Evidence-based medicine

Evidence-based medicine (EBM) is a medical movement based upon the application of the scientific method to medical practice, recognizing that many long-established medical traditions are not yet subjected to adequate scientific scrutiny. According to the Centre for Evidence-Based Medicine, "Evidence-based medicine is the conscientious, explicit and judicious use of current best evidence in making decisions about the care of individual patients."

Overview
Using techniques from science, engineering and statistics, such as meta-analysis of the existing literature, risk-benefit analysis, and randomized controlled trials, it aims for the ideal that all doctors and other healthcare professionals should make "conscientious, explicit, and judicious use of current best evidence" in their everyday practice.

Evidence-based medicine categorizes different types of clinical evidence and ranks them according to the strength of their freedom from the various biases that beset medical research. For example, the strongest evidence for therapeutic interventions is provided by randomized, double-blind, placebo-controlled trials involving a homogeneous patient population and medical condition. In contrast, patient testimonials, case reports, and even expert opinion have little value as proof because of the placebo effect, the biases inherent in observation and reporting of cases, difficulties in ascertaining who is an expert, and more.

Practising evidence-based medicine implies not only clinical expertise, but expertise in retrieving, interpreting, and applying the results of scientific studies, and in communicating the risks and benefit of different courses of action to patients.

For all its problems, evidence-based medicine has very successfully demoted the ex cathedra statement of the "medical expert" to the least valid form of evidence, and all "experts" are now expected to be able to reference their pronouncements to the relevant literature. One way that physicians facilitate the integration of evidence-based medicine in daily practice is via participation in a journal club.

History
Professor Archie Cochrane was a Scottish epidemiologist whose book Effectiveness and Efficiency: Random Reflections on Health Services (1972) and subsequent advocacy caused increasing acceptance of the concepts behind evidence-based practice. Cochrane's work was honoured through the naming of centres of evidence-based medical research &mdash; Cochrane Centres &mdash; and an international organisation, the Cochrane Collaboration. The term "evidence-based medicine" first appeared in the medical literature in 1992 in a paper by Guyatt et al.

Classification
Two types of evidence-based medicine have been proposed.

Evidence-based individual decision making
Evidence-based individual decision (EBID) making is evidence-based medicine as practiced by the individual health care provider. There is concern that current evidence-based medicine focuses excessively on EBID. The American Academy of Family Physicians (AAFP) has determined that DynaMed (http://www.ebscohost.com/dynamed) may be of assistance to family physicians in answering clinical questions with high-quality evidence.

Process and progress
Using techniques from science, engineering, and statistics, such as meta-analysis of medical literature, risk-benefit analysis, and randomized controlled trials (RCTs), EBM aims for the ideal that healthcare professionals should make "conscientious, explicit, and judicious use of current best evidence" in their everyday practice. Ex cathedra statements by the "medical expert" are considered to be least valid form of evidence. All "experts" are now expected to reference their pronouncements to scientific studies.

The systematic review of published research studies is a major method used for evaluating particular treatments. The Cochrane Collaboration is one of the best-known, respected examples of systematic reviews. Like other collections of systematic reviews, it requires authors to provide a detailed and repeatable plan of their literature search and evaluations of the evidence. Once all the best evidence is assessed, treatment is categoried as "likely to be beneficial", "likely to be harmful", or "evidence did not support either benefit or harm".

Generally, there are three distinct, but interdependent, areas of EBM. The first is to treat individual patients with acute or chronic pathologies by treatments supported in the most scientifically valid medical literature. Thus, medical practitioners would select treatment options for specific cases based on the best research for each patient they treat. The second area is the systematic review of medical literature to evaluate the best studies on specific topics. This process can be very human-centered, as in a journal club, or highly technical, using computer programs and information techniques such as data mining. Increased use of information technology turns large volumes of information into practical guides. Finally, evidence-based medicine can be understood as a medical "movement" in which advocates work to popularize the method and usefulness of the practice in the public, patient communities, educational institutions, and continuing education of practicing professionals.

Ranking the quality of evidence
Evidence-based medicine categorizes different types of clinical evidence and ranks them according to the strength of their freedom from the various biases that beset medical research. For example, the strongest evidence for therapeutic interventions is provided by systematic review of randomized, double-blind, placebo-controlled trials involving a homogeneous patient population and medical condition. In contrast, patient testimonials, case reports, and even expert opinion have little value as proof because of the placebo effect, the biases inherent in observation and reporting of cases, difficulties in ascertaining who is an expert, and more.

US Preventive Services Task Force
Systems to stratify evidence by quality have been developed, such as this one by the U.S. Preventive Services Task Force  for ranking evidence about the effectiveness of treatments or screening:
 * Level I: Evidence obtained from at least one properly designed randomized controlled trial.
 * Level II-1: Evidence obtained from well-designed controlled trials without randomization.
 * Level II-2: Evidence obtained from well-designed cohort or case-control analytic studies, preferably from more than one center or research group.
 * Level II-3: Evidence obtained from multiple time series with or without the intervention. Dramatic results in uncontrolled trials might also be regarded as this type of evidence.
 * Level III: Opinions of respected authorities, based on clinical experience, descriptive studies, or reports of expert committees.

National Health Service
The UK National Health Service uses a similar system with categories labeled A, B, C, and D. The above Levels are only appropriate for treatment or interventions; different types of research are required for assessing diagnostic accuracy or natural history and prognosis, and hence different "levels" are required. For example, the Oxford Centre for Evidence-based Medicine suggests levels of evidence (LOE) according to the study designs and critical appraisal of prevention, diagnosis, prognosis, therapy, and harm studies:


 * Level A: Consistent Randomised Controlled Clinical Trial, cohort study, all or none (see note below), clinical decision rule validated in different populations.
 * Level B: Consistent Retrospective Cohort, Exploratory Cohort, Ecological Study, Outcomes Research, case-control study; or extrapolations from level A studies.
 * Level C: Case-series study or extrapolations from level B studies.
 * Level D: Expert opinion without explicit critical appraisal, or based on physiology, bench research or first principles.

GRADE Working Group
A newer system is by the GRADE Working Group and takes in account more dimensions that just the quality of medical evidence. "Extrapolations" are where data is used in a situation which has potentially clinically important differences than the original study situation. Thus, the quality of evidence to support a clinical decision is a combination of the quality of research data and the clinical 'directness' of the data.

Despite the differences between systems, the purposes are the same: to guide users of clinical research information about which studies are likely to be most valid. However, the individual studies still require careful critical appraisal.

Note: The all or none principle is met when all patients died before the Rx became available, but some now survive on it; or when some patients died before the Rx became available, but none now die on it.

Categories of recommendations
In guidelines and other publications, recommendation for a clinical service is classified by the balance of risk versus benefit of the service and the level of evidence on which this information is based. The U.S. Preventive Services Task Force uses:
 * Level A: Good scientific evidence suggests that the benefits of the clinical service substantially outweighs the potential risks. Clinicians should discuss the service with eligible patients.
 * Level B: At least fair scientific evidence suggests that the benefits of the clinical service outweighs the potential risks. Clinicians should discuss the service with eligible patients.
 * Level C: At least fair scientific evidence suggests that there are benefits provided by the clinical service, but the balance between benefits and risks are too close for making general recommendations. Clinicians need not offer it unless there are individual considerations.
 * Level D: At least fair scientific evidence suggests that the risks of the clinical service outweighs potential benefits. Clinicians should not routinely offer the service to asymptomatic patients.
 * Level I: Scientific evidence is lacking, of poor quality, or conflicting, such that the risk versus benefit balance cannot be assessed. Clinicians should help patients understand the uncertainty surrounding the clinical service.

Statistical measures
Evidence-based medicine attempts to express clinical benefits of tests and treatments using mathematical methods. Tools used by practitioners of evidence-based medicine include:

Likelihood ratio
The pretest odds of a particular diagnosis, multiplied by the likelihood ratio, determines the post-test odds. (Odds can be calculated from, and then converted to, the [more familiar] probability.) This reflects Bayes' theorem. The differences in likelihood ratio between clinical tests can be used to prioritize clinical tests according to their usefulness in a given clinical situation.

AUC-ROC
The area under the receiver operating characteristic curve (AUC-ROC) reflects the relationship between sensitivity and specificity for a given test. High-quality tests will have an AUC-ROC approaching 1, and high-quality publications about clinical tests will provide information about the AUC-ROC. Cutoff values for positive and negative tests can influence specificity and sensitivity, but they do not affect AUC-ROC.

Number needed to treat / harm
Number needed to treat or Number needed to harm are ways of expressing the effectiveness and safety of an intervention in a way that is clinically meaningful. In general, NNT is always computed with respect to two treatments A and B, with A typically a drug and B a placebo (in our example above, A is a 5-year treatment with the hypothetical drug, and B is no treatment). A defined endpoint has to be specified (in our example: the appearance of colon cancer in the 5 year period). If the probabilities pA and pB of this endpoint under treatments A and B, respectively, are known, then the NNT is computed as 1/(pB-pA). The NNT for breast mammography is 285; that is, 285 mammograms need to be performed to diagnose one breast cancer. As another example, an NNT of 4 means if 4 patients are treated, only one would respond.

An NNT of 1 is the most effective and means each patient treated responds, e.g., in comparing antibiotics with placebo in the eradication of Helicobacter pylori. An NNT of 2 or 3 indicates that a treatment is quite effective (with one patient in 2 or 3 responding to the treatment). An NNT of 20 to 40 can still be considered clinically effective.

Quality of clinical trials
Evidence-based medicine attempts to objectively evaluate the quality of clinical research by critically assessing techniques reported by researchers in their publications.


 * Trial design considerations. High-quality studies have clearly-defined eligibility criteria, and have minimal missing data.


 * Generalizability considerations. Studies may only be applicable to narrowly-defined patient populations, and may not be generalizable to clinical practice.


 * Followup. Sufficient time for defined outcomes to occur can influence the study outcomes and the statistical power of a study to detect differences between a treatment and control arm.


 * Power. A mathematical calculation can determine if the number of patients is sufficient to detect a difference between treatment arms. A negative study may reflect a lack of benefit, or simply a lack of sufficient quantities of patients to detect a difference.

Limitations
Although evidence-based medicine is becoming regarded as the "gold standard" for clinical practice there are a number of limitations and criticisms of its use.

Ethics
In some cases, such as in open-heart surgery, conducting randomized, placebo-controlled trials is commonly considered to be unethical, although observational studies may address these problems to some degree.

Cost
The types of trials considered "gold standard" (i.e. large randomized double-blind placebo-controlled trials) are expensive, so that funding sources play a role in what gets investigated. For example, public authorities may tend to fund preventive medicine studies to improve public health, while pharmaceutical companies fund studies intended to demonstrate the efficacy and safety of particular drugs.

Generalizability
Furthermore, evidence-based guidelines do not remove the problem of extrapolation to different populations or longer timeframes. Even if several top-quality studies are available, questions always remain about how far, and to which populations, their results are "generalizable". Furthermore, skepticism about results may always be extended to areas not explicitly covered: for example, a drug may influence a "secondary endpoint" such as test result (blood pressure, glucose, or cholesterol levels) without having the power to show that it decreases overall mortality or morbidity in a population.

The quality of studies performed varies, making it difficult to compare them and generalize about the results.

Certain groups have been historically under-researched (racial minorities and people with many co-morbid diseases), and thus the literature is sparse in areas that do not allow for generalizing.

Publication bias
It is recognised that not all evidence is made accessible, that this can limit the effectiveness of any approach, and that efforts to reduce various publication bias and retrieval bias is required.

Failure to publish negative trials is the most obvious gap, and moves to register all trials at the outset, and then to pursue their results, are underway. Changes in publication methods, particularly related to the Web, should reduce the difficulty of obtaining publication for a paper on a trial that concludes it did not prove anything new, including its starting hypothesis.

Treatment effectiveness reported from clinical studies may be higher than that achieved in later routine clinical practice due to the closer patient monitoring during trials that leads to much higher compliance rates.

The studies that are published in medical journals may not be representative of all the studies that are completed on a given topic (published and unpublished) or may be misleading due to conflicts of interest (i.e. publication bias). Thus the array of evidence available on particular therapies may not be well-represented in the literature. A 2004 statement by the International Committee of Medical Journal Editors (that they will refuse to publish clinical trial results if the trial was not recorded publicly at its outset) may help with this, although this has not yet been implemented.

Populations, clinical experience, and dubious diagnoses
EBM applies to groups of people but this does not preclude clinicians from using their personal experience in deciding how to treat the person in front of them. In The limits of evidence-based medicine, Tonelli advises that "the knowledge gained from clinical research does not directly answer the primary clinical question of what is best for the patient at hand." and suggests that evidence-based medicine should not discount the value of clinical experience.

David Sackett writes that "the practice of evidence based medicine means integrating individual clinical expertise with the best available external clinical evidence from systematic research".

Political criticism
There is a good deal of criticism of evidence based medicine, which is suspected of being - as against what the phrase suggests - in essence a tool not so much for medical science as for health managers, who want to introduce managerialist techniques into medical administration. Thus Dr Michael Fitzpatrick writes: "To some of its critics, in its disparagement of theory and its crude number-crunching, EBM marks a return to 'empiricist quackery' in medical practice . Its main appeal, as Singh and Ernst suggest, is to health economists, policymakers and managers, to whom it appears useful for measuring performance and rationing resources."

In psychiatry
Standard knowledge about mental illnesses, such as the Diagnostic and Statistical Manual of Mental Disorders, have been criticized as incompletely justified by evidence. In many cases, it is unknown whether a particular "disease" has one, several, or no underlying biological causes (controversy arising over whether some diseases are merely an artifact of the attempt to construct a unified classification scheme, rather than a "real" disease).

While some experts point to statistics in support of the idea that a lack of adoption of research findings results in suboptimal treatment for many patients, others emphasize the importance of the skill of the practitioner and the customization of the treatment to fit individual needs. There is some controversy over whether mental illnesses is too complex for broad population studies to be helpful.

History
While some find traces of evidence-based medicine's origin in ancient Greece, others trace its roots to ancient Chinese medicine. Although testing medical interventions for efficacy has existed since the time of Avicenna's The Canon of Medicine in the 11th century, it was only in the 20th century that this effort evolved to impact almost all fields of health care and policy. Professor Archie Cochrane, a Scottish epidemiologist, through his book Effectiveness and Efficiency: Random Reflections on Health Services (1972) and subsequent advocacy, caused increasing acceptance of the concepts behind evidence-based practice. Cochrane's work was honoured through the naming of centres of evidence-based medical research &mdash; Cochrane Centres &mdash; and an international organization, the Cochrane Collaboration. The explicit methodologies used to determine "best evidence" were largely established by the McMaster University research group led by David Sackett and Gordon Guyatt. The term "evidence based" was first used in 1990 by David Eddy. The term "evidence-based medicine" first appeared in the medical literature in 1992 in a paper by Guyatt et al. Relevant journals include the British Medical Journal's Clinical Evidence, theJournal Of Evidence-Based Healthcare and Evidence Based Health Policy. All of these were co-founded by Anna Donald, an Australian pioneer in the discipline.

EBM and ethics of experimental or risky treatments
Insurance companies in the United States and public insurers in other countries usually wait for drug use approval based on evidence-based guidelines before funding a treatment. Where approval for a drug has been given, and subsequent evidence based findings indicating that a drug may be less safe than originally anticipated, some insurers in the U.S. have reacted very cautiously and withdrawn funding. For example, an older generic statin drug had been shown to reduce mortality, but a newer and much more expensive statin drug was found to lower cholesterol more effectively. However, evidence came to light about safety concerns with the new drug which caused some insurers to stop funding it even though marketing approval was not withdrawn. Some people are willing to take their chances to gamble their health on the success of new drugs or old drugs in new situations which may not yet have been fully tested in clinical trials. However insurance companies are reluctant to take on the job of funding such treatments, preferring instead to take the safer route of awaiting the results of clininal testing and leaving the funding of such trials to the manufacturer seeking a licence. .

Sometimes caution errs in the other direction. Kaiser Permanente did not change its methods of evaluating whether or not new therapies were too "experimental" to be covered until it was successfully sued twice: once for delaying IVF treatments for two years after the courts determined that scientific evidence of efficacy and safety had reached the "reasonable" stage; and in another case where Kaiser refused to pay for liver transplantation in infants when it had already been shown to be effective in adults, on the basis that use in infants was still "experimental." Here again, the problem of induction plays a key role in arguments.

Application of the evidence based model on other public policy matters
There has been discussion of applying what has been learned from EBM to public policy. In his 1996 inaugural speech as President of the Royal Statistical Society, Adrian Smith held out evidence-based medicine as an exemplar for all public policy. He proposed that "evidence-based policy" should be established for education, prisons and policing policy and all areas of government.

Note

 * Guyatt G, Cairns J, Churchill D, et al. [&#8216;Evidence-Based Medicine Working Group&#8217;] Evidence-based medicine. A new approach to teaching the practice of medicine. JAMA 1992;268:2420-5. PMID 1404801.