enrichment may attenuate anxiety and stress and restore immune response (Benaroya-Milshtein et al. 2004). It can also slow disease progression, a consequence that in some circumstances might interfere with the research aims but that may also provide insights into new or better treatments or new research avenues (Hockly et al. 2002). Enrichment can thus improve welfare, reduce stress, and improve the quality of data obtained from the animals in situations where such enrichment does not compromise the anticipated research outcomes. However, the effect of environmental enrichment on stress responses can vary depending on species or strain, the type of enrichment used, the stressor employed, and the type(s) of stress response(s) evaluated (Bardo et al. 2001; Belz et al. 2003; Green et al. 2002; Lawson et al. 2000; Marashi et al. 2003; Moncek et al. 2004; Roy et al. 2001; Schrijver et al. 2002; Sharp et al. 2005; Tsai et al. 2002).
Ideally, enrichment devices or strategies should draw on previous literature or research that shows that they are beneficial to the animals and have no unexpected adverse effects on their health, and that their use does not jeopardize experimental outcomes and research goals through the introduction of uncontrolled variables, increased variability, and/or inter-experimental variance leading to a need for more animal studies (Baumans 2005; Bayne 2005; FELASA Working Group Standardization of Enrichment 2006). Benefiel and colleagues suggest the need for evidence-based evaluation of “mandatory” enrichment practices for all laboratory animal species (Benefiel et al. 2005). Meier and colleagues have shown that enrichment (in the form of various housing supplements) can increase the acute stress response (as evidenced by elevated heart rate and body temperature) of individually housed mice (Meier et al. 2007). Recent evaluation of the effect of enriched environment on genetically engineered fibulin-4 knockout mice (fibulin-4+/−) has shown that knockouts in enriched cages had fewer disorganized regions on their arterial walls than knockout littermates housed in standard cages. These results suggest that the type of housing environment may interfere with the expected phenotype of genetic manipulations and with the experimental outcomes (Cudilo et al. 2007). However, despite a lack of adequate pilot studies, background data, or published information, even such highly controlled conditions as toxicology studies have effectively adopted appropriate enrichment enhancements (Dean 1999; Turner et al. 2003). Faced with the absence of unequivocal scientific evidence for data-driven enrichment standards and aware of the potential for unexpected consequences by the indiscriminate use of enrichment strategies, the Committee makes its recommendations guided by best practices and expert professional judgment in an attempt to balance the need to safeguard animal welfare while maintaining scientific excellence.