Good #Science Is Good Science | Boston Review
▻http://bostonreview.net/science-nature/marc-lipsitch-good-science-good-science
I share Medawar’s pragmatic vision of scientific reasoning. Scientists must resist the temptation to excessive skepticism : the kind that says no evidence is ever quite good enough. Instead they should keep their eyes open for any kind of information that can help them solve problems. Deciding, on principle, to reject some kinds of information outright, or to consider only particular kinds of studies, is counterproductive. Instead of succumbing to what Medawar calls “habitual disbelief,” the scientist should pursue all possible inputs that can sharpen one’s understanding, test one’s preconceptions, suggest novel hypotheses, and identify previously unrecognized inconsistencies and limitations in one’s view of a problem.
This conception of science leads me to disagree with some elements of the philosopher of medicine Jonathan Fuller’s recent essay about two sects within epidemiology, defined by what kinds of evidence they consider meaningful and how they think decisions should be made when evidence is uncertain. Fuller sees in the contrast two “competing philosophies” of scientific practice. One, he says, is characteristic of public health epidemiologists like me, who are “methodologically liberal and pragmatic” and use models and diverse sources of data. The other, he explains, is characteristic of clinical epidemiologists like Stanford’s John Ioannidis, who draw on a tradition of skepticism about medical interventions in the literature of what has been known since the 1980s as “evidence-based medicine,” privilege “gold standard” evidence from randomized controlled trials (as opposed to mere “data”), and counsel inaction until a certain ideal form of evidence—Evidence with a capital E—justifies intervening.
Fuller rightly points out that this distinction is only a rough approximation; indeed, there are many clinical epidemiologists who do not share the hardline skepticism associated with the most extreme wing of the evidence-based medicine community. But the distinction is also misleading in a subtle way. If the COVID-19 crisis has revealed two “competing” ways of thinking in distinct scientific traditions, it is not between two philosophies of science or two philosophies of evidence so much as between two philosophies of action.
In March, as health systems in Wuhan, Iran, and Northern Italy teetered under the weight of COVID-19 cases, Ioannidis cautioned that we really didn’t know enough to say whether a response was appropriate, warning of a “once-in a-century evidence fiasco” and suggesting that the epidemic might dissipate “on its own.” (I replied to that argument, explaining why we do know enough to act decisively against this pandemic.) To my knowledge, Ioannidis has never stated that early interventions should have been avoided, but by repeatedly criticizing the evidence on which they were based, he gives that impression.