To the Editor:

I appreciate the kind and constructive comments of Dr. Wright regarding my 2004 AAOS Presidential Address. They are especially meaningful since they originated from an institutin that is a leader in clinical research and evidence-based medicine in our specialty.

The two suggestions in his letter have been partially addressed by the American Academy of Orthopaedic Surgeons. Starting with the February 2005 annual meeting in Washington, D.C., all submitted abstracts must be accompanied by a grade of their level of evidence. By 2006, all abstracts will be weighted according to their level of evidence so that those studies with better scientific methodology will have an improved chance of acceptance on the program.

The AAOS has also started grading levels of evidence in some of its publications. The two recently released IMCA (Improving Musculoskeletal Care in America) documents on osteoarthritis of the knee and hip provide levels of evidence and grades of recommendation for most of their diagnostic and therapeutic conclusions. Similarly, the Evidence-Based Practice Committee of the AAOS has issued multiple clinical practice guidelines on specific disease entities and position statements with comparable grading. Because of the number of authors involved and the lack of sufficient training in levels of evidence, grading in all AAOS educational review publications, including JAAOS, is not feasible at this time.

At the June AAOS Board of Directors meeting, a strategic discussion on how to incorporate evidence-based medicine into our educational offerings and into the orthopaedic practices of our fellows was conducted. Michael Goldberg, chair of the AAOS Evidence-Based Practice Committee, presented a compelling argument to move ahead aggressively in this area.

It is clear that in the near future, the public, the government, and hopefully our fellowship will demand levels of evidence in most, if not all, of our educational materials. The Journal of Bone and Joint Surgery has been an orthopaedic leader in this arena, and its example should be followed.

To the Editor:

In his AAOS Presidential Address (1), Dr. Bucholz, as they say, has hit the nail on the head. Knowledge is the business of the American Academy of Orthopaedic Surgery. As he so eloquently states, the academy is “the authority for up-to-date, accurate knowledge of musculoskeletal diseases and patient care”. With that thought in mind, I would propose that the American Academy begin to integrate Levels of Evidence into their activities. Levels of Evidence are a way of grading the quality of the studies based on their methodology whereby Level 1, randomized clinical trials, is the highest level and Level 5, expert opinion, is the lowest level of evidence.

Levels of Evidence were introduced to the Journal of Bone and Joint Surgery in January 2003 (2) Since that time, every clinical article published in JBJS is accompanied by a Levels of Evidence rating at the end of the abstract.

How could Levels of Evidence be used by the AAOS? First, Levels of Evidence ratings could accompany abstracts submitted for the AAOS Annual Meeting. For the past three years the Pediatric Orthopaedic Society of North America has required abstracts submitted to the Annual Meeting have an accompanying Levels of Evidence rating. After ensuring the levels of evidence were accurate, these ratings could accompany the published abstract and thereby immediately place the study into context for surgeons. Furthermore, Levels of Evidence ratings are likely to provide subtle pressure and, with time, promote better quality studies in orthopaedics.

Another way in which Levels of Evidence ratings could be used is in the major educational outputs of the AAOS, such as Orthopaedic Knowledge Update or the Journal of the American Academy of Orthopaedic Surgery. In these review-type publications, Levels of Evidence for individual studies could be summarized together using Grades of Recommendation, such as those of the Canadian Task Force on Preventative Health Care, where Grade “A” indicates high quality, consistent evidence, Grade “B” fair evidence, and Grade “C” indicates conflicting evidence in support of treatment recommendations.

Dr. Bucholz states new strategies are needed. Levels of Evidence could be one strategy to maintain the AAOS as the leader in orthopaedic knowledge.

Yours truly,

James G. Wright, MD, MPH, FRCSC Robert B. Salter Chair of Surgical Research Associate Surgeon-in-Chief Senior Scientist and Program Head, Population Health Sciences, Research Institute, The Hospital for Sick Children Professor of Surgery, Public Health Sciences, and Health Policy, Management and Evaluations, University of Toronto

References

1. Bucholz,RW, Knowledge Is Our Business, JBJS, 86-A,1575, 2004

2. Wright, J. G.: Introducing Levels of Evidence to the Journal. JBJS, 85 -A(1): 1-3, 2003.