Evidence Based Medicine and Pain Management MLWhitworth, MD
Evidence based medicine (EBM) is a process by which medical decision making is applied in order to maximize the results or to decide if a particular drug/procedure/device has any merit. The process involves asking a question, searching the literature for relevant studies, evaluating these studies as providing different levels of evidence, then reaching a conclusion, and acting on that conclusion. EBM incorporates a hierarchy of evidence that is weighted towards more scientific studies. For instance, a study using patient outcomes which were determined by the person conducting the experiment is given less weight than a study in which the outcomes are determined by an independent evaluator (blinded study). Placebo controlled trials are given more weight than those without a placebo control. Multisite investigations using strict protocols are given more weight than a single investigator's results. Systematic reviews are given more weight than individual studies. There are several important factors in conducting clinical trials in order to eliminate the chance that the result is due to a random error or a recurring flaw in the design of the study. Reduction in such errors gives us higher assurances the study is valid and when repeated under the same circumstances, would achieve the same results. Some of these important factors include: 1.Prospective instead of retrospective (this means the study was designed and carried out with a goal in mind and care was being taken to follow the protocol. Retrospective studies are on the other hand an afterthought....a way to evaluate whatever data is available after the fact.) 2. Randomized vs. non-randomized. Randomized means the patient is to receive treatment A or treatment B and the choice of the treatment is generated randomly. Non-randomized trials shift patients into one limb of the study or another and may result in study bias that can invalidate the study. 3. Placebo controlled study. The best studies are performed using a null control group: that is one in which the control group is given a treatment that is not supposed to have any effect, one way or another. 4. Comparative group study. Sometimes placebo controlled studies are impossible to conduct therefore, another treatment is selected which is thought to be equivalent, and part of the patient's receive the comparative treatment. This approach only works if the study design is randomized. 5. Double Blinded vs. single blinded vs. non-blinded study. The double blind study is that where both the observer and the participant do not know what the treatment received actually is (placebo vs. study drug, etc) where a single blinded study is one in which the patient does not know which treatment is being received but the observer may. A non-blinded study is one in which both observer and patient know what is being administered. Double blind studies eliminate both the bias of the observer and the participant in the study, and are thought to the the most accurate of the three blinded states of a study. While the best single study is a placebo controlled double blind randomized prospective study, the use of multiple sites simultaneously conducting the study theoretically will reduce bias and expand the consistency of the outcomes produced, although a multi-site study is very difficult and expensive to carry out. A series of cases is not a study. Nor is a doctor's own experience with a particular result. Both of these have very low reliability ratings. Often a meta-analysis of numerous studies is performed. The meta-analysis takes several studies which may be only related by one factor, and tries to mathematically attempt to derive meaningful conclusions about a particular question. It should be noted FDA approval does not mean the appropriate EBM studies have been performed. For instance, Coblation nucleoplasty was launched without any clinical studies at all. So was pulsed radiofrequency and several other techniques in pain management. These devices received FDA approval without any evidence of effectiveness. Drugs are far more stringently regulated than are devices by the FDA. Procedures are not regulated at all. Where can studies be found? The best studies are published in the major printed medical journals that meet the criteria for medline indexing. There are many journals not medline indexed, but their publication quality is not as rigidly controlled as are the major journals. Medline abstracts (remember the abstracts may not accurately reflect the answer to the EBM question which was asked), may be viewed at pubmed.com. Systematic reviews of pain management procedures and surgical procedures may be found in the major medical journals and also in the Cochrane Database. However, one must remember each review may draw incorrect conclusions due to a lack of understanding of the principles of EBM. EBM does not require placebo controlled randomized double blinded trials as evidence of efficacy, but some organizations (especially insurance companies) impute that requirement. EBM only requires the literature is evaluated, and the best available data is used to determine efficacy. Organizations such as Cochrane and insurance company's medical policies incorrectly conclude a technique is "experimental" or "doesn't demonstrate effectiveness" if there do not already exist multiple clinical trials which are placebo controlled randomized prospective in nature. ASIPP is a pain management organization which has performed systematic reviews of the available data to support pain techniques. The National Guidelines Clearinghouse is another source to determine guidelines on clinical practices. Some of the greatest controversies in pain management revolve around procedures or devices that are relatively new and do not have any evidence at all or extremely limited evidence of efficacy when introduced into the market. Pulsed RF was introduced in approximately 1999 and has been used hundreds of thousands of times with almost no scientific evidence of efficacy. Similarly, coblation nucleoplasty was introduced as a new technique in approximately 2001 without one clinical study published and even to this day has only a hand full of non-major journal publications demonstrating efficacy and safety. Clearly the FDA is asleep at the switch. The FDA is given the task of evaluating new technologies and requiring clinical trials (not merely bench studies or internally conducted studies) prior to approval. Pain management is increasingly see new and very expensive techniques being introduced without any demonstration of patient safety or that the devices actually work at all. Whereas physicians use EBM to determine safety and efficacy, the FDA permitting companies to manufacture devices to be used in pain management and introduce them into the marketplace without sufficient safeguards is reprehensible. Pain management, without a residency program anywhere in the USA, is incapable of generating the type of research necessity to legitimize many of the pain management techniques being used across america. Therefore, it becomes increasingly important for patients to seek out their own resources on efficacy of pain management procedures being scheduled by pain management physicians.
