Evidence-based medicine offers the promise of improved health outcomes for our patients. The problem is that the evidence is often difficult to find, conflicting or insufficient to guide our practice. Sometimes, it is just impossible to keep up with the latest changes. For physicians who are interested in evidence-based medicine, clinical practice guidelines offer a solution. Guidelines can synthesize a comprehensive review, letting us know what the bulk of the evidence says. When the evidence does not exist or is insufficient, a good guideline, explicit in its limitations and methods and written by consensus, can still offer direction.

In this issue of American Family Physician, Kushner¹ reviews concussion in sports, relying on a guideline for the management of minor closed head injury in children that was jointly developed by the American Academy of Pediatrics and the American Academy of Family Physicians (AAFP) and published in 1999.² As one of AAFP's representatives on the panel that developed this guideline, I take special interest in seeing how others apply it. There are three key issues for the family physician to consider in reading Kushner's article. First, what are the limitations of the original guideline? Second, how do these limitations affect Kushner's view? Finally, does the article exhibit a subspecialist bias?

The panel members who developed the guideline were disappointed by the paucity of evidence on the subject. Our literature search identified 542 articles, and 64 articles were included in the review. Despite all this “evidence,” we never found some key data, such as a description of the natural history of minor head trauma (e.g., a person with minimal symptoms and a mild abnormality or small bleed demonstrated on computed tomographic scanning). Without this information, it is not possible to determine the true benefit of any intervention, nor is it possible to ascertain the risks of watchful waiting. In the end, we had to make a decision, and the decision was made by consensus. While not evidence-based, this decision and the process behind it are explicitly stated in the guideline. The bottom line is that the guideline Kushner used to support his article lacks some of the critical evidence we would like, but it is the best document available.

The question of how to manage a child with mild head trauma has been controversial, and the guideline, written as a collaborative product of the two major specialties caring for children, clearly sets the parameters for good care. However, the guideline refers to a specific patient group: “. . . previously neurologically healthy children of either sex two through 20 years of age, with isolated minor closed head injury . . . who have normal mental status at the initial examination, who have no abnormal or focal findings on neurological (including funduscopic) examination, and who have no physical evidence of skull fracture . . . who may have experienced temporary loss of consciousness (duration <1 minute) . . . evaluated by a health care professional . . . within 24 hours [of the] injury.”^{2(p. 1407–8)} In other words, this is a fairly specific patient group, and the guideline does not address all children with head trauma. Any attempt to apply the guideline to other types of patients should be interpreted with caution.

An important point to emphasize when considering whether a child can be observed at home is the definition of an “observer.” While Kushner's article refers to a “reliable observer,” the original guideline describes a “competent observer” who is able to comply with instructions for home observation. A key component of these instructions is the ability to return for care if the child's condition deteriorates. The ability to travel (e.g., living relatively close to medical care and being able to return at any hour) is an important part of the clinical assessment.

Finally, there is the question of how Kushner as a subspecialist—not only a neurologist but the medical director of a brain injury program—is able to relate to the needs of the practicing family physician. From my perspective, he did an excellent job.