How Can Physicians Stay Current on Prescription Drugs?

Robert M. Centor, MD; Pennie Marchetti, MD; R.W. Donnell, MD; Roy M. Poses, MD

Disclosures

August 30, 2006

Roy M. Poses, MD: Embrace Evidence-Based Medicine

The idea that physicians are faced with information overload has become a cliche. There is no magic solution to the problem.

However, evidence-based medicine can help us cope. Briefly, evidence-based medicine has been defined as "the integration of best research evidence with clinical expertise and patient values."[29] Evidence-based medicine is best viewed as a process with several steps: identifying a clinical question; systematically searching for clinical studies that address this question; critically reviewing these studies to eliminate those that are poor quality or irrelevant, and interpreting the rest as best as possible; and then basing the answer to the question on those superior studies. Evidence-based medicine is well-suited to address which treatment, including pharmaceutical treatments, might be best for a particular patient. Note that "best" here means the treatment most likely to maximize benefits and minimize harms to the patient, allowing for how the patient values particular benefits and harms.

To learn more about evidence-based medicine, the short book by Sackett and colleagues[29] and one by Guyatt and Rennie[30] are very helpful. For quick, practical, and understandable discussions about how critical reviews of clinical research are done, see the "Users' Guides" series originally published in JAMA.[30] In particular, the 2 "Users' Guides" about treatment give clear, practical directions for reviewing clinical trials of pharmaceuticals.[31,32]

Of course, it is not practical for individual physicians to go through the whole evidence-based medicine process themselves to keep up to date on pharmaceutical treatments. However, they can use sources, textbooks, review articles, continuing medical education courses, etc., that are based on the evidence-based medicine process. Some of my personal favorites include the Cochrane Collaboration,[33] Clinical Evidence by the BMJ,[34] and the US Preventive Services Task Force.[35]

Unfortunately, far too much that is written and said about pharmaceuticals is "interest based" rather than "evidence based"; that is, it is based on the usually financial, sometimes ideological interests of the writers and speakers. There are many ways that the design, conduct, and analysis of clinical research can be manipulated to increase the likelihood that the results will fit the interests of the research's sponsor. See Smith[36] and Brophy[37] for catalogs of methods used to perform such manipulations. A really rigorous evidence-based medicine approach may help detect how the clinical literature may be manipulated to advance such interests. For example, a perspicacious review by Wright in 2002 warned of the adverse effects of Cox-2 inhibitors long before these effects were widely discussed.[38]

Knowing that research was done at a well-known academic institution, and that an article reporting it was written by well-known academics, is no guarantee against manipulation. Due to financial pressures and other motivations, some academic institutions have been willing to hand over control of clinical research to the corporate sponsors of the work, such as pharmaceutical companies. Mello and colleagues[39] documented the sorts of contractual provisions that academic institutions may accept, such as giving the sponsor rather than the investigators ownership of the data; permitting the sponsor, but not the investigators, to alter the study design; and giving the sponsor control over statistical analyses. In addition, not all journals require authors to reveal conflicts of interest that could affect how they report or discuss clinical research, and some authors have not always readily admitted conflicts of interest even when asked.[40,41]

Thus, one should be particularly skeptical of reviews, opinion articles, and CME presentations that are not rigorously evidence-based. Failing to look for all relevant research, or failing to understand the shortcomings of individual studies, may lead to unintentionally biased results. Many CME presentations are done by "key opinion leaders," who were financed by pharmaceutical companies and, hence, may tend to favor their products.[42] Worse, some reviews and opinion articles may be ghost-written by financially interested organizations, particularly medical education companies running marketing campaigns for pharmaceutical companies.[43,44]

Finally, even the most rigorous evidence-based medicine methods do not work well when research is suppressed by those who are displeased with its results. In that case, even a systematic search would not produce a good reflection of the clinical truth. How often this occurs is unknown, but there have been some notorious cases.[45] There was even one attempt to suppress the report of a new disease that may have resulted from an occupational exposure.[46] Having pharmaceutical and device companies register all clinical trials before they can go forward may lessen the likelihood that trial results can be hidden, but mandatory registration is not yet common practice, much less the law.[47] Physicians should be vigilant for attempts to suppress clinical research and vigorously protest those they find.

In summary, there ought to be some straightforward ways to stay up to date on advances in pharmaceutical treatments. Right now, however, the world of pharmaceutical research and promotion is something of a jungle. Physicians who want to take the best care of their patients need training, the right mental equipment, and trusty guides equipped with evidence-based machetes to hack through it.

Comments

3090D553-9492-4563-8681-AD288FA52ACE
Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.

processing....