The study by Sheth et al. examines the citation rates for published studies comparing internal fixation with arthroplasty for the treatment of hip fractures, and the authors report an overall citation rate of only 56% even when a five-year lag in publication is granted. More importantly, Sheth et al. provide the results of a cumulative meta-analysis of twelve studies revealing “a stable estimate of at least a 60% reduction in risk of revision with arthroplasty” present after the fifth study, performed in the year 2000. On the basis of the provided information, at least one, and possibly more, of the trials began enrolling patients after the year 2000. As a result, the authors ultimately question “the continued exposure of patients to interventions that have proved to be less effective,” calling into question the timing of the existence of clinical equipoise in the treatment of hip fractures.
The foundation of all research rests in the clinical question, which serves as the first and most important step in the design of a potential trial. Once a question is posed, the next logical step involves a thorough review of the existing literature to seek an answer. In the past, this required reliance on textbooks and hand searching of articles and associated references, but, more recently, online search engines and access to journal archives allow quicker retrieval. More and more clinicians are becoming aware of the principles of evidence-based medicine and the attributed hierarchy inherent to various study designs and thus are able to scrutinize publications for both internal and external validity prior to incorporating results into clinical practice. As shown previously, as the methodological rigor associated with a particular trial increases, so generally does the impact of the publication1. In addition, as the impact factor of the publishing journal increases, so does the subsequent citation rate2. With this in mind, the results of the current study revealing that 60% of the “highly cited” studies were published in The Journal of Bone and Joint Surgery (American or British Volume) are not surprising.
Clinical equipoise describes the legitimate uncertainty that exists among clinicians regarding various treatment options and persists as long as results comparing one intervention with another are too weak to influence the judgment of the community of clinicians3. In concept, this seems straightforward, but in reality many additional issues are involved. Miller and Joffe call into question the concept of clinical equipoise as a guide for the conduct of trials due to the fallibility of expert opinion as well as a lack of consensus regarding what constitutes the “collective indifference” required among experts for equipoise to exist4. Their argument is best relayed in the following statement: “Assuming that the relevant expert community can be identified, what is the minimal proportion of members who must favor treatment A over treatment B as an appropriate therapy for patients with a given medical condition?”4 Unfortunately, the existence of clinical equipoise is not as straightforward as it seems, and it is possible that, for some experts, it did not exist in the year 2000 for hip fracture treatment despite the aforementioned results of the cumulative meta-analysis.
An excellent example in the orthopaedic literature concerning the controversy surrounding the concept of clinical equipoise is shown by the 2002 study by Moseley et al. regarding a randomized trial of arthroscopic surgery for the treatment of knee osteoarthritis5. In that study, patients at a Veterans Affairs hospital with osteoarthritis were randomly assigned to one of three treatment arms consisting of arthroscopic debridement, arthroscopic lavage, or placebo surgery with incisions only. The results revealed no difference in outcome between arthroscopic lavage or arthroscopic debridement as compared with the placebo group. Although the methods were well described and the results were straightforward, representing the highest level of available evidence on the topic, much controversy arose following publication. Were the results truly generalizable to the population at large? The predominantly male population, the lack of reporting of limb malalignment, and the possibility of alleviation of mechanical symptoms in early stages of osteoarthritis call into question the external validity of the results of the study. Despite the existence of the study by Moseley et al., one could still question at what stage of knee osteoarthritis does arthroscopy fail to serve as a viable treatment option?
This article by Sheth et al. serves as a potent reminder to those involved in medical research. As the article alludes, it is not always feasible for institutional review boards to have members with expertise across multiple subspecialties of medicine. Ultimately, the responsibility rests with the researchers designing a potential study to perform a thorough literature review prior to instigating a new clinical trial. This includes an exhaustive literature search, which would likely benefit from the methodological transparency required of a systematic review. However, the presence or absence of clinical equipoise remains controversial and difficult to determine.