Evidence in practice: answers are within your reach.

As a graduate student in the 1970s, I was introduced to the revolutionary idea that a computer could be used to search the literature. I was skeptical that this new technology could both save time and discover articles that could be missed in a traditional "hard copy" search of journals and indexes. Nevertheless, my dissertation was looming, and I was willing to try anything to make life easier:

To perform a computerized search, students had to schedule an appointment with the university librarian. Only one librarian was trained in using this new technology, so appointments often had to be scheduled weeks in advance. When the big day finally came, I was ushered into a small, dimly lit, windowless room. The computer--which was the size of a minivan--filled most of the room. It emitted a soft, droning hum that sounded like a Gregorian chant. The librarian, seated solemnly at the keyboard, went through the ritual of dialing into the database. I sat some distance off to the side. I believed that this colossus could sense fear, and I did not want to get too close.

If the planets and stars were aligned correctly, the computer would connect to the database, and we could begin the search. I would offer specific keywords, and the librarian would enter these keywords and report the number of relevant articles or "hits" related to each keyword. We then would combine specific keywords and try to narrow the search to a reasonable number of articles. The problem, however, was that the time spent online was very costly. The user needed to fire out specific terms, pick the best combinations of these terms, and log off the system before the meager funds for these searches ran out. It was like a game show or playing "beat the clock," but there was no lovely parting gift (such as a year's supply of car wax or instant rice) for contestants who failed to use this new technology quickly and efficiently.

Computers are now the size of notebooks, and various databases can be accessed directly via the Internet. I find it especially remarkable that anyone can log onto the online version (PubMed) of the National Library of Medicine's database (MEDLINE), and, at no cost, have immediate access to abstracts of articles from more than 3,000 journals published since 1966. Other databases offer various features that can make the search easier or more focused on a specific question. As a result, online database searches are no longer restricted to students and academics squirreled away in cloistered alcoves. A magnificent tool has been placed literally at our fingertips.

As clinicians, we have a responsibility--both to our patients and to our profession--to make optimal use of this tool. We now have the opportunity to search the literature for evidence that can guide our practice and help us make clinical decisions. Many clinicians have embraced the technology; others are less familiar with the details of performing a search using a particular database.

To help clinicians use online technology, Physical Therapy began publishing Evidence in Practice--or EiP--in January 2002. EiP articles use patient examples to illustrate the process of how evidence can be gathered and used to guide clinical decision making. EiP articles describe how physical therapists can use online databases (MEDLINE, CINAHL, and so forth) alone or in combination to find evidence about the effectiveness of an intervention or another aspect of patient management. EiP highlights search strategies and techniques and the "bells and whistles" of a database that readers can use to make their own literature searches more effective and efficient. We hope that EiP can demystify and explain the process of the literature search and show physical therapists how they can apply the concepts of evidence-based practice in their workplace.

Judging from reader input, EiP has enabled readers to better understand the technical aspects of finding and applying evidence in the literature. Some readers, however, have interpreted these articles as practice guidelines or an attempt to document best practice, which was never our intention. We underscore the fact that EiP articles are examples of how evidence can be obtained to address specific clinical situations. Rather than promoting specific approaches to patient management, we are showing how evidence can be obtained and applied in our practice.

We have tried to show that it is sometimes difficult to find evidence that definitively answers each question. As clinicians, we must find articles that represent the best evidence for each question, but we then must use our expertise to interpret this evidence. I suspect that some physical therapists might be frustrated if they cannot find the "smoking gun" article that indicates the "ideal" method for intervening in specific clinical situations. But thankfully, such articles are the exception rather than the rule. Imagine how dogmatic and uninspiring our profession would be if evidence in the literature dictated a specific, rigid course of action for each and every patient! We are fortunate to be members of a profession in which the literature is often equivocal, because we have the opportunity to use this evidence to guide our approach rather than dictate a uniform and inflexible intervention in each situation.

Evidence also can indicate that an intervention is ineffective. This fact, which can be difficult for some clinicians to accept, was illustrated in several EiP articles. We must be willing, however, to give up long-standing beliefs that are based on anecdotal or relatively low forms of evidence when higher forms of evidence fail to support these beliefs. EiP articles have provided examples showing that not all evidence is created equal and that certain articles in the literature (randomized controlled trials, systematic reviews) can provide better evidence for clinical decisions than other articles (eg, expert opinion). Again, the clinician must ultimately weigh all the evidence, but an intervention that has been shown to be ineffective for a specific type of patient should not be incorporated into the plan of care for that type of patient.

Like any successful enterprise, EiP is the result of the efforts of many people. In particular, we were fortunate to recruit an incredibly bright and dedicated group of Editorial Advisors--Julie Fritz, Christopher Powers, Lisa Riolo, and David Scalzitti. These scholar-practitioners were instrumental in developing EiP, and they contributed most of the initial installments of this feature, with the indispensable editorial guidance of Assistant Editor Steve Glaros. We also draw heavily on the expertise of Jules Rothstein, Editor in Chief.

Above all, we want EiP to serve the needs of our readers and our profession. We welcome suggestions about how this feature might be adapted or improved in the future. We also welcome your ideas for articles. If you want to contribute an article to the EiP feature, please do not hesitate to contact me or one of the Editorial Advisors.

The ability to gather evidence is now within the reach of each and every clinician. The goal of EiP articles is to illustrate just how accessible this evidence can be and how it can be used to guide our interventions. The technology used to search for evidence will certainly continue to evolve--and the Journal will be right there with it, helping physical therapists incorporate evidence into practice.

Interested in writing an article for EiP? Have a topic that you believe would be appropriate for EiP? Please contact Charles Ciccone, Editor for Evidence in Practice, [e-mail].