How to Select Humane Endpoints  | Body Condition Scoring  | Welfare Assessment Scoring Systems 


Evaluating Health and Welfare in Research Rodents

The accuracy and translatability of animal-based research is directly dependent on the health and welfare of the animals used. Animals that are distressed, painful or excessively stressed are not normal and data from these animals are likely to be inaccurate. Investigators are responsible for recognizing when situations and procedures may adversely affect their animals. Sources of adverse effects can include:

  • Genetic background and manipulation
  • Experimental procedures (e.g., substance administration, blood collection, surgery, etc.)
  • Husbandry (e.g., housing, breeding, diet, etc.)
  • Handling (including routine)
  • Transport (including within the research facility)


For each experimental study, the investigator should develop a welfare assessment plan that identifies all procedures and situations may cause distress, pain and/or excessive stress in their animals and indicate any physiological or behavioral signs expected as a result. These signs will be used as indicators when evaluating each animal’s health and welfare. Monitoring frequency and humane endpoints must also be specified for each identified procedure and situation. As the study progresses, welfare indicators may be reviewed and updated as needed to improve effectiveness.

In addition to identifying welfare indicators, investigators must specify what treatments or actions will occur when indicator(s) are noticed. These may include intervention points and/or humane endpoints. Intervention points are specific therapies or treatments provided to an animal when certain clinical signs or indicators appear. For example, fluid therapy may be provided if signs of dehydration are found, or especially palatable food provided if weight loss is >/= ten percent. Humane endpoints (the time at which an animal is removed from the study; usually via euthanasia in rodents) must also be defined.

Selecting Appropriate Welfare Indicators

To be effective, welfare indicators must be objective, easily and reliably recognized, relevant to the project and species and practical to use. Indicators may be behavioral or physiological and can be generally applicable or specific for an individual model or project. Rather than relying on a single indicator, using multiple behavioral and physiological indicators, both general and specific, will provide a more detailed and complete picture of an animal’s welfare and prevent interpretation errors1.

Potential indicators of poor animal welfare can include:

  • Changes in physical condition such as loss of body weight, abnormal coat condition or posture; lameness; excessive licking or scratching.
  • Changes in heart rate, respiratory rate and character, blood pressure, or stress hormone levels.
  • Deviations from “normal” behavior (e.g., apathy or withdrawal, increased aggression, stereotypic behavior). These may also include changes in the use of enrichment or behavioral time budgets (e.g., decreased nest building or excessive sleeping).


Welfare Assessment Scoring Systems

It is often helpful to use a numerical scoring system to assess the health and welfare of animals on study. A form called a score sheet is used to record the health status of individual animals at regular, predetermined intervals. The welfare indicators monitored on a score sheet must be objective, easy to evaluate and relevant to the clinical signs expected for that procedure or situation. Scores from multiple indicators are often added up and the result used to determine whether action is needed according to a predetermined key. Scoring systems must be adapted for each study/model so that appropriate indicators are used.

Investigators are expected to provide training to staff and students who will be using a scoring system to ensure they are competent and able to recognize and deal with welfare issues. Outside observers such as veterinary or IACUC personnel should be able to easily read and understand how scores are determined and decisions made based on the scores received. Analysis of an animal's score sheet must clearly show the effect of the procedure/experiment on the animal's health including a pattern of either recovery or deterioration over time. It is also recommended that a description of the welfare assessment protocol be included in the methods section of peer-reviewed journal papers as part of the scientific method1

Examples of Welfare Assessment Scoring Systems

Detailed examples of scoring systems for various rodent and rabbit models may be found in the online “European Commission Expert Working Group: Examples to illustrate the process of severity classification, day-to-day assessment and actual severity assessment” (2013) document - scroll down the page to "Severity assessment – illustrative examples pdf". These examples are meant to illustrate the development and use of welfare assessment protocols for specific laboratory animal studies. Please refer to this document for detailed information on how scoring systems may be used.


  1. Hawkins P, Morton DB, Burman O, Dennison N, Honess P, Jennings M, Lane S, Middleton V, Roughan JV, Wells S, Westwood K; UK Joint Working Group on Refinement BVAAWF/FRAME/RSPCA/UFAW. June 2010. A guide to defining and implementing protocols for the welfare assessment of laboratory animals: eleventh.


Humane Endpoints

Humane endpoints refer to one or more predetermined physiological or behavioral signs that define the point at which an experimental animal’s pain and/or distress is terminated, minimized or reduced by taking actions such as euthanizing the animal, terminating a painful procedure or giving treatment to relieve pain and/or distress (CCAC). Humane endpoints function as an alternative to experimental endpoints and provide investigators with an effective way to refine their research. The establishment of humane endpoints prior to the start of an experiment allows the investigator to prevent unnecessary animal pain and distress while ensuring accurate and timely data collection.

To be effective, humane endpoints must be clearly defined and based on objective criteria. Non-specific signs of illness such as inactivity, hunched posture or a rough coat are an indication that an animal should be examined more closely. By themselves these signs do not often constitute an endpoint. Familiarity with the animal model in use is necessary to select endpoints that are both humane and scientifically sound. As experience with and data collected from a specific animal model accrue, endpoints can be refined or modified. Further information on humane endpoints may be found on these and other websites:

Humane Endpoints in Laboratory Animal Experimentation

UC Davis Center for Animal Alternatives

USDA National Agricultural Library

Canadian Council on Animal Care

Investigators should include a precise definition of the humane endpoint(s), including specific assessment criteria, when describing how humane endpoints will be used in their IACUC protocols. The frequency of animal observation and assessment must also be clearly stated. Note: Normal, healthy experimental animals must be observed at least once a day. Animals in studies involving pain and/or distress will often require more frequent observations to effectively determine the time at which a specific endpoint has been reached. An appropriate monitoring schedule must be specified in the IACUC protocol for each study. In addition, the IACUC protocol must describe the training provided for personnel responsible for observation and assessment and the action(s) to be taken when an animal reaches a humane endpoint.

Type of Humane Endpoints

Humane endpoints are often based on the following:

  • Clinical signs
  • Pathophysiological changes
  • Behavioral changes
  • Biochemical changes
  • Hormonal changes

The exact time of the endpoint (the point at which an animal is removed from study) will depend on the objective of the experiment but should occur before the onset of distress (i.e., unable to adapt completely to a stressor) or as soon as possible thereafter.

Moribund Animals and Death as an Endpoint

The term moribund refers to an animal that is near death or in the process of dying. Animals in this state are often comatose (unresponsive and unaware of stimuli) and so beyond awareness of suffering. However, an animal may have experienced much pain and distress prior to reaching a moribund state. Stating that animals will be euthanized when they become moribund is not an appropriate humane endpoint as this may not reduce or alleviate any suffering that the animal will experience. The purpose of identifying endpoints is to prevent or minimize animal pain and distress. While certain types of studies have historically used death of the animal as a scientific endpoint, this is now rarely accepted and investigators must present conclusive evidence to support the use of such an endpoint.

How to select and use humane endpoints:

Choose appropriate endpoints that are objective and relevant for the assessment of pain/distress in the species. This may include:

  • Body weight changes
  • External physical appearance
  • Behavioral changes
  • Physiological changes (e.g., body temperature, hormonal fluctuations, clinical pathology, etc…)

Research personnel responsible for observing and evaluating animals must be adequately trained and experienced in the recognition of these signs for the species being used. Especially when using behavioral assessment, personnel must be familiar with “normal” before they can be expected to recognize “abnormal”. Investigators are responsible for ensuring these students and employees are appropriately trained and have the skill and authority to treat or euthanize animals who have reached an endpoint.

Pilot studies (experiments) can be useful in determining endpoints, especially when the effects of an experimental treatment in animals are not well known. They may also function to refine experimental studies by allowing for the establishment of earlier endpoints and provide training for personnel in the recognition of endpoints.

Examples of humane endpoints:

Deteriorating body condition score

  • Objective and easy to use for assessing the condition of animals used in research, especially studies where animals may experience some degree of debilitation as the study progresses.
  • Scoring methods have been developed for many species including mice.

Weight loss

  • Rapid weight loss of 15-20 percent within a few days. This requires frequent monitoring of body weight.
  • Gradual weight loss - over an extended period of time leading to emaciation. The degree of weight loss should be specified in terms of % or quantity (grams, pounds, kg).
  • Note: Certain debilitating conditions such as tumor growth and ascites may mask true weight loss.

Changes in nest building behavior in mice

  • Studies have shown that use of nesting material and nest quality decrease with increasing pain/distress in mice.
  • Mice must be housed individually and provided a sufficient amount of material to construct a nest. In addition, the mouse must build a new nest each day for accurate evaluation to occur (e.g., either replacing the nest with new, unconstructed material or disrupt the nest structure and evaluate the mouse's ability to reconstruct using the used material).

The inability to rise or ambulate

  • Correlates with inability to access food or water.
  • Visually obvious, objective and easy to assess.

Tumor size

  • Usually measured as mean diameter of the mass or tumor volume as a percentage of body weight (i.e. greater than 1.5 cm diameter in mice or greater than 10% of body weight)

The presence of labored respiration

  • The animal shows increased respiratory rate and/or effort. Labored respiration is often accompanied by a strong abdominal component to breathing.


  • The skin looses its elasticity. In a hydrated animal, skin pinched over the back quickly returns to its normal position after it is released. In a dehydrated animal the skin is slow to return to normal or remains tented. Please note, dehydration must be moderate to severe before skin tenting will be noticable. Elderly, very young, obese or thin animals may show altered responses.

Ulcerated, necrotic or infected tumors.

  • The presence of open wounds.


Suggestions for the use of humane endpoints in selected studies:

Chronic studies where some degree of debilitation is expected.

  • Body Condition Scoring
  • Weight Loss
  • Loss of ability to ambulate (inability to access food or water).
  • Labored respiration may be associated with lung pathology or abdominal enlargement placing pressure on the diaphragm.


Experimental Neoplasia

  • Tumor size - tumor volume or mean diameter.
  • Tumor ulceration, infection, or necrosis.
  • For internal tumors body condition scoring may be more useful than weight loss.
  • Loss of ability to ambulate (inability to access food or water).
  • Labored respiration may be associated with lung pathology or abdominal enlargement placing pressure on the diaphragm.


Acute Studies

  • Weight loss
  • Dehydration
  • Loss of ability to ambulate (inability to access food or water).
  • Labored respiration may be associated with lung pathology or abdominal enlargement placing pressure on the diaphragm.


Experimental Surgery

Many of the same signs as listed above in addition to conditions specific to post-surgical infections or other complications including:

  • Pain, swelling, redness or discharge from surgical incisions.
  • Dehiscense (splitting apart) of surgical incisions.

These signs may not by themselves be endpoints but are medical issues that suggest there are deficiencies in surgical techniques or care that require attention.

Body Condition Scoring

Body condition scoring is a health assessment method useful for many species and can be used as a humane endpoint. Body condition scoring systems were originally designed to evaluate fat and muscle development in livestock. Scoring techniques for many species, including rodents, have since been published and are widely used to assess health and fitness in animals. Body condition scoring offers an objective and easy to use assessment method that can be incorporated into humane endpoints. Use of body condition scoring can provide a more accurate determination of health and fitness than body weight measurements, especially in studies where animals may develop tumor masses or fluid accumulation that obscures true weight loss. In addition, body condition scoring is useful in chronic studies where animals may lose muscle mass and fat over time.

A body condition scoring method for mice was developed by Ullman-Cullere, MH and CJ Foltz. Body condition scoring: A rapid and accurate method for assessing health status in mice. Laboratory Animal Science 1999, 49(3):319-323.