Generalizability, Robustness and Replicability When Evaluating Wellbeing of Laboratory Mice with Various Methods

Area of Science:

• Animal research ethics
• Laboratory animal science
• Veterinary medicine

Background:

• Objective assessment of animal wellbeing is crucial for ethical in vivo research.
• Non-invasive methods are needed to reliably measure animal welfare.
• Current methods' generalizability across diverse research models is uncertain.

Purpose of the Study:

• To evaluate the performance of non-invasive methods for assessing animal wellbeing in mice.
• To determine the reliability and generalizability of these methods across different surgical interventions and gastrointestinal diseases.
• To provide data supporting the optimization of animal welfare in research settings.

Main Methods:

• Evaluation of non-invasive methods: body weight, burrowing activity, nesting behavior, distress score, and fecal corticosterone metabolites.
• Assessment in healthy mice, post-surgery, and during disease progression.
• Performance analysis using receiver operating characteristic curves to differentiate healthy from diseased states.

Main Results:

• Method performance varied significantly between different surgical interventions and gastrointestinal diseases, indicating low generalizability.
• Methods demonstrated high robustness within a single gastrointestinal disease model, often independent of drug treatments.
• High replicability was observed when assessing two distinct cohorts with identical surgical interventions.

Conclusions:

• Non-invasive methods can yield valid conclusions on animal wellbeing within a specific animal model.
• The lack of generalizability across models suggests that comparisons using single methods may be misleading.
• A combination of multiple methods is recommended for comprehensive and accurate animal welfare assessment in research.