BBI Contributing Writer

In 1990, Gordon Guyatt, MD, currently professor on the faculty of the Health Sciences Center of Clinical Epidemiology & Biostatistics at McMaster University (Hamilton, Ontario), coined the phrase evidence-based medicine (EBM). The concept of evidence-based medicine is not new; its use has been under investigation since the mid-1980s. Under Title IX of the U.S. Public Health Service Act, the Agency for Healthcare Research and Quality (AHRQ) is charged with enhancing the quality, appropriateness, and effectiveness of healthcare services.

One of the ways that AHRQ accomplishes these goals is through research and promotion of improvements in clinical practice. As part of its mandate, AHRQ has oversight for the U.S. Preventive Services Task Force (USPSTF), a panel of healthcare experts that evaluates the latest scientific evidence on preventive medicine. The first task force, established in 1984, published the Guide to Clinical Preventive Services in 1989. A second task force released a second edition of the guide in December 1995. This edition, evaluating common screening tests, is available on the Internet at www.ahrq.gov/clinic/uspstfix.htm. The third task force has released more than 50 practice guidelines.

In theory, EBM implies that physicians make patient care decisions based on integrating individual clinical expertise with the best available external clinical evidence derived from a systematic search of medical literature, Guyatt said in a plenary address during this year's annual meeting of the American Association for Clinical Chemistry (Washington) in Orlando, Florida, in late July. This implies that physicians are aware of the evidence in support of using a specific intervention, and the strength of that evidence.

The practice of EBM by physicians, Guyatt said, involves several steps. The first is to formulate an appropriate clinical question, such as "Is this patient's anemia a result of iron deficiency or some other condition?" The physician then conducts a literature search to find lab test clues that will help in finding a solution. The physician embarks on a search of the medical literature, critically appraises the articles found and applies the results of the search to the patient's care.

McMaster University has included a course on the use of EBM in its medical residency program. The course teaches residents to follow the EBM analysis steps. For more common conditions, the EBM team has developed likelihood ratios for many lab tests based on traditional statistical analyses and bioinformatic treatment of data from the medical literature. Likelihood ratios provide a numerical probability score for pre- and post-hypothesis testing that a particular lab procedure will be of value in solving a defined diagnostic problem. "Many of our residents now carry a likelihood ratio card along with their pagers and other tools," Guyatt said.

But new tests and new applications of old tests are implemented every day. An AHRQ-supported study presented at the second annual conference of the Health Legacy Partnership (HELP) in Washington last October indicated that primary care physicians in the U.S. see a need for more outcomes research and easier access to research findings. However, a study reported in the Journal of the American Medical Association in October 1998 found that physicians have an unmet information need for two out of every three patients, but in most cases the clinician does not research a question due to the time requirements of journal literature searches.

Information technology is providing a solution to this dilemma. Electronic medical databases that provide diagnostic tools may make it easier for clinicians to obtain answers at the site of patient care. These databases, formulated as electronic clinical guidelines, have attracted interest in the healthcare industry, both in the U.S. and internationally. In its 2001 report, "Crossing the Quality Chasm: A New Health System for the 21st Century," the Institute of Medicine (Washington) encourages the use of clinical guidelines that make scientific evidence more useful and accessible to clinicians.

A number of healthcare and information technology companies have responded to the demand for these tools. In May, Cerner (Kansas City, Missouri) bought Zynx Health, a subsidiary of Cedars-Sinai Medical Center (Los Angeles, California). Zynx has developed evidence-based medicine solutions that will be integrated into the Cerner Millennium technology platform. In June of last year, Johns Hopkins University (Baltimore, Maryland) and American Healthways (Nashville, Tennessee) launched an Outcomes Verification Program to independently evaluate and verify the effectiveness of a wide range of models of outcomes interventions and their clinical and financial results at Johns Hopkins.

The U.S. Commerce Department's National Institute of Standards and Technology Advanced Technology Program has awarded a $9.2 million, three-year project for the development of a software infrastructure aimed to create electronic clinical guidelines — best practice benchmarks that provide clinicians with patient-specific recommendations at the point of care.