Paper: The social and political lives of zoonotic disease models: Narratives, science and policy, Melissa Leach, Ian Scoones
Participants: Bhagya Chengat, Stefano Catalano, Becca Bodenham, Jennika Virhia, Mahbubur Rahman, Laura Craighead
The paper was chosen as something that might hold relevance to the different disciplines represented within the cohort.
In general the group felt the paper’s content was relevant as using modelling techniques to inform policy is highly topical especially in light of recent pandemic threats. The paper highlights the need to combine modelling approaches and sources of data but on the whole it was felt that the paper had a somewhat negative tone. Although the suggestion of triangulation instead of integration was introduced early on no specific suggestions or successful examples of doing so were offered within the text. It was highlighted that the paper made some sweeping unreferenced statements such as suggesting Ebola has ‘outbreaks occurring nearly every year in East and Central Africa.’
When looking at the first example given in the paper of H5N1 the speed of producing a publication after the outbreak was highlighted and discussed. In pandemic situations with potentially massive global impact a quick response is required to inform policy on what actions to take. As highlighted in the paper this is often at the expense of hard data to parameterise models and many assumptions are made. We discussed whether these limitations and assumptions are always properly acknowledged in policy decisions if the suggested outcome is financially profitable to pharmaceutical companies is there a tendency to steer towards certain models suggestive of particular interventions effectiveness? This also led to discussion as to the frequency that modelling informs policy during pandemic versus endemic disease situations. It was highlighted that in pandemics often confidence is gained from a solid numerical answer derived from a complex mathematical model. Discussions concluded that in such a time sensitive situation a broad model with sensitivity testing would be an appropriate first step to identify sparse data areas and to prioritise which data has the greatest impact on model outcome and therefore where to prioritise field work for further data collection if feasible. This led to the discussion of using open source data sets from varied fields to parameterise models, how easily searchable and usable such open sources were however would indicate how useful this may be.
As highlighted at the beginning of discussions there was disappointment that the paper did not highlight possible frameworks for utilising combinations of data sources and modelling approaches from different disciplines or offer examples of it’s effective use. We discussed how you might go about combining qualitative and quantitative data and discussed coding methods for open interview transcripts which most of us were unfamiliar with.
In general the discussion highlighted many issues that hold relevance to our different projects and generally posed lots of interesting questions about influences and actions involved in the science and policy realm. At the end of discussion it was suggested that future papers could be very discipline specific as this would be useful for those unfamiliar with the discipline to gain further insight and ask questions from a different perspective.