Evidence Matters

Purpose

“When it comes to health—Evidence Matters!” This is a statement heartily endorsed by Evidence Matters Chief Executive Officer and Cofounder Ofer Allan Avital. The general purpose of the Website is to provide quick and easy access to medical evidence extracted from peer-reviewed research articles, allowing the clinician to choose the best treatment options for a patient with a confirmed diagnosis.

General description

Evidence Matters has been described as a “new generation online database” [1], meaning there has been a progression from manually searching indexes (such as Index Medicus) to searching the next generation of electronic (linear) bibliographic databases (such as MEDLINE and CINAHL). The electronic databases allow the user to search by keyword and retrieve a list of references and, in some cases, the full text; however, the clinician must then sort through the articles, read the pertinent ones, critically evaluate them, and apply the results or answers to their patients.

The third-generation database, Evidence Matters, is a clinical knowledge management system that uses a simple “ask-a-question” interface to retrieve the latest peer-reviewed research. The results or answers from the articles are synthesized into article summaries, tables, and graphs with links to the original references. The database provides users with the answers first and references second, focusing on the effectiveness, safety, and costs of therapy options and utilizing an evidence-based approach to searching the medical literature.

At present, subscribers to this database are institutions; although there are plans to offer a pricing model for individual subscribers in the future. Institutions may choose access based on a fixed number of concurrent users, or they may select the unlimited access option. Subscriptions are on an annual basis, and institutions may register for increased functionality, so prices can vary.

Content

Information included on the Evidence Matters Website is available in English and French. Currently, the Evidence Matters database has three main modules, covering twenty major diseases. Six modules are scheduled to be added to the database in 2006. The existing modules are:

• Cardiology

• Endocrinology

• Oncology

Diseases are chosen based on the following criteria:

• search requests

• burden of illness to society

• volume of research turnover

An example of high-volume research turnover is the oncology module that includes types of cancer involved in more than 90% of cancer care incidents. The main cancers currently included are: breast cancer, cervical cancer, lung cancer, colon cancer, and prostate cancer, with new diseases being added every few months.

Whenever possible, Evidence Matters links the user from the search results and article summaries to the full-text version of resources. During 2006, the availability of full text should increase due to a distribution agreement between Evidence Matters and ProQuest that will provide access to all full-text resources available in ProQuest products.

Intended audience

The intended audience for Evidence Matters includes clinicians as well as librarians, teachers, researchers, students, and patients. The “ask-a-question” interface is designed with the end user in mind. It can be more intuitive for the average user than the Boolean queries required when searching many other databases.

In this era of the “empowered” patient, search tools that assist the user with accessing reliable health information are in demand. Evidence Matters does an excellent job of translating complex scientific evidence into standardized templates that present the results in a consistent way regardless of the length or complexity of the original articles. Nevertheless, patients should seek guidance from clinicians when interpreting search results—just as they would when interpreting results from searching any other medical database.

Major features

• The PICO model: A main feature of this database is its “ask-a-question” interface that employs the patient, intervention, comparison, and outcome (PICO) model for formulating a clearly focused clinical question. The user may apply advanced filters as well, such as: timing of outcome measurement, sex, race, age range, and blinding characteristics. The advanced filters are context sensitive, so when the user creates a new clinical question, the advanced filters change accordingly.

• Information sources and quality: To obtain new articles for Evidence Matters, electronic database searches are conducted daily or several times weekly for each disease area, with the data coming primarily from research appearing in peer-reviewed scientific journals. Over 12,000 journals are scanned via databases such as MEDLINE, CINAHL, and Cochrane. Articles are selected that deal with therapeutic effectiveness, safety, and costs. To date, approximately 400 journals are included in the Evidence Matters database, signifying that articles in this subset of journals satisfy the abstractors' search criteria. The listing of journals is an ever-expanding, “open list” because new journal titles are added as the abstractors locate relevant articles. References go as far back as 1970, with a 5-year span of coverage being the minimum for most journals.

• Search protocol: Searches are conducted daily in databases such as MEDLINE, CINAHL, and Cochrane, with search terms including the appropriate names and synonyms for disease areas. Abstractors also perform manual searches of disease-specific resources. Studies are retained that describe trials of interventions on humans, and all study designs are accepted, from randomized controlled trials to case reports.

• Data extraction and review: Most of the Evidence Matters reviewers have medical degrees, and some have masters' degrees in library science. Information extracted from articles undergoes a three-tiered review process. First, specially trained reviewers conduct an initial data extraction and review, and a second blind review of the generated article summaries is performed by staff editors. A third review is done by any user who thinks there might be a discrepancy or error in any statistic or article summary. The user who notices a discrepancy can contact the Evidence Matters staff from the related Web page. If there are no further corrections after a six-month period, the article in question is marked as having been viewed for that period of time with no corrections.

Summary of features

• Simple “ask-a-question” template allows the user to build a clearly focused clinical question.

• Search results are displayed clearly in article summaries, tables, and graphics with links to original references.

• Thousands of scientific research articles from authoritative sources have been indexed by clinicians and medical librarians.

• Advanced search filters are available: sex, age, ethnicity, etc.

• Information has been pre-abstracted by trained professionals and linked to a search.

• Statistics and outcomes are standardized, using the same type of statistic to evaluate similar clinical outcomes.

• Disease areas are updated on a regular basis

Usability

According to Avital, the usability of the Evidence Matters Website has been tested through focus groups and questionnaires. The ask-a-question format makes it easy for the user to build a clinical question, and the Website provides an animated demo to guide the user through the process of using the Evidence Matters database [2].

Advantages

Evidence Matters is accessible via the Website, and a personal digital assistant (PDA) version is under development. The main advantage of this product is its user-friendly, ask-a-question interface. The “Question Wizard” makes it possible for even a beginner or patient to build a clearly focused question. Of course, a patient should seek the guidance of a clinician for assistance with the interpreting the search results.

Disadvantages

The strength of Evidence Matters is its ability to simplify a large body of research in a therapeutic area. Its weakness, however, is that it does not cover rare diseases about which a small number of studies may have been published. A second weakness of the database is its limited number of modules. There are plans, however, to add the following modules in the near future: neurology, respiratory disorders, gastroenterology, nephrology, orthopedics, gynecology, and pediatrics.

Similar products

Evidence Matters is comparable to Google Scholar, PubMed, Cochrane, and UpToDate, as shown in Table 1.

Databases and tools for evidence-based treatment decisions

Future plans

Future plans include the implementation of additional modules, filters, and other search features—an ambitious, labor-intensive, and costly undertaking. If these plans come to fruition, then Evidence Matters will become an even more valuable resource for the busy clinician who wants to practice evidence-based medicine in the most efficient and effective manner possible.

Conclusion

Evidence Matters offers some unique features that go far beyond providing a list of references. It synthesizes the data in a meaningful way, customizes the content, and gives the user the power to choose safe, effective therapies based on outcomes. The context-sensitive filters provide enhancements that make Evidence Matters the ultimate tool for the optimum search. The potential of this valuable search tool will be more fully realized as modules, with their accompanying disease areas, are added to the database.