Chartered Institute for IT initiative follows concerns over Imperial College modelling on Covid-19 lockdown
The computer code behind the scientific modelling of epidemics should meet independent, professional standards to ensure public trust, according to IT experts following controversies over Imperial College London’s pandemic modelling by epidemiologist Neil Ferguson.
In a policy paper, BCS, the Chartered Institute for IT, said that the quality of software behind scientific models appears to “rely too much on the individual coding practices of the scientists who are not computer scientists”, rather than “professional software development practices being publicly evidenced against appropriate standards”.
Ferguson team’s modelling that informed the UK lockdown in March when it predicted up to 510,000 Covid-19 deaths without mitigation measures faced a range of criticisms, including that it was not openly available for scrutiny from the outset and, later, that the code initially developed for flu modelling is based on an old programming language and may not be robust enough or applicable to coronavirus.
“The affair has seen legitimate scientific concerns and debate mixed up with efforts to undermine the lockdown and deflect responsibility for policy decisions,” Neil Chue Hong and Simon Hettrick the director and deputy director, respectively, of the Software Sustainability Institute wrote recently in an article for Research Professional News.
Now, Bill Mitchell, director of policy at BCS, says that “the politicisation of the role of computer coding in epidemiology has made it obvious that our understanding and use of science relies as much on the underlying code as on the underlying research”.
The BCS paper published on 28 May says: “We believe professional software development standards should be adopted when implementing computational methods for conducting scientific research where that research could be relied on by policymakers and which could have critical consequences for society, such as for example healthcare, criminal justice, or climate change, amongst others.”
The underlying software code should be open sourced, it added.
According to BCS, professionalising and using best practice software development in scientific research would have many benefits, including the ability of researchers to “share, combine, adapt and build upon software implementations of computational models, no matter whether they are in the same discipline, institution or country”.
It would also enable scientists to correctly modify software implementations of computational models in times of crisis “as rapidly as possible” as well as facilitate “reproducibility of research findings”.
Finally, the institute says, it would provide “reassurance to the public that policy decisions are based on robust evidence of the highest quality”.
BCS added that it would approach a number of organisations to discuss how to professionalise software development practice in scientific research, including the Centre for Data Ethics and Innovation, the Alan Turing Institute, the Cabinet Office, UK Research and Innovation and Public Health England, among others.