Do we, as members of science-based professions, misuse science? Prominent scientists may take on a position akin to religious leaders in certain pockets of society. This practice of holding beliefs as dogma is called scientism and it can be problematic. Science can be a useful way of understanding the world, but it isn’t perfect. There are numerous ways science can be misused and much of these boil down to a poor understanding of what science represents. This blog is not intended to dissuade you from using science or evidence in your practice; that - in my humble opinion - would be insane! This blog is to explain what science is and what it isn’t. I will give a few examples of the misuse of science. I will summarise by presenting solutions to the issue of science misuse.
Science isn’t perfect
Science isn’t a perfect system for understanding the world, it’s just the best we have. Although science represents a noble pursuit of the truth there have been prominent figures - within science - who vehemently argue this is not the case.
John Ioannidis argues, in his standout 2005 paper, that most published research findings are false (Ref). Science, for the most part, is theory driven and theories can be seen as a vehicle to explain your results. Research findings can fit a hypothesis, but isolated results never prove a theory. Ionnidis explains that research findings are a product of the pre-study odds and lead to post-study probability. Lower pre-study odds means a lower post-study probability of positive research findings. Theory driven research is rarely directly replicating previous findings so most studies have very low pre-study odds, meaning the probability that the positive research finding is true is actually very low. This is an important realisation, as it opens the door to doubt many scientific ‘doctrines’ with a healthy, humble dose of skepticism.
Ionnidis also argues that many research findings can be viewed as a reflection of the prevailing bias of the time. This argument is reflected in Thomas Kuhn’s revolutionary book (originally published in 1962) The Structure of Scientific Revolutions. Kuhn originally presented the viewpoint that many scientific discoveries are simply a reflection of the prevailing ‘scientific paradigm’ rather than, as they are typically portrayed, a slow accumulation of facts.
Scientists can misuse science
The growth of the open science movement is a reaction to scientists realising what non-scientists already knew; there’s a communications crisis. When scientists conduct their work in secrecy or with self-serving motives, the public lose trust and disengage with the scientific process. ‘Anti-science’ sentiment can be powerful leverage for politicians and we see this today in climate change denial. And scientists can be at fault.
P hacking is the selective manipulation of analyses or data eligibility in the pursuit of finding a statistically significant result (Ref). A primary driver to do this may be positive (significant) results continue to enjoy higher publication rates (Ref, Ref). Thankfully there is a significant push from within science to pre-register all aspects of the study - including the analysis you plan on performing - to make these disingenuous practices harder (Ref). The open science movement will continue to improve the way science is performed, but it can do nothing for the way science is used.
Clinicians can misuse science
Science continues to uncover more and more about how we think and it’s clear we are unfortunately riddled with biases. There are numerous ways that we can misuse science and many of them have roots in these biases. The following list represents a small selection:
To cherry-pick is to appeal to just those scientific results or theories that support your world view (Ref, Ref). Related to this is the practice of giving insufficient weight to scientific results that conflict with your view (Ref). We can see this when people ‘use’ (misuse) science to prove their stance regardless of its quality or data.
Using available ‘high profile’ evidence
The availability bias is a mental shortcut, in which we generalise the specific. More recent and readily available answer and solutions are favoured because they are easy to recall and seemingly more important (Ref). An example of this may be using recent ‘high profile’ evidence to influence decisions in treatment when it does not apply to the patients you work with.
Arguing from scientific authority
The fact that a study has produced certain results, and published in a reputable journal and profiled heavily within the media does not make it true. Science does have a perception of authority, but it maybe us that provide it with that sheen. We may think that any scientific results are better than theorising, but bad or underdeveloped science, or science that has not yet gone through the rigorous scrutiny necessary to become scientific consensus, is not better than no science at all (Ref).
Science is a guide for understanding the conditions we treat, explaining the links between these conditions and treatment choices. It is a framework for thinking but it can’t think for us. We need to be better thinkers first and better users of science second. This begins with having an awareness of what science is and what it isn’t. It isn’t something that replaces good reasoning. Knowing the limits to what you can claim and what you can’t is important as is an understanding of our faulty thinking apparatus. Holding science on a pedestal may only serve to continue the misuse of science and hold our science-based professions back. We need to ‘use’ science better, and stop misusing science.