Amalgamating Evidence About Causes: Medicine, the Medical Sciences, and Beyond

点击翻译按钮获取中文摘要

In many areas of science, one can rely on a variety of evidence from different methods, experts, and disciplines to derive causal statements. Accordingly, the adequate amalgamation of evidence to produce causal knowledge is a widespread challenge for scientists and those aiming to rely on scientific causal claims in decision-making. This is particularly true in the biomedical sciences and medical practice. Here, there are at least four particular areas in which practitioners must synthesize causal knowledge: (i) treating a series of patients in routine clinical practice, (ii) measuring individual effect sizes from multiple medical trials and combining them into an overall effect size, (iii) drawing conclusions about intervention effects based on widely varying evidence, and (iv) synthesizing the judgments of a group of experts. In each area, the evidence for presumed causal relationships has different forms and properties, as well as different levels of reliability, and the way in which this different evidence can be synthesized varies from area to area. It turns out that there are numerous controversial and unresolved debates in each of these areas. One might be tempted to think that these are purely scientific issues that could be resolved by technical developments in medicine or statistics. In fact, however, the existence of these unresolved debates is indicative of deeper philosophical challenges. Accordingly, the goal of this project is to systematically explore steps toward a better understanding of the range of possibilities for amalgamating evidence of causes in each of the four areas described above, using the tools of philosophy of science. In particular, we aim to formulate formal constraints, prescriptive principles, and methodological heuristics that could guide practitioners in medicine and be used as evaluative norms in this practical, policy-oriented context. To do this, we use case study investigation, formal work, and computer simulations.