Evidence-based medicine drives recommendations

If you ever serve on a work group charged with developing a clinical practice guideline (CPG) for the AAOS, you may want to watch what you say. The following phrases are out of place, according to William C. Watters III, MD, chair of the Guideline and Technology Oversight Committee:

“In my opinion …”
“My experience is …”
“I feel that …”
“I am sure that …”
“I was trained to …”

It’s not that individual opinions aren’t respected or that experience and training are discounted. It’s just that clinical practice guidelines arise from evidence-based medicine (EBM), which rests not on opinion or tradition, but on a critical appraisal of the best available clinical research findings—the evidence.

The Institute of Medicine describes a CPG as a “systematically developed statement to assist practitioner and patient decisions about appropriate health care for specific clinical circumstances.” At the Academy, the process of developing guidelines has evolved from a consensus-driven process prone to reflecting the participants’ biases to one that more faithfully follows the EBM tenets.

“The early attempts to develop clinical guidelines often were algorithms—at this point you do this, at this point you do that,”Dr. Watters says. “They had pitfalls such as lack of transparency and a real vulnerability to reflecting the viewpoints of either the individuals involved in their development or the source of the funding, which might be insurance companies, government, or medical societies.”

CPG developers painstakingly scour the literature on a given topic to identify high-quality investigations such as controlled, randomized trials with human patients that yield scientifically valid support of a treatment or procedure. (For a breakdown of levels of evidence as they apply to various types of studies, see below)

“The work group ranks these studies as good, not so good, and bad, and then just uses the good ones,” Dr. Watters says. “It’s not always obvious. You need to look at the design of the paper and decide how well that design was carried out. With enough good papers, the work group can make a recommendation and assign a level of strength to the recommendation. Or it might recommend something as an option and call for more research. Or it can say, ‘Just don’t do this; it doesn’t work. You’re wasting your time.’”

In the final published guideline, each recommendation has an assigned level of strength, and the language of the recommendation corresponds formally to the strength level (Table 1).

Sifting studies
One of the primary challenges of writing guidelines is the scarcity or absence of high-quality human studies for treatment approaches and procedures.

Dr. Watters notes that concepts and hypotheses in orthopaedic surgery are not intrinsically resistant to testing by trials with patients, and researchers have adapted to the new EBM priorities.

“We often hear that high-level data don’t exist,” Dr. Watters says, “but an increasing number of high-quality studies are being reported in orthopaedic journals. The impact of guidelines is being seen in the literature.”

EBM may be rooted in good science, but economics is driving the momentum for its comprehensive implementation. Both government and insurance companies are promoting evidence-based guidelines as a way to control costs.

“Government encourages evidence-based comparisons. Insurance companies will reimburse for proven techniques and products that bring real value to the patient,” says Dr. Watters.

Turning around an initial reluctance
Although the EBM concept has permeated the healthcare system, physicians have not necessarily rushed to embrace it in everyday practice, and the guidelines met some initial resistance.

“Physicians are individualists,” Dr. Watters says. “They want the patient to do well and the literature shows that physicians will modify their practices based on better outcomes.”

Dr. Watters advises physicians to view EBM and guidelines as a three-legged stool, supported by the best research evidence, clinical experience, and patient values. “Look at what the literature says, apply your experience, and listen to the patient,” he says.

The Academy is committed to promoting EBM and clinical practice guidelines, by presenting them in both scholarly publications (Journal of the AAOS) and educational programs (webinars and courses).

EBM fits squarely within the quality initiative launched by AAOS President John J. Callaghan, MD, as well. It provides a way to measure value. “Value is quality divided by cost over time,” explains Dr. Watters. “You have to have a basis of evidence before you can even discuss issues of quality and value. Quality is based on validated patient outcomes. The less expensive treatment may give a higher-quality outcome.”

The recently released guideline on “The Treatment of Osteoporotic Compression Fractures of the Spine” is the tenth to be issued by the Academy. The AAOS has also given approval to a guideline on lower back pain issued by the American College of Physicians and the American Pain Society. Also under way is a joint effort with the American Dental Association to develop a guideline for antibiotic prophylaxis in dental patients with joint replacements.

Levels of Evidence for Primary Research Question (PDF)

Terry Stanton is senior science writer for AAOS Now. He can be reached at tstanton@aaos.now