| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Dialogue on Evidence In Practice |
Directors,
Centre for Evidence-Based Physiotherapy
School of Physiotherapy
University of Sydney
PO Box 170
Lidcombe NSW 1825
AUSTRALIA
We encourage Journal readers to respond to articles in "Evidence in Practice." E-mail your contribution to the dialogue to steveglaros{at}apta.org.?
We would like to suggest an alternate search strategy for the "Evidence in Practice" article in the March issue written by Charles Ciccone, PT, PhD.
Dr Ciccone selected the CINAHL database because he wanted "to focus on journals that might deal specifically with physical therapy interventions." CINAHL is not restricted to physical therapy interventions. In contrast, the Physiotherapy Evidence Database (PEDro) (ptwww.cchs.usyd.edu.au/pedro/) is a database that is restricted to systematic reviews and randomized controlled trials (RCTs) of physical therapy interventions. We believe, therefore, that it might offer a potentially more efficient method to search for external evidence on the effectiveness of physical therapy interventions than generic databases, such as CINAHL or MEDLINE, could. In our opinion as the developers of PEDro, PEDro is a much more complete database of RCTs and systematic reviews in physical therapy than either MEDLINE or CINAHL are.
A search of PEDRO was conducted on March 6, 2002. By typing lymphedema in the abstract search box, 6 records were retrieved—1 systematic review and 5 RCTs—all of which were relevant to the clinical question. If the 2 alternate spellings of lymphedema (lymphedema and lymphoedema) were typed in the abstract search box and the OR function was activated, an additional 3 records were retrieved, and 1 of these, an RCT, was directly relevant to the clinical question.
For readers who are interested only in the high-level evidence provided by systematic reviews and RCTs, the CINAHL search would be disappointing because it retrieved only 1 relevant record (a systematic review1) from 40 articles or hits. In contrast, the first PEDro search yielded 6 relevant records (the same systematic review and 5 RCTs) from 6 hits, and the second search yielded 7 relevant records (the same systematic review and 6 RCTs) from 9 hits. In this case, the efficiency of the PEDro search is clearly greater than the CINAHL search.
One benefit of the CINAHL search was that it retrieved the Canadian clinical practice guidelines for the care and treatment of breast cancer2 that was not retrieved by the PEDro search, because PEDro does not currently include clinical practice guidelines. However, as the developers of PEDro, we have been aware of this limitation and have recently secured funding to expand the PEDro database to include evidence-based clinical practice guidelines.
We believe that PEDro is an efficient way to search for evidence of the effects of therapies, but we do not claim that PEDro includes all relevant reviews and trials. We suggest that systematic reviewers continue to use databases such as CINAHL and MEDLINE in addition to PEDro.
 | References |
|---|
Editor—"Evidence in Practice" and Reviews
Physical Therapy
I want to thank Dr Maher and colleagues for illustrating an important point. Databases vary in terms of their features and the types of information they can retrieve. Through"Evidence in Practice," we hope to show readers a diversity of approaches to finding evidence. Clinicians should be aware that they need to use different search strategies to answer specific questions and that they need to explore various databases to find the resource that works best in a given situation.
I selected CINAHL for the search in the March issue because it purportedly covers both the physical therapy literature and the nursing literature. The latter resource seemed important because of the role that nurses play in the management of conditions such as lymphedema. CINAHL offers access to the full text of certain articles, a feature that many clinicians find invaluable. I did not, however, intend to promote one database as superior to any other. My goal was to show that there are alternatives to using MEDLINE (the database used in our first article) and that this particular alternative was selected for the reasons stated above.
One issue raised by Dr Maher and colleagues may need some clarification. In their letter, they imply that clinicians may want to use a specific database if they are "interested only in the high-level evidence provided by systematic reviews and RCTs [randomized controlled trials]." Furthermore, they state that only these types of articles would be "relevant" for clinicians interested in high-level evidence.
Although I would agree that RCTs and systematic reviews theoretically provide the highest level of evidence, I would also emphasize that RCTs and systematic reviews vary in terms of quality and, therefore, cannot be judged categorically as providing a high level of evidence based on only their design. Moreover, information contained in another type of article such as a retrospective study may ultimately provide the most appropriate and relevant information to answer a specific clinical question. A retrospective study, for example, may be more helpful than an RCT if the retrospective study deals with patients or interventions that more closely resemble a given patient situation. It is important for clinicians to retrieve a variety of articles and then carefully decide which articles seem most relevant before ultimately judging the strength of the evidence based on the quality of each relevant article.
I again thank Dr Maher and colleagues for their insight, and I applaud the work they have done to promote evidence-based practice. We hope to incorporate some of their ideas in future articles that show readers how different databases may offer certain advantages in answering specific clinical questions and in guiding interventions. We also hope to do a head-to-head comparison of 2 different databases so that readers can see that the search results may vary depending on the database used.
Related Article
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Physical Therapy 2002 82: 276-282.
This article has been cited by other articles:
|
|
 |
|
  C. G Maher, C. Sherrington, M. Elkins, R. D Herbert, and A. M Moseley Challenges for Evidence-Based Physical Therapy: Accessing and Interpreting High-Quality Evidence on Therapy Physical Therapy, July 1, 2004; 84(7): 644 - 654. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
