Periodic Reporting for period 1 - PreclinSR (Increasing clinical translation of experimental stroke research: new approaches to systematic review)
Okres sprawozdawczy: 2016-11-01 do 2018-10-31
Stroke is a leading cause of death and disability; despite substantial investment in preclinical research and clinical trials, few effective treatments are available. A possible reason for this is that preclinical data are not used efficiently in the design of clinical trials, through a lack of rigorous critical appraisal or objective assessment of the preclinical evidence offered. The objectives here were firstly to establish a role for network meta-analysis in establishing the effectiveness of novel compounds in the context of the evidence for effectiveness of existing compounds, to establish whether the new compound offers material advantages; and secondly to establish a framework for the collation of such data, collected in the context of systematic review, using a common standard format to ensure inter-operability.
The network meta-analysis approach was first piloted in a review of the protective effect of anaesthetics in rodent models of traumatic brain injury, and that work is now published (doi 10.1016/j.bja.2018.07.024). Application of the approach in the stroke literature, and a collaborative review article setting out the strengths and weaknesses of this approach, are close to submission. We have collated and cleaned data for outcomes assessed in over 3500 publications describing animal models of stroke. These data are currently in the last stages of validation (using text mining approaches) after which they will be made available in standard ontology, identified with a DoI and published in the Edinburgh Data Archive. Accompanying this we are in the late stages of developing data visualisation tools which will be made available online through the CAMARADES website.
The most exciting outputs are firstly the demonstration of the feasibility of network meta-analysis as applied to data from animal studies; and secondly the application of standard ontologies and data visualisations to information already collected in the context of systematic reviews. While funders have been hesitant to recognise the transformative potential of this approach, we believe that linking these tools to emerging technologies in living systematic reviews will have a major impact, allowing researchers access to curated, up to date summaries of existing knowledge annotated for the risks of bias in individual studies.