 |
| Professor Jonathan Michaels |
This blog post originated in the Journal of Medical Ethics. The published version is available at: https://blogs.bmj.com/medical-ethics/2020/05/29/evidence-based-injustices/
In healthcare, and many other areas of endeavour, policy and guidance claims legitimacy on the basis that it is evidence-based and follows the best scientific advice. Expert advisory committees collect, consider and interpret extensive, and often complex, scientific evidence. As we have seen in the diverse responses to Covid-19, evidence and expertise does not necessarily lead to purely ‘objective’ responses or unequivocal decisions. Instead, it may provide a veneer of scientific authority that masks a host of underlying subjective influences, uncertainties, biases, and value judgements.
Over my career as a vascular surgeon I developed an interest in decision theory, perhaps in the expectation that it would give me reassurance in the complex decisions affecting life and limb that the specialty demands. Instead, rather than providing clarity in such questions, it helped to highlight the underlying structure of decisions, which depend both upon predictive evidence and the value judgements inherent in dealing with uncertainty and the necessary trade-offs between potentially conflicting objectives. Through this interest I became involved in the development of guidelines and technology appraisals carried out by the National Institute of Health and Care Excellence (NICE) in the UK.
Over the years, I observed the growing political and commercial influence on all aspects of such processes, from the selection of subject matter, through the generation and interpretation of evidence, to the value frameworks and decision-making criteria that determine the final outcomes. Complex decision-making processes, a culture of expertise, formal and bureaucratic processes and an extensive ‘case-law’ have developed over the years. This has the effect of excluding those without the necessary background knowledge, economic resources and academic credibility from contributing to, or challenging, such decisions.
A few years ago, I came across Miranda Fricker’s description of epistemic injustice. She suggests that “there is a distinctively epistemic genus of injustice, in which someone is wronged specifically in their capacity as a knower”. This results in individuals or groups being systematically disadvantaged through exclusion from the processes that generate the knowledge upon which so many decisions depend. This resonated with my own experience of the way in which evidence-based principles appear to be applied in healthcare decisions, particularly those involving rationing of scarce resources.
In exploring the principles of distributive and procedural justice that underpin healthcare policy and guidance, I highlight the factors that may lead to injustices and consider how these might be addressed. Evidential failures may occur due to distortion in the selection of subject matter for research, the limitations of preferred research methods, the choice of comparators and outcome measures, or the reporting and interpretation of results. Epistemic exclusion may result in the failure of guidance to reflect the interests and values of particular patient, service-user, societal or professional groups. Opportunities to contribute to or challenge decisions may be inaccessible to those most likely to be disadvantaged by the outcome.
Apart from the many underlying biases and distortions that are inherent in the published evidence, there are implicit and explicit value judgements in the decision-making processes that may remain obscured or reflect unwarranted epistemic privilege assigned to particular groups. Some of the potential remedies, such as the regulation of trial design and transparent reporting, are well documented, but still fail to be implemented. Significant improvement in other aspects may require a fundamental shift in the research agenda to reflect areas of societal concern, greater emphasis on independent evaluation of all the evidence, including explicit adjustment for likely biases, and a means to represent the interests of marginalised or excluded stakeholders.
Download the full paper from the Journal of Medical Ethics:
https://jme.bmj.com/content/early/2020/05/26/medethics-2020-106171