Evidence Based Medicine


Modern medicine was born in XIX c and developed during the first half of XX c on the basis of personal experience of many brave scientists who dared to research new areas of chemistry,  physiology, physics. Their discussions shaped modern medicine and gave it the power to fight diseases, create new drugs, discover different aspects of human body, and develop surgery. However the number of new produced and patented molecules rapidly grew after the World War II, and it was impossible for any physician to personally try and know them all. In order to help patients and doctors in making reasonable decisions during the treatment new, statistic ways of creating evidence was proposed in the 1990s. This statistical approach was called Evidence Based Medicine and very fast became a standard for researching and presenting truth in medicine. Following article presents the definition of Evidence Based Medicine, and discusses main problems related with this idea.

Evidence Based Medicine is, according to the definition presented in Sackett’s article of 1996, the meticulous, precise and prudent use – in clinical procedures- of best possible scientific evidences concerning the efficiency, effectiveness and safety of a given therapy. (source)1  

Giving hope for obtaining objective data, EMB was ethusiastically adopted in the environments managing healthcare system.  For the less informed, it has been the disposer of absolute truth and realization of dream on the ideal medicine, free from influences of personal convictions. However, this image is very far from the truth and EBM itself faces numerous methodological problems.  It is easier to understand what is and what is not the Evidence Based Medicine when you have reviewed the history of medicine of the preceding 150 years.

Until the second half of the twentieth century medical practitioners decided on therapy on the basis of their education and experience they had acquired during their practice. Mostly, they shared the information with their professional colleagues and this fact is expressed today in the Code of Medical Ethics. The most renown specialist in a given discipline discusses certain issues between themselves, often in public during the meetings of Scientific Associations. This system allowed for the development of medicine as science, and inventions/discoveries of single physicians now and then changed the course of history.  It is enough to mention the works of Louis Pasteur and Robert Koch on bacteria, experiments of Sigmund Freud concerning the treatment of neuroses or Horace Wells using the “laughing gas” for analgesia in tooth extraction. 

The development of medicine of the twentieth century was inseparably connected with Chemistry and Pharmacotherapy. The breakthrough in their application took place in the World War II period, which was interconnected -on one hand- with intense researches and- on the other hand- with market reorganization. After the war pharmaceutical companies patented more and more particles and no medical practitioner was able to assess their efficiency individually as there were too many of them.  What is more, after tragic experience with Thalidomide public opinion started demanding the safety control of chemical medications.

While to a limited extent the statistics had been used in medicine since half of the eighteenth century (see the renowned trials by James Lind) but its was only two hundred years later when it became inevitable in the face of the growing number of produced and patented particles which were supposed to become medicinal drugs. In 1947, British Medical Journal published Austin Bradford Hill’s call to physicians to start using statistics in their research(source)2. But it is the name of Archie Cochrane that shall be associated with the general approval of statistics as a tool for describing reality in clinical trials. Since the late seventies the systematic reviews of available literature – initiated by Lind- have become, at least in theory, a “golden” standard in undertaking therapeutic decisions.  

In the view of its supporters, Evidence Based Medicine is the ideal tool which facilitates decision-making process n the basis of best possible plan of trials. In reality it is much different. This is why the idea of EBM, and -in particular- the manner of its implementation, is highly debatable.

In the following part I shall discuss most commonly reported remarks and problems concerning EBM:

First of all, EBM does not fulfil its own methodological presumptions- it has not been proved that the care over patients is better if the decisions are based on statistical information(source)3 The idea of IBM was generally introduced with the evasion of researches and evidences it stipulates.

PSecondly, EBM is based on statistics concerning large groups, so its application in treatment of an individual patient is limited(source)4. Statistics are used for description of trends occurring in populations. In order to describe these trends it uses simplifications, assuming similarities between individuals. Meanwhile, patients differ from one another. These differences are essential in clinical works and sometimes they decide on the effectiveness of therapy. What is more, in case of individual patient the trends or average means are not important. It is important whether a given treatment works in this specific case or not.  Statistical research requires large groups of patients involved. EBM precludes the research on rare diseases or description of rare events and cases. It causes marginalization of the sick for whom treatment has no “good quality scientific evidence”.

Another problem of EBM, connected with the aforementioned individuality of a patient is limiting his autonomy in undertaking best possible decisions for himself/herself. Whilst in the already quoted article Sackett and co-authors warned against using EBM for creating “recipe books” in medicine and influencing the decision-making process of medical practitioners and patients, though this is what EBM has mostly been used for.  Additionally, regardless of the intentions of its creators, EBM is particularly susceptible to the influence of the coalition of pharmaceutical industry and establishment managing the healthcare system.  Using the authority of medical trials, they are able to favour certain techniques of treatment and therapies.  In the time of growing crisis in healthcare it is not difficult to imagine the temptation to manipulate information passed on to medical doctors and patients in such a way that they- for instance- choose cheaper solutions, but not necessarily having better efficiency.

While mentioning the pharmaceutical industry we cannot forget about the aspect of financing clinical trials. EBM experiences the essential malfunction in the very beginning of the process of collecting data for analysis. Carrying out clinical trials is very expensive. The costs grow proportionally to the quality of research which means the number of patients included in trials, means involved, external quality assurance systems and so on. In reality, clinical trials are financed, to a large extent, by pharmaceutical companies and, as a result, are connected with medicinal drugs produced by them and cause the situation that the industry takes the control over science, having impact on every stage of the trial: research plan, carrying out research and, what is even more important, the industry become the sole disposer of trial results deciding on their publication or encryption. This situation leads to insufficient financing of pioneering trials or these researches referring to therapies not subject to patenting procedures.  Moreover, in making decisions both patients and doctors are deprived of access to 1/3 researches as their results have never been published(source)5.   

The methodology offered by EBM is also problematic as, what was mentioned before, it has originated from statistics and epidemiology. Like these disciplines of knowledge, it evaluates the sources of information, favouring data obtained from controlled clinical trials, additionally “good quality ones” which means these least prone to interferences. This situation results in underestimation of information obtained from other sources, such as quality trials, observations, clinical experiments, ethnographic research. It causes several problems.  

Clearly the knowledge generated by EBM ignores non-statistical sources of medical information.  Upshur lists four equivalent types of data in medicine: qualitative-personal, quantitative-personal, qualitative -general and quantitative-general(source)6. EBM is focused merely on the last type of information, mitigating the importance of the former three types.

By no manner of means it has been proved that information obtained in controlled clinical trials and meta-analyses is better or more essential than information acquired from the remaining types of research(source)7,(source)8,(source)9. What is more, it seems that the similarly-planned controlled trials, enquiring the same issue, often provide divergent findings(source)10.

Also, EBM “suffers from” problems connected with its philosophical grounds. Namely it is completely based on empiricism, it presumes the observation free from observer-related interferences is possible. Meanwhile, since the second half of the twentieth century it is widely known that you cannot carry out objective observation, not susceptible to theories and independent from an observer(source)11. This is why it is so important for the science to take into account the beliefs and the world vision of the observer, which form the research questions as well as the interpretation of results.  As EBM does not take these into account, some scholars call it not only un-scientific but even anti-scientific(źródło)12,(źródło)13.

As we can see the ides of Evidence Based Medicine,, implemented without research and discussions, is not free from numerous, often cardinal methodological and ethical problems. It turned out that in practice it marginalizes the individual experiences, is used for limiting the autonomy of patients and physicians in decision-making process. It also gave the pharmaceutical industry a tremendous influence on the direction and process of the development of medicine. Even more interesting is the fact that by means of this methodology and standards set by EBM, courageous scholars worldwide, in hundreds of clinical trials and dozens of meta-analyses, confirm the effectiveness, safety and mechanisms of activity of acupuncture.

This article raises issues concerning:

  • EBM objectivity.

  • Quality methodological assumptions.

  • Contradictions in EBM.

It is not evidence based

  • It was introduced with the disregard to methodology it has been stipulating

it is based on statistics referring to large groups and generalizations

  • Its application for treatment of a particular patient is very limited.
  • It does not take into account factors differentiating patients (intake of medications, life style, diet, etc.).
  • It marginalizes rare diseases.

It limits patient’s and a physician’s autonomy in decision making process

  • it results in creating “recipes”- simplified procedure schemes.

It favours information from multi-centre controlled clinical trials, although they provide discrepant results, and it was not proved that these are better than results from other sources:

  • It subordinated the findings from the source of financing
  • It makes the sponsor the disposer of the outcomes- 1/3 trials is never published.
  • it neglects other sources of information

it falsely assumes that the objective observation – free from observer’s convictions – exists.

1 David L Sackett, William M C Rosenberg, J A Muir Gray, R Brian Haynes, W Scott Richardson. “Evidence based medicine: what it is and what it isn’t.” BMJ 1996;312:71. 312, s. 71, 1996

2 A. Bradford Hill. Statistics in the medical curriculum? BMJ. 1947; ii: 366.

3 Haynes RB. “What kind of evidence is it that evidence-Based Medicine advocates want health care providers and consumers to pay attention to?” BMC Health Serv Res 2002;2(1):3.

4 Cohen AM , Stavri PZ , Hersh WR “A categorization and analysis of the criticisms of Evidence-Based Medicine.” Int J Med Inform. 2004 Feb;73(1):35-43

5 Jones CW, Handler L, Crowell KE, Keil LG, Weawer MA, Platts-Mills TF “Non-publication of large randomized clinical trials: cross sectional analysis” BMJ 2013;347:f6104

6 R.E. Upshur, E.G. VanDenKerkhof, V. Goel, “Meaning and measurement: an inclusive model of evidence in health care”, J. Eval. Clin. Pract. 7 (2) (2001) 91—96.

7 Haynes RB. “What kind of evidence is it that evidence-Based Medicine advocates want health care providers and consumers to pay attention to?” BMC Health Serv Res 2002;2(1):3.

8 Sox H. “Screening mammography for younger women: back to basics”. Ann Intern Med 2002;137(5 Part 1):361–2.

9 Benson K, Hartz AJ.”A comparison of observational studies and randomized, controlled trials”. Am J Ophthalmol 2000;130(5):688.

10 Cohen AM , Stavri PZ , Hersh WR “A categorization and analysis of the criticisms of Evidence-Based Medicine.” Int J Med Inform. 2004 Feb;73(1):35-43

11 E. Harari, “Whose evidence? Lessons from the philosophy of science and the epistemology of medicine”, Aust. N. Z. J. Psychiatry 35 (6) (2001) 724—730.

12 B.G. Charlton, A. Miles “The rise and fall of EBM” QJM 91 (5) (1998) 371—374.

13 A. Miles, P. Bentley, A. Polychronis, J. Grey, C. Melchiorri “Recent developments in the evidence-based healthcare debate” J. Eval. Clin. Pract. 7 (2) (2001) 85—89.