Surgeons use the literature to determine the best treatment for their
patients. Although practice can be changed by a single study, we often find
multiple studies more convincing. The appropriate literature can be identified
in many ways. First, we can search the literature ourselves. However, this
takes a lot of time and the search may not be comprehensive. Second, we can
identify meta-analyses or systematic overviews, which collect and combine
studies together to assemble a larger sample size in order to provide greater
precision around treatment effects. However, meta-analyses pose many
methodologic challenges and are not always available. Finally, we can use
review articles such as the Current Concepts Reviews published by The
Journal as a source for treatment recommendations. The difficulty with
review articles is uncertainty about the quality of the recommendations they
contain.
In this issue, The Journal introduces the latest in a series of
changes to help surgeons appraise the literature to make the best decisions
with and for their patients. In June 2000, The Journal featured a new
section called Evidence-Based
Orthopaedics1. On a
quarterly basis, structured abstracts derived from high-quality evidence from
journals other than JBJS are published with a commentary from a
clinical expert. This section provides the latest and highest-quality evidence
for orthopaedic surgeons. In January 2003, The Journal began
publishing a Levels of Evidence rating at the end of the abstract for every
clinical article2.
This system rates the quality of the evidence with use of five levels that are
based on the study methods. Levels of Evidence ratings provide a measure of
the quality of clinical articles published in JBJS.
With this issue, The Journal extends the evidence-based approach
to selected Current Concepts Reviews by applying Grades of Recommendation to
selected treatments discussed in the reviews. Authors writing Current Concepts
Review articles are encouraged to provide definitive and explicit
recommendations regarding the care of patients. Definitive recommendations are
most appropriate either when there is minimal disagreement among clinicians or
when the evidence is consistent and of high quality. If authors choose to make
their recommendations explicit, they should be formatted in a standard way.
Recommendations must be included in a separate table entitled
"Recommendations for Care." Authors should base their
recommendations on a complete evaluation of the appropriate literature.
Support for recommendations regarding treatment, diagnosis, or prognosis are
based on JBJS Levels of Evidence
(;
last accessed February 9, 2005). Levels of Evidence ratings for multiple
studies addressing a clinical care recommendation are to be summarized with
use of Grades of Recommendation. A Grade of Recommendation is to be attached
to each recommendation in the table entitled "Recommendations for
Care." Grade-A recommendations are based on consistent Level-I studies.
Grade-B recommendations are based on consistent Level-II or III evidence.
Grade-C recommendations represent either conflicting evidence or are based on
Level-IV or V evidence. A grade of I indicates that there is insufficient
evidence to make a treatment recommendation
(Table I).
Grades of Recommendation have been used most commonly in the development of
practice guidelines. In this process, organizations assemble all of the
appropriate literature on a clinical question. The quality of the literature
is appraised with use of Levels of Evidence. Finally, the overall quality of
the literature is summarized together as a Grade of Recommendation. The
JBJS Grades of Recommendation, modified from previously published
Grades
(;
last accessed February 9, 2005), were developed by a consensus group that
included representation from the American Academy of Orthopaedic Surgeons,
The Journal of Bone and Joint Surgery (American and British
editions), and Clinical Orthopaedics and Related Research.
Using Grades of Recommendation will improve the review articles published
by JBJS in two ways. First, the review articles will provide explicit
treatment recommendations for surgeons. Second, Grades of Recommendation will
guide surgeons as to whether they should definitely (Grade A) or probably
(Grade B) change their practice. Finally, by using this system, we also will
know those things that we do not know. In some clinical situations the surgeon
will not have good evidence, or even any evidence at all, upon which to make
an important decision, and in such cases the decision will have to be made
with the surgeon relying heavily upon professional judgment alone.
We look forward to your feedback on this important initiative.