No abstract

Animal research on anxiety and anxiety disorders relies on valid animal models of anxiety. However, the validity of widely used rodent behavioural tests of anxiety has repeatedly been questioned, as they often fail to produce consistent results across independent replicate studies using different study populations or different anxiolytic compounds. In this study, we assessed the sensitivity of behavioural tests of anxiety in mice to detect anxiolytic effects of drugs prescribed to treat anxiety in humans. To this end, we conducted a pre-registered systematic review of studies reporting tests of anxiolytic compounds against a control treatment using common behavioural tests of anxiety in mice. PubMed and EMBASE were searched on August 21 st 2019 for studies published in English and 814 papers were identified for inclusion. Risk of bias was assessed based on Syrcle’s risk of bias tool and the Camarades study quality checklist on a randomly selected subsample of 180 papers. Meta-analyses on effect sizes of treatments using standardized mean differences (Hedges’ g) showed that only two of 17 test measures reliably detected effects of anxiolytic compounds other than diazepam. Further, we report considerable variation in both direction and size of effects of most anxiolytics on most outcome variables, indicating poor replicability of test results. This was corroborated by high heterogeneity in most test measures. Finally, we found an overall high risk of bias. Our findings indicate a general lack of sensitivity of common behavioural tests of anxiety in mice to anxiolytic compounds and cast serious doubt on both construct and predictive validity of most of those tests. The use of animals to model human conditions can be justified only if the expected results are informative, reproducible, and translatable. In view of scientifically valid and ethically responsible research, we call for a revision of behavioural tests of anxiety in mice and the development of more predictive tests .