Strengthening the evidence-base in severity assessment by systematic literature analyses

• Applicants: Professor Dr. André Bleich; Professorin Dr. Heidrun Potschka
• Subject Area: Molecular and Cellular Neurology and Neuropathology ; Sensory and Behavioural Biology ; Veterinary Medical Science
• Term: since 2017
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 321137804

Systematic reviews (SRs) use a transparent and replicable methodology to retrieve, select and analyse all relevant publications on a specific topic. While SRs are common in clinical research, and accepted to represent the highest level of scientific evidence, animal welfare-related SRs remain relatively scarce. In the preceding funding period, the SR-part of P16 has contributed to method development for SRs of animal studies, and used these methods in numerous reviews relevant to severity assessment, in close collaboration with multiple consortium partners. Because the consortium’s experimental studies can only assess severity-related outcomes for a selected number of interventions in certain experimental set-ups, and in a limited number of animals, animal strains, animal age groups and laboratories, SRs are crucial to fill the remaining knowledge gaps. Our SRs analysed various parameters relevant to pain and severity assessment, e.g., corticosterone, burrowing, nest-building, and mouse and rat grimace scales, which complemented the experimental findings by the consortium. In funding period 3, we will further use the high level of evidence generated through SRs to enable overall validation of parameters and scoring systems critical for severity assessment in the frame of FOR 2591. Our analyses will focus on e.g., corticosterone, body weight and various clinical scores. These data will be extracted from relevant literature systematically collected by FOR 2591 for previous SRs. To effectively expand the consortium’s primary data, we need to validate and further develop approaches tackling the challenges related to SRs focussed on severity assessment. Among other things, we will test (combinations of) proxy measures to counteract poor reporting of severity assessment in publications of animal studies; we will compare convenience sampling strategies for efficiency and coverage to counteract poor indexing of severity assessment in the large medical literature databases (Embase, PubMed); and we will determine the efficiency and accuracy of artificial intelligence (screening prioritisation algorithms) to decrease the workload associated with screening large numbers of references. In addition to this tailored method development for severity assessment SRs, two selected case studies will result in knowledge about severity associated with food and water restriction and different blood sampling techniques, for which literature data will be compared with normative severity scoring. Besides this work, we will continue to contribute to all SRs throughout the consortium to complement the experimental data. By using these strategies, P16b will complement the data science approaches for validation of parameters and methods as elaborated by FOR 2591. Synthesising this evidence from groups and models outside FOR 2591 is vital to formulate generally applicable guidelines for evidence-based severity assessment.

• DFG Programme: Research Units
• Subproject of: FOR 2591: Severity assessment in animal-based research