1

Guidelines for Diet Control in Behavioral Animal Studies

Introduction
The purpose of this document is to provide investigators with guidelines for the proper use of diet
control in behavioral studies in which the investigator must control food or fluid consumption and
food or fluid consumption is not the independent variable in the research design. Consideration must
be given for the species, strain or stock, gender, age, prior experimental manipulation, body
condition, and/or hydration status of each animal. In all situations, professional judgment must be
used to ensure the well-being of the animal throughout the period of study.
Behavioral research often requires that an animal perform a task for which it receives a food or fluid
reinforcement.
1,2,5
This situation resembles conditions in the wild, in which animals must forage,
travel extended distances, solve problems, or otherwise work to obtain their food and water. In the
professional judgment of many investigators, veterinarians, and animal behaviorists, performing a
task for reinforcement is behaviorally enriching for laboratory animals. As noted in the Guide, the
least restriction that will achieve the scientific objective while maintaining animal well being should
be used.
8
In the development of protocols
16,17
utilizing food or water control, investigators must
address three fundamental issues: a) the necessary level of control; b) the potentially adverse
consequences of control; and c) the methods for assessing the health and well-being of their animals.
In addition, the species being used and the animals size, age, health status and concurrent
treatments must also be considered. Consideration of these factors can facilitate the establishment
of interventional endpoints to maintain the health and well-being of the animals under study.
6

It is recommended that experimental animals on food or fluid control be weighed several times a
week, ideally before experimental sessions.
7
At a minimum, body weights should be recorded at least
weekly and more often for animals requiring greater restrictions.
6
Written records should be
maintained for each animal to document daily food and fluid consumption, hydration status and any
behavioral and clinical changes which are to be used as criteria for the temporary or permanent
removal of an animal from study.
8
Each ACUC must evaluate the pain-distress categorization of
animals that experience diet control in accordance with USDA Animal and Plant Health Inspection
Service, Animal Care Policy #11
([Link]
Young, developing animals have additional dietary requirements for maintaining their normal rate of
growth. Investigators working with young animals should specifically address, in their animal study
proposal, their expectation for any retardation of growth rate and adult size. Comparisons with litter
mates, with similar control animals, or with growth standards will prove useful when assessing
growth in young, developing animals, and in many genetically modified animal models. In many
situations, when caloric or fluid restriction has been justified in developing animals, animals may
never reach their projected adult size, but will in all other respects develop into normal adults. In all
situations, young, developing animals on restricted food or fluid regimens should be carefully
monitored by investigative, veterinary and husbandry personnel.

Food
Whenever an animal obtains any portion of its diet through food reinforcement, the investigator
must ensure that the sum of the nutritional value of the food earned through reinforcement and of
the food provided "free" (without the necessity of earning it) is sufficient to maintain the animal in a

2
healthy state. When possible, the food reinforcement should be a substance and size that is
sufficiently reinforcing and motivating (e.g. through the use of well-defined pellets containing a
uniform sugar or high-fat component that are commercially prepared for reinforcement procedures)
such that dietary restriction is minimized. Some invariant degree of dietary control is most often
required to provide consistent motivation for reliable control over the experimental behavior, though
specifics may, depend on the species, the behavioral task, and the requirements of the research
design.
1,3

Though ad lib feeding is the normal practice in the vivarium, ad lib feeding is not the norm for rodents
in the wild. Many standard control rats and mice used in biomedical research are sedentary, obese,
glucose intolerant, and on a trajectory to premature death.
14
Three dietary restriction protocols have
been reported to positively impact life-span and overall health of the animals: every-other-day
feeding (unlimited access to regular food pellets, on an alternate day feeding schedule); paired
feeding (equal amounts of reduced-calories pellets by weight as the ad libitum group) and limited
daily feeding (access to food for a limited time window of 1 hour per day). Calorie restriction is an
accepted practice for long-term housing of some species such as some rodents and rabbits, and as an
adjunct to some clinical and surgical procedures
8
. Animals on long term caloric restriction regimens
should be weighed at least weekly, and observed daily for signs of distress. As weight loss is expected
with long term caloric restriction, close monitoring of the animals by the veterinarian, and periodical
assessment of clinical health (e.g., serum chemistry) is advised.
In most cases, some food should be provided every day, unless a specific exception is described in an
approved animal study proposal. Experience has demonstrated that short periods, generally forty-
eight hours or less, of markedly reduced food intake or fasting may be required during the initial
phases of diet control, or after periods of increased food intake (e.g., ad libitum food availability).
Experience has demonstrated no adverse consequences to a short period (e.g. 24 hours even in the
smallest species) without food intake in normal healthy animals.
13
In contrast to ad lib feeding, long
term diet control has been shown to greatly enhance longevity and significantly improve animal
health and well-being. Brief periods of fasting of 24 hours or less, preceded by ad lib feeding of at
least forty-eight hours is not considered to be diet control and does not require the recording of the
animals weight.
Consideration must be given for the species being used and the animals size, age, health status, body
condition and concurrent treatments. However, if any period of markedly reduced food intake or
fasting is required, the principal investigator should provide a clear justification for the reduced food
intake, as well as the extent and duration of food reduction in his or her animal study proposal. It is
recommended that animals be gradually reduced to a target weight and acclimated to the feeding
schedule to mitigate the stress response.
6
Ideally, the diet restriction should be limited so that the
body weight is reduced not more than 10% per week.
Many investigators have maintained normal healthy animals with a 15% weight loss or more.
Therefore, there appears to be a low risk to the animals health in using the 15% weight-reduction
limit. However animals with a 15% weight loss under dietary control may be more susceptible to the
deleterious effects of a short-term fast and may require closer monitoring than other animals.
Animals on diet control should be allowed a short-term unrestricted feeding period prior to any
surgical procedure to avoid the development of hypoglycemia during the recovery period.
When caloric control is an experimental requirement, other aspects of the animals diet should
remain balanced (e.g., vitamins, minerals, etc.). Animals weights must be recorded and weight

3
records must be kept for all animals on dietary control. Weight records must be updated a minimum
of once each week, and should be available for examination by the veterinary staff and the institute
Animal Care and Use Committee. In the case of chronic food restriction, the use of species-, age-, and
strain-specific target growth rates is more appropriate than using a fraction of age-matched free-fed
animal weights as a target.
13
Where these data are not available, if an animal shows a loss in body
weight of more than 15% during the period of study, when compared to the pre-diet control weight
of the animal, the animal must be evaluated in consultation with a veterinarian and, if deemed
necessary, its food increased appropriately unless scientifically justified. The maximum percentage of
body weight loss while an animal is on diet control should not exceed 20% of its initial body weight.
15

Exceptions are allowed if scientifically justified and/or if the attending or facility veterinarian
determines that the weight loss does not endanger the animals health, e.g. for species that
experience seasonal fluctuations in body weight. One example of an exception to the above rule
involves obese animals that are placed on caloric restriction. When evaluating an animal with a 15%
weight loss that was previously obese, the veterinarian may determine a weight of the animal that is
closer to its ideal weight for the animal. In such situations, the animals permanent record must
clearly indicate the weight that should be used rather than their pre-diet control weight for future
15% weight loss assessments. Physical evaluation of the animal by a veterinarian, changes in palpable
muscle mass and evaluation of serum chemistry (e.g., serum protein, albumin levels, etc.) can be
helpful for assessing clinical health in animals under dietary control. Rodents with clinical evidence of
malocclusion should not be entered on studies involving control of food or fluid consumption. Clinical
signs resulting from subclinical malocclusion in rodents may be exacerbated during periods of diet
control and necessitate their removal from study. Pale eye color in albino rodents or low hematocrit
may be attributable to nutritional deficiency anemia. Low blood total solids by refractometry may be
attributable to protein deficiency. In addition, it may at times be helpful to monitor an animal for
signs of ketosis or metabolic acidosis. Special attention should be given to ensure that the diet fed
meets the animals nutritional needs.
6
In general, the total caloric intake of a food-regulated animal is
50-70% of that associated with ad libitum feeding.
6,9
With the exception of short term fasts, it is
recommended that animals should be fed a daily ration containing at least 30% of their minimum
caloric requirements when establishing the target weight.
It may be useful in some long-term research designs involving diet control (often termed chronic
preparations) to intermittently allow animals a period of ad libitum feeding sufficient to establish a
new unrestricted feeding body weight plateau. This may be necessary if the animal stops performing
or the attending or facility veterinarian determines that the animals current weight endangers its
health. When transitioning an animal from a controlled food access paradigm to ad libitum access,
careful monitoring of the animals dietary intake is recommended to aid in the prevention of
deleterious gastrointestinal complications (e.g., bloat in primates and dogs, a condition in which
gastric distension can become life threatening). If animals are subsequently placed on diet restriction,
the new unrestricted feeding weight may be less than the previous one, and a physical and clinical
exam is advisable. It is noteworthy that long-term calorie restriction, at up to 40% restriction in
caloric intake without changes in any other dietary component, significantly extends life span in
several animal models, including mice, rats and primates.
10,11
This dietary manipulation also greatly
reduces the incidence of age-related cancers and age-associated neurocognitive impairment.
12
The
cognitive abilities of rats maintained on restricted feeding schedules or provided access to a running
wheel are superior to rats maintained under standard housing conditions.
14



4

Fluid
As with food intake, whenever an animal obtains any portion of its fluid requirements through fluid
reinforcers in behavioral testing, the investigator must ensure that the sum of the fluid earned
through reinforcement and the supplemental fluid provided outside of the experiment is sufficient to
maintain the animal in a healthy state. Experience has demonstrated that the transition of an animal
to a controlled water access paradigm is best accomplished through a gradual, systematic limitation
of fluid intake over a several-day period.
1
If possible and consistent with the experimental paradigm,
concurrent with the systematic limitation of available free-choice water, animals should be provided
with an opportunity to work for additional water until satiated. In some cases, the restriction often
can be relaxed or reduced after the animal becomes proficient at a given task.
6,7
However as with
control of food intake, some invariant degree of fluid restriction is most often required to provide
consistent motivation ensuring reliable control over the experimental behavior.
Experience has demonstrated that short periods without or with markedly reduced fluid intake may
be required during the initial phases of a research design requiring water control. The duration of the
period will vary with the species and hydration status of the animal. Many, larger species of
nonhuman primates do well with markedly reduced or no fluid intake for periods up to thirty-six (36)
hours, but smaller species, especially some New World species, (such as squirrel monkeys), may be
especially susceptible to the effects of fluid restriction. Experience has demonstrated no adverse
consequences of short periods without fluid intake in normal, healthy animals.
4,13,15
However,
consideration must be given for the species being used and the animals size, age, health status, body
condition and concurrent testing and/or treatments. If any period without or with markedly reduced
fluid intake is required after initial phases, the principal investigator should provide a clear
justification for the reduced fluid intake, as well as the extent and duration of fluid reduction in his or
her animal study proposal.
It is recommended that at the start of a new research protocol the amount of fluid consumed, body
weight and hydration assessment be recorded daily for each animal.
6
Once a baseline fluid intake has
been established on a given task, each animal should be allowed to earn fluids to satiety or to a level
approved in the Animal Study Proposal. As needed, fluid intake should be appropriately
supplemented on a daily basis. In cases in which supplements are required, the minimum amount of
fluids to be provided each day should be equivalent to the amount typically consumed by the animal
when it is permitted to earn fluids to satiety or to the level approved in the Animal Study Proposal. It
is recognized, however, that to ensure the animals welfare and experimental integrity, daily
adjustments in fluid intake are often required during the course of the research. Once an animal has
learned a behavior, the daily amount of fluid provided should be increased to the maximum level that
will ensure adequate and reliable performance of the task.
Vacations from controlled fluid intake paradigms may be advisable in some protocols. A vacation
is a period of time, ranging from a day to a few weeks in duration, when the animal is provided
unrestricted or a markedly increased fluid allocation, commonly >1.5-3 times their routine daily
consumption. In studies utilizing water as a reinforcer, during periods of several weeks duration or
longer, when experimental sessions involving that reinforcement are suspended, gradually increasing
the animals consumption to ad libitum access is recommended. In addition, it is recommended that
animals be provided with additional access to fluid for some period on days when research
procedures are not scheduled, unless scientifically justifiable reasons preclude such fluid

5
supplementation.
7
When transitioning an animal from a controlled water paradigm to ad libitum fluid
access, careful monitoring of the animals dietary intake is recommended to aid in the prevention of
deleterious gastrointestinal complications (e.g., bloat in primates, a described above for food
control). Following a vacation period, an animal may require a period without fluid intake to regain
the motivation to perform their learned task. The ACUC must balance the benefits of a vacation
period with the subsequent need for marked fluid restriction.

Assessment of Adequacy of Fluid Intake
Mice and rats may be acutely water deprived for as long as 24 hours to provide motivation for
drinking or tasting tests. Mice and rats tolerate this without clinically overt signs of physiologic
distress or behavioral abnormalities. Each animal under fluid control must be observed daily for its
health status by the animal care or investigative staff. Normal physiological responses to fluid control
routinely result in robust changes in the animals clinical pathological status. For example, fluid
control will often result in elevated blood parameters (e.g., Hematocrit, Serum Total Protein, etc.),
while physical and behavioral assessment of the animal indicates that the animal is healthy and
adapting normally to the controlled access paradigm.
1,2,22
If at any time the attending veterinarian
determines that an animal is not adapting sufficiently to the controlled fluid paradigm, the
veterinarian will consult with the investigator to develop a plan to maintain the health of the animal.
Even though animals typically learn to work in a manner that earns their entire daily fluid
requirement during the testing session, a number of precautions must be taken to avoid the
detrimental effects of fluid control. The nature (e.g., water, fruit juice) and, if applicable,
concentration of the fluid reinforcement should be specified in the animal study proposal. Daily
records of fluid intake must be maintained and be available for review by the veterinary staff and the
institutional ACUC. The daily record should indicate the fluid earned during the testing session and
any supplemental fluid and/or fruit provided to the animal. Therefore, during acute water
deprivation, if water is available for a minimum of fifteen minutes per day with ad lib food, body
weights records are not required.
15
Similarly, food and water deprivation for periods of 24 hours or
less preceded by ad lib food and water for at least 48 hours do not require body weight records.
However, paradigms that require repeated periods of acute water deprivation over several weeks to
months should be considered to be a chronic controlled fluid access paradigm.
Some animals on a chronic controlled fluid access paradigm may decrease their total caloric intake in
response to changes in their access to water. Because food intake is correlated to the amount of fluid
consumed, monitoring food consumption in addition to body weight can be a valuable tool. In most
cases, the decreased caloric intake is minor and does not result in a body weight loss greater than
fifteen percent. However, in the case of obese animals or those experiencing chronic fluid deficiency,
loss of body weight in excess of 15% has been observed. This weight loss does not pose a problem in
the case of obese individuals, but can lead to severe complications in the case of a chronic fluid
deficiency. Therefore, as a precaution against chronic fluid deficiency during a chronic controlled
access paradigm, the animal's weight must be measured and recorded at no less than weekly
intervals. If an animal shows a loss in body weight of more than 15% during the period of study, when
compared to the pre-diet control weight of the animal, the animal must be evaluated by a
veterinarian and, if required, its fluids or food increased appropriately. Exceptions are allowed only if
the attending or facility veterinarian determines that an animal is adequately hydrated and that the
weight loss does not endanger the animals health or there is a scientific justification.

6
One such exception to the above rule involves obese
18
animals which are placed on fluid restriction.
When evaluating a previously obese animal with a 15% weight loss, the veterinarian may determine a
weight of the animal that is closer to its ideal weight for the animal. In such situations, the
veterinarian must clearly indicate in the animals permanent medical record the weight to be used
rather than their pre-diet control weight for future 15% weight loss assessments.

Summary
It is imperative that investigators, animal care staff and veterinarians working with animals on food or
water controlled access paradigms know the species-typical signs of distress for the animals with
which they are working. Animals routinely adapt well to the research design and display no signs of
distress.
1
Animals must be carefully monitored on a daily basis to ensure that they are healthy,
adapting normally, and consume sufficient food and/or water to maintain good health. Close
monitoring is particularly important when an animal is initially acclimated to food or water control,
during transition back to an ad libitum state or when increasing the difficulty of the behavioral task.
In all situations, the details of the training paradigm used and accountability of the individuals
involved must be clearly outlined in the approved animal study proposal. Experience has
demonstrated that diligent record keeping on the daily food or fluid volume consumed, hydration
status, appearance, general affect, experimental performance, and routine weighing are reliable for
ensuring that subjects remain in a healthy state. Records should be maintained as appropriate,
reviewed regularly, and accessible when necessary. The daily records should indicate the food or fluid
earned during the recording session and any supplemental food or fluid provided to the animal. In
addition, a plan of action, complete with endpoints for therapeutic intervention, should be
considered when the experimental Animal Study Proposal is developed.
6



7
References
1. Toth, L. A., and Gardiner, T. (2000) Food and Water Restriction Protocols: Physiological and
Behavioral Considerations. Contemporary Topics In Laboratory Animal Science, 39: 9-17
2. Claassen, V. (1994) Food and Water Intake. p. 267-287. In J. P. Huston (ed.) Techniques in the
Behavioral and Neural Sciences. Elsevier, New York
3. Weindruch, R. (1996) Caloric restriction and aging. Sci. Am. 274:46-52
4. Hughes, J. E., Amyx, H., Howard, J. L., Nanry, K. P., and Pollard, G. T. (1994) Health effects of
water restriction to motivate lever-pressing in rats. Lab. Animal Sci. 44:135-140
5. Van Sluyters, R., Oberdorfer, M. D. (eds.) (1991) Preparation and Maintenance of Higher
Mammals during Neuroscience Experiments, Report of a National Institutes of Health
Workshop.
6. NRC (2011) Guidelines for the Care and Use of Mammals in Neuroscience and Behavioral
Research. Washington, DC: National Academy Press.
[[Link] ]
7. NIH (2002) Methods and Welfare Considerations in Behavioral Research with Animals.
Washington, DC: U.S. Government Printing Office [[Link]
writing-and-application-process/[Link]]
8. NRC (1996) Guide for the Care and Use of Laboratory Animals. Washington, DC: National
Academy Press []
9. Bucci, T. J. (1992) Dietary restriction: Why all the interest? An overview. Lab. Animal, 21:29-34
10. Heilbronn LK, Ravussin E (2003) Calorie restriction and aging: review of the literature and
implications for studies in humans. Am J Clin Nutr 78: 361-369.
11. Mattison JA, Roth GS, Lane MA, Ingram DK (2007) Dietary restriction in aging nonhuman
primates. Interdiscip Top Gerontol 35: 137-158.
12. Martin B, Mattson MP, Maudsley S (2006) Caloric restriction and intermittent fasting: two
potential diets for successful brain aging. Ageing Res Rev 5:332-353.
13. Rowland, N. E (2007) Food or fluid restriction in common laboratory animals: Balancing
welfare considerations with scientific inquiry. Comp Med 57:149-160.
14. Martin, B., Sunggoan, J., Maudsley, S., Mattson, Mark P. (2010) Control Laboratory rodents
are metabolically morbid: Why it matters. PNAS 107 (14): 6127-6133.
15. Heiderstadt, K.M., McLaughlin, R.M., Wright, D.C., Walker, S.E., Gomea-Sanchez, C.E. (2000)
The effect of chronic food and water restriction on open-field behavior and serum
corticosterone levels in rats. Lab. Animals 34:20-28.
16. Crawley, J.N., (2000) Whats Wrong With My Mouse? Behavioral Phenotyping of Transgenic
and Knockout Mice. Chapter 7. Feeding and Drinking. pp. 131-149.
17. Current Protocols. Current Protocols in Neuroscience. Behavioral Neuroscience. Unit 8, Online
ISBN: 9780471142300.
18. Ullman-Culler, M.H, Foltz, C.J. (1999) Body Condition Scoring: A Rapid and Accurate Method
for Assessing Health Status in Mice. Lab An Sci 49(3): 319-323.
19. Hayes, K.E., J.A. Raucci, N.M. Gades, L.A. Toth (2000) An Evaluation of Analgesic Regimens for
Abdominal Surgery in Mice. Contemporary Topics 39 (6): 18-23.
20. Bachmanov, A.A., D.R. Reed, G.K. Beauchamp, M.G. Tordoff (2002) Food Intake, Water Intake,
and Drinking Spout Side Preference of 28 Mouse Strains. Behav Genet 32(6): 435443.
21. Prescott, M.J., [Link], P.A. Flecknell , D. Gaffan, K. Garrod , R.N. Lemon, A.J. Parker, K.
Ryder, W. Schultz, L. Scott, J. Watson, L. Whitfield (2010) Refinement of the use of food and

8
fluid control as motivational tools for macaques used in behavioural neuroscience research:
Report of a Working Group of the NC3Rs. Journal of Neuroscience Methods 193:167-188.
22. Bekkevold, Christine M; Robertson, Kimberly L; Reinhard, Mary K; Battles, August H; Rowland,
Neil E (2013) Dehydration Parameters and Standards for Laboratory Mice. JAALAS 52 (3): 233-
239.





Adopted by ARAC - 6/13/90
Reapproved - 5/8/96, 2/10/99
Revised - 3/27/02, 3/9/05, 12/18/07, 01/12/11, 12/11/13