| Verified against primary record | |
| Canonical definition | The conscientious, explicit and judicious use of current best evidence in making decisions about the care of individual patients[1] |
|---|---|
| Stated negative boundaries | Not cookbook medicine, not a cost-cutting tool, and not restricted to randomised trials[1] |
| Lowest certainty tier | Case series and case reports sit at the bottom of the standard certainty hierarchy[2] |
| Independently reported | |
| Information environment in medical travel | Decisions rested more on former patients’ testimonials and word of mouth than on the patient’s own physician[3] |
| Bands apply only to the rows beneath them. The widely reproduced three-circle formulation of best evidence, clinical expertise and patient values is a later restatement; the 1996 editorial’s own definition is two-part and treats patient preferences separately. | |
Evidence-based medicine is the practice of using the best available research evidence, together with clinical expertise and the patient’s own circumstances and preferences, to decide about an individual patient’s care. The reason it matters to this subject is structural: the information available to someone choosing treatment abroad sits at the bottom of the evidence hierarchy almost by construction.
What the original definition says
The founding editorial defines the practice as the conscientious, explicit and judicious use of current best evidence in making decisions about the care of individual patients, and describes it as integrating individual clinical expertise with the best available external clinical evidence, with patients’ predicaments, rights and preferences as a further element that clinical decisions must account for. It also states what the practice is not: not cookbook medicine, not a cost-cutting device, and not restricted to randomised trials, since the best study design depends on the question asked.[1] The tidy three-circle diagram of best evidence, clinical expertise and patient values that now stands in for the definition is a later crystallisation, and the 1996 text should not be cited as having stated it.
The indexing vocabulary gives a compatible but differently weighted formulation, describing an approach to practising medicine with the goal of improving and evaluating patient care, integrating the best available research evidence with patient values to guide medical decisions.[4]
Grading certainty
Modern practice does not treat evidence as present or absent but as carrying a level of certainty. The internationally used grading approach assesses the balance of benefits and harms, the quality of the evidence, applicability, certainty of baseline risk and resource cost, and was developed because pre-existing grading systems differed enough to limit their usefulness.[5] It produces four certainty levels, high, moderate, low and very low, and each licenses different language: definitive phrasing at high certainty, probably or likely at moderate, may or the evidence suggests at low, and an explicit statement that the evidence is very uncertain at very low.[6]
The mapping from study design to certainty is what makes this relevant here. Randomised trials start at high certainty, observational studies at low, and case series and case reports at very low.[2] Testimonial and single-case material is therefore not weak evidence in a loose sense; it occupies the defined bottom tier of the hierarchy that clinical guidance is built on.
The internal criticism
The movement has been criticised from inside. One widely cited paper argues that the practice has produced real benefits but also negative unintended consequences, and that renewal requires usable evidence combined with context and professional expertise.[7] A sharper critique argues that influential randomised trials are largely run by and for industry, that meta-analyses and guidelines have become a production line serving vested interests, that research funding flows to work with little relevance to health outcomes, and that clinical medicine has been reshaped by financial considerations, while concluding that evidence-based medicine remains an unmet but worthy goal.[8] Both are reported here at the level of their published abstracts and summaries.
Why this bears on treatment abroad
Qualitative research with 32 Canadians who paid out of pocket for surgery abroad found that their decisions rested more on former patients’ testimonials and word of mouth than on their family physicians, that few assessed the reliability of their sources, and that neutral third-party information was conspicuously absent, with facilitators, destination hospitals and industry websites filling the gap; the authors question whether medical travellers can give genuinely informed consent when positive testimonials skew the perception of risk.[3] That is a small retrospective qualitative sample and nothing should be quantified from it. The structural observation it supports is that the dominant information source in this market is the category that formal evidence grading places at the lowest certainty tier, and that the parties supplying it are the parties being paid.
See also
- Clinical Practice Guidelines, the instruments through which graded evidence reaches practice
- Systematic Reviews of Medical Tourism, what the higher tiers of the hierarchy contain on this subject
- Case Studies in Medical Tourism Research, the form most of the published material takes
- Experimental Treatment, where the absence of graded evidence is the defining feature
References
- Sackett DL, Rosenberg WMC, Gray JAM, Haynes RB, Richardson WS. Evidence based medicine: what it is and what it isn’t. BMJ 1996;312:71-72, 13 January 1996; copy hosted by the Centre for Evidence-Based Medicine, University of Oxford. Verified against primary record: the editorial was opened and its definition and its stated negative boundaries read. Retrieved 10 October 2026.
- Cochrane. Levels of quality of a body of evidence in the GRADE approach, Cochrane Handbook for Systematic Reviews of Interventions, archived version 5.0.2, table 12.2.a. Verified against primary record: the table was opened and the design-to-level mapping read. Cited as an archived version. Retrieved 10 October 2026.
- Johnston R, Crooks VA, Snyder J. A qualitative study of the decision-making processes of Canadian medical tourists. Globalization and Health 2012;8:23, 7 July 2012. Verified against primary record: the article was opened and the findings on information sources and the authors’ conclusion read. Retrieved 10 October 2026.
- United States National Library of Medicine. Medical Subject Headings descriptor Evidence-Based Medicine (D019317). Introduced 1997. Verified against primary record: the descriptor record was opened and its scope note and tree numbers read. Retrieved 10 October 2026.
- GRADE Working Group. Grading quality of evidence and strength of recommendations. BMJ 2004;328:1490, 17 June 2004. Independently reported: only the abstract and metadata were retrievable at this address, so the certainty-level names are taken from the two sources below rather than from this paper. Retrieved 10 October 2026.
- United States Centers for Disease Control and Prevention. Communicating findings from the GRADE certainty assessment, ACIP GRADE Handbook chapter 11. Page dated 22 April 2024; last published 16 September 2024. Verified against primary record: the chapter was opened and the four certainty levels and the language each licenses read. Retrieved 10 October 2026.
- Greenhalgh T, Howick J, Maskrey N. Evidence based medicine: a movement in crisis? BMJ 2014;348:g3725. Independently reported: the repository record and its abstract were opened; the published article was not read in full. Retrieved 10 October 2026.
- Ioannidis JPA. Evidence-based medicine has been hijacked: a report to David Sackett. Journal of Clinical Epidemiology 2016;73:82-86. Abstract read at a works index hosting the abstract. Independently reported: only the abstract was accessible, and only abstract-level claims are attributed here. Retrieved 10 October 2026.
Sourcing note: the four GRADE certainty levels are cited to two sources that state them, rather than to the 2004 paper that is their canonical reference, because only the abstract of that paper was retrievable. The 1996 editorial is cited from a copy hosted by an academic centre rather than from the journal. The two critical papers are reported at abstract and summary level only, which is recorded in their references.