The observable and the unobservable of Evidence-Based Medicine

In 1997 Sackett et al. defined evidence-based medicine as follows:

Evidence based medicine is the conscientious, explicit, and judicious use of current best evidence in making decisions about the care of individual patients. The practice of evidence based medicine means integrating individual clinical expertise with the best available external clinical evidence from systematic research. By individual clinical expertise we mean the proficiency and judgment that individual clinicians acquire through clinical experience and clinical practice. Increased expertise is reflected in many ways, but especially in more effective and efficient diagnosis and in the more thoughtful identification and compassionate use of individual patients' predicaments, rights, and preferences in making clinical decisions about their care. By best available external clinical evidence we mean clinically relevant research, often from the basic sciences of medicine, but especially from patient centred clinical research into the accuracy and precision of diagnostic tests (including the clinical examination), the power of prognostic markers, and the efficacy and safety of therapeutic, rehabilitative, and preventive regimens. External clinical evidence both invalidates previously accepted diagnostic tests and treatments and replaces them with new ones that are more powerful, more accurate, more efficacious, and safer.[My emphasis]

Good doctors use both individual clinical expertise and the best available external evidence, and neither alone is enough. Without clinical expertise, practice risks becoming tyrannised by evidence, for even excellent external evidence may be inapplicable to or inappropriate for an individual patient. Without current best evidence, practice risks becoming rapidly out of date, to the detriment of patients.

Evidence based medicine has brought about lots of support and criticism over the years. As it happens with other concepts it may serve well, but it also has potential for being misused, misunderstood, and misapplied. Guyatt and Djulbegovic in their recent article Epistemologic Inquiries in Evidence-Based Medicine try to clarify what the practice is, address a few criticisms, and direct where the framework is still incomplete. I would like to address a section in their article that deals with how they approach what's observable and not directly observable in EBM. They write:

EBM makes a normative claim about when some kinds of medical knowledge can genuinely be taken as knowledge. It also makes a normative claim about medical practice: Wherever possible, the choice of diagnostic test, preventive measure, or treatment should be based on the best available evidence about the available interventions. EBM thus makes medical knowledge central to ethical medical practice in a rather strong way and also uses a stringent criterion for what should count as medical knowledge.

The authors write about the distinction between knowledge, evidence, belief, truth, and facts in EBM. They also comment on the roles of skepticism and uncertainty in EBM practice:

Although not universally agreed upon,¹³ there is a crucial difference between evidence and knowledge. The evidentiary basis for belief may not be necessarily true, while “true” belief is a prerequisite for the definition of knowledge. However, as explained later, what counts as “true” scientific knowledge is tentative and remains revisable as science progresses. EBM accepts a notion of gradation of evidence,¹⁷ implying that particular claims can be confirmed by given evidence to various degrees and that evidence can be misleading.¹³ The emphasis of EBM on skepticism and uncertainty — we will never be sure of the magnitude of the effects of our treatments or the power of our diagnostic tests — is central to the approach and agrees with the philosophical view that scientific knowledge is never complete and ultimately fallible.

On the observable, the observed, theories, and unobservable reality they add:

Both EBM and logical-positivism suggest that the role of theories is not to accurately describe the world but to accurately predict empirical observations.^27,30 Theories do not need to be true to serve as useful links between observed reality and observable (but not observed) as well as unobservable reality. [...] Like positivism, EBM suggests we should restrict ourselves to the observed reality, and when we go beyond our observations, the focus is on extending our inferences about the unobserved world. Reality, however, remains ultimately unknowable.^28,29 [My emphasis] This positivist approach has some affinities with EBM, which for instance tends to privilege knowledge about what can be shown to work over deeper questions about why it does so. However, the logical-positivist stance is too restrictive. Logical-positivists consider any statements that are not true by definition (analytic) or that correspond to empirical findings (synthetic statements) as nonsense.³² In other words, anything that is in principle observable but not observable at a given time would be classified by logical-positivists as nonsense.³² EBM, on the other hand, often proceeds as in the example of the collation of studies in meta-analysis to allow convergence on the underlying truth that is only indirectly observable. [...] Hence, EBM need not be taken as positivist science, and advocates of EBM are not committed to positivist theses about meaning and truth in science and medicine or about the kind of facts that can be known or not known.

I think at this point it's important to touch upon the distinction between scientific realism and anti-realism which I think relates to the concepts Guyatt and Djulbegovic are writing about. In this video interview Bas van Fraassen, a philosopher of science who advocates for a constructive empiricism approach which falls under the anti-realist umbrella, explains the difference. The empiricists approach, he explains, is to limit the explanation of science's successful endeavor to what's directly observable as opposed to scientific realism which goes beyond the observable. According to him scientists accept theories and models because they are empirically adequate to what's directly observable, not because they are true and describe an unobservable reality.

The contrast between scientific realism and empiricism may have implications of everyday medical practice. What's really observed in everyday practice are data collected from taking a history, doing a physical exam, results from laboratory and imaging studies. These results are put together and matched to possible disease entities and events which are not directly observable. Scientific realism explains that although the myocardial infarction is not directly observed, it is believed to be a real identity and can be deduced from the collected data. As stated before by Guyatt and Djulbegovic, EBM emphasizes uncertainty, skepticism, fallibility, and incompleteness of knowledge. These concepts need further exploration as they may help explain the origins of being wrong.

As per the authors the practice of EBM should also be based on the totality of the evidence. But if the practice of medicine is to be based on the totality of the evidence, EBM as normative practice should take into account factors that affect research and clinical practices. These factors include, but are not limited to, publication bias, misapplication of statistical procedures, questionable research practices, etc. Being a scientific realist without knowing the limitations of the evidence does not make sense, but a structure based only on what's observable doesn't seem sufficient either. Scientific explanations are difficult, but they should aim to take what's important into account.

Tweet to @jvrbntz