However, the committee considered food deprivation a type of punishment and abandoned this idea because of the general principle that rewards should be used in preference to punishments whenever possible.

The experts' information about using high preference foods and drinks as rewards played an important role in the committee's final decision. This significant fact emerged only from communication with experts. Neither the scientific literature nor policy statements provided such information, emphasizing the importance of exploring many avenues of advice.

LITERATURE SEARCH

The committee pursued another avenue of information-gathering by seeking help from the scientific literature, which showed that a complicated set of factors are involved in thirst (see for instance the classic paper of Adolph et al., 1954) and that there is considerable biological variation among individuals within a species· However information on variables that might influence the effects of water deprivation such as species, age, or other factors was not really helpful· In general, the literature provided a sparse source of help. For instance, the committee asked the investigator to provide publications to substantiate his claim that this pattern of water deprivation is not deleterious to the animals· Although similar procedures are used by others, the investigator could provide no published research articles in which the humaneness or justification of this procedure is discussed·

Nothing was found in the literature concerning the effects on animals of prolonged and repeated daily periods of water deprivation· However, in research on thirst, some researchers reported the effect on rhesus monkeys of a single 24 hour period of water deprivation, "The monkeys experienced significant cellular dehydration, as indicated by a 5.8% elevation in plasma osmolarity that exceeded the threshold for thirst, and a significant hypovolemia, as indicated by elevated plasma protein and hematocrit values," (Wood et al., 1982)· These investigators reported that following a single period of 24-hour water deprivation, a "mean volume of 32.1 ±12.2 roi.kgq was drunk over 15 minutes .... The mean weight loss during the deprivation period was 35 ±7.8 g.kg'l. The plasma osmolarity increased from 295 ±2 to 312 ±3 mosmol.kgd ·..The magnitude of the cellular dehydration produced by water deprivation is indicated by the increases observed in plasma osmolarity.., and sodium concentration."

Thus, there appears to be merit in the concerns, voiced by some experts, that cellular dehydration should be recognized as a potentially confounding factor in interpreting the researcher's results.

Committees like to know what other committees have done when faced with a similar situation, and the literature was searched for such information· IACUCs are understandably troubled when they learn that they are making decisions that differ from those made by other committees when reviewing the same procedures· Anecdotal information about inconsistencies of IACUC decisions abound, but to date, scholarship (and therefore accounts in the literature) on analysis of IACUC decisions is in its infancy. In what is probably the first such endeavor, Professor Rebecca Dresser, J.D., reports on a survey in which IACUCs reviewed identical protocols, one of which is similar to the one under discussion in this paper -- prolonged water deprivation that is not part of research on thirst (Dresser, 1989)· In the case described by Dresser, six squirrel monkeys were to be deprived of water 5 days a week except for one randomly timed 30-minute period daily·

Dresser reports that the 31 committees reviewing the case "typically wanted evidence that the monkeys would be adequately hydrated and often asked whether the investigator could substitute positive reinforcement or food deprivation, which is viewed as less detrimental to the animals·" One committee immediately disapproved the water deprivation section of the protocol and others wanted additional information· For instance, one committee required that the investigator justify the need for water deprivation; another wanted documentation that water deprivation is safe and required the investigator to describe the monitoring plan to ensure against dehydration; and another wanted information on how nutritional needs of the animal would be satisfied during water deprivation because they would be unable to eat when their mouths were dry from severe thirst. (This point is of significance for all species of animals and in particular to the case presented in this article because rhesus and other macaques tend to drink while eating· Even within species, wide variations can be found from animal to animal with respect to differences in habits of eating with or without water consumption)· As one would expect, these 31 committees did not show uniform judgments, although there is certainly a sense that prolonged water deprivation raised significant concerns among them.

POLICIES

The committee was grappling with the problem of determining where the burden of proof lies and sought guidance from national policies· Should the burden be to show that the procedure is acceptable, or the reverse, that the procedure is inhumane? This is an exceedingly important and difficult issue to address because the choice inescapably biases the case against the side that must accept the burden of proof· Should one assume that the animal is not necessarily suffering (and, therefore, initiate monitoring procedures), or should one assume that the animal probably is suffering (and, therefore, substitute alternative procedures)? Current public policy offers little guidance on how to approach these issues·

The Public Health Service (PHS) Policy on Humane Care and Use of Laboratory Animals does state, however, that in the absence of established evidence to the contrary, "Investigators should consider that procedures that cause pain or distress in human beings may cause pain or distress in animals" (PHS, 1986)· This statement played a part in influencing the committee's final decision·

The issue of water deprivation as part of a research protocol is not specifically addressed in either the PHS policy or the Animal Welfare Act. However, it is mentioned in several voluntary codes from professional organizations both in the United States and abroad· On the whole, the guidelines that were found tended to place 22 hours of water deprivation into a very questionable category, and some would prohibit it.

Since the investigator was a psychologist, it was appropriate that guidance should be sought from the official code of the American Psychological Association (APA) (1985)· Part of this code states, "Procedures involving extensive food or water deprivation should be used only when minimal deprivation procedures are inappropriate to the design and purpose of the research." Similarly, in another context, the guidelines state, "Psychologists should adjust the parameters of [aversive] stimulation to levels that appear minimal." The investigator was confronted with this statement in the early part of the negotiations with the IACUC. He said that if he used a shorter period of water deprivation, such as 6 hours, then the monkeys refused to perform the required tasks after a relatively short time. The investigator said he needed an hour or more to collect what he considered to be a satisfactory volume of results each day. Thus, it seemed that the investigator's decision was based on convenience to himself --namely, the faster accumulation of data--and not on the benefit to the animals. Several policies in various contexts state that when there is a conflict between investigator convenience and welfare of the animal, then the animal shall take precedence (see for instance the Canadian Council on Animal Care, 1980).

Other provisions of the APA guidelines, not specifically concerning water deprivation per se, are: (1) "Psychologists are encouraged to test painful stimuli on themselves wherever reasonable," and (2) "Whenever consistent with the goals of the research, consideration should be given to providing the animal with control over the painful stimulation." One of the experts suggested that the investigator and the committee follow these provisions as the best guidelines available.

Another set of pertinent guidelines comes from the New York Academy of Sciences which state, in part:

The specifics of the U.S. Department of Agriculture's (USDA's) policies are not included in the Academy's guidelines. However, referring back to the USDA Animal Welfare Regulations, it is found that nonhuman primates must be given water "at least every 12 hours" once having been accepted for transport. Since the monkeys were deprived of water for a period considerably longer than 12 hours, the Academy's waiver would take effect and approval of the full 1ACUC would be required.

There are also relevant policies in both Canada and the United Kingdom. Unlike the guidelines of the New York Academy of Sciences, the national standards of Canada allow no waivers. The Guide to the Care and Use of Experimental Animals of the Canadian Council on Animal Care unreservedly states, "Water deprivation should be used only when thirst is the specific subject of the study (Canadian Council on Animal Care, 1980)." Perhaps the most uncompromising policy of all those encountered is that of The Biological Council (1987) in the United Kingdom. This Council is made up of representatives of ten professional associations, including the British Veterinary Association, the British Laboratory Animals Veterinary Association, the Physiological Society, and others. This policy states, in part, "Water should always be available to all animals." Under these guidelines, the protocol could not be approved.

On the whole, therefore, the voluntary codes reflect the same concerns that were expressed by the IACUC.

CONCLUSION

The committee found that, in its totality, the information from experts, the literature, and policy statements tended to substantiate its original concerns. Though committee members anguished over the resistance encountered from the investigator, they finally came to a firm conclusion that this procedure could not be approved.

The decision-making process followed by a conscientious committee is complex and painstaking. It is important that multiple sources of information be sought because some types of information are available only from certain sources. Throughout the process, inevitably, there was ambiguity on some questions, and many difficult choices had to be made. It is a tribute to conscientious committees that they fulfill their mandated role with such rigor.

NOTE

Case studies such as this one are useful tools for training purposes. They can be used in seminars for students or in training courses for laboratory personnel. To use a case study for these purposes, the participants role-play as IACUC members, and a leader acts as the chairperson. The protocol is described (but not the conclusion) and discussion is opened. The participants are invited to identify and analyze the humane issues and come to a conclusion to approve, disapprove, or modify the protocol. Case studies can help to sensitize people to animal welfare concerns and the dilemmas faced by conscientious committee members.

ACKNOWLEDGEMENTS

The author wishes to thank the following people for their constructive comments on this manuscript: Harry C. Rowsell, O.C., D.V.M., Ph.D., Former Executive Director, Canadian Council on Animal Care; Rebecca Dresser, J.D., School of Law, Case Western Reserve, Cleveland, Ohio; and David M. Moore, M.S., D.V.M., Dip. ACLAM, Department of Pathology, Virginia Polytechnic Institute, Blacksburg, Virginia.

REFERENCES

Adolph, E. F., J. P. Barker, and P. A. Hoy. 1954. Multiple factors in thirst. Am. J. Physiol. 178:538-562.

Animal Welfare Regulations. Code of Federal Regulations, Title 9 (Animals and Animal Products), Subchapter A (Animal Welfare), Parts 1-3 (9 CFR 1-3) (Available from: Regulatory Enforcement and Animal Care, U.S. Department of Agriculture, Federal Building, Room 268, Hyattsville, MD 20872.

APA (American Psychological Association). 1985. P. 6 in Guidelines for Ethical Conduct in the Care and Use of Animals. Washington, D.C.: American Psychological Association. (Available from: APA, Order Department, 1400 North Uhle Street, Arlington, VA 22201.)

Biological Council, The. 1987. Guidelines on the Use of Living Animals in Scientific Investigations. London: The Biological Council.

Canadian Council on Animal Care (CCAC). 1980. P. 44 in Guide to the Care and Use of Experimental Animals, Volume I. Ottawa, Ontario: Canadian Council on Animal Care. (Available from: CCAC, 1105-151 Slater Street, Ottawa, Ontario KIP 5H3, Canada.)

Dresser, R. 1989. Developing standards in animal research review. Am. J. Vet. Med. Assoc. 194(9):1184-1191.

New York Academy of Sciences. 1988. P. 8 in Interdisciplinary Principles and Guidelines for the Use of Animals in Research, Testing, and Education. New York: New York Academy of Sciences.

PHS (Public Health Service). 1986. P. 82 in Policy on Humane Care and Use of Laboratory Animals. Washington, D.C.: U.S. Department of Health and Human Services. (Available from: Office for Protection from Research Risks, Building 31, Room 4B09, National Institutes of Health, Bethesda, MD 20892.)

Wood, R. J., E. T. Rolls, and B. J. Rolls. 1982. Physiological mechanisms for thirst in the nonhuman primate. Am. J. Physiol. 242(5):R423-R428.





Copyright © 2008. National Academy of Sciences.
All rights reserved.
500 Fifth St. N.W., Washington, D.C. 20001.
Terms of Use and Privacy Statement