Groups Use Different Methods in Developing Guidelines

BUFFALO, NY—Evidence-based practice guidelines are needed to improve the quality of cancer care, several representatives from the professional cancer societies told the President’s Cancer Panel at a meeting hosted by Roswell Park Cancer Institute . But expert physicians also play a critical role in the creation of practice guidelines, particularly in situations where evidence is lacking or controversial.

“In the late 1980s, physicians missed significant opportunities to direct practice guidelines,” said William T. McGivney, PhD, chief executive officer of the National Comprehensive Cancer Network (NCCN). “Our process seeks to restore physician authority in determining appropriate practice guidelines.” The NCCN guidelines are based on scientific evidence integrated with expert medical opinion. “Medical autonomy must be returned to practicing oncologists,” he said.

In just 3 years, the NCCN has completed three volumes of guidelines that cover 90% of all cancers. NCCN guidelines are current, specific in their recommendations, and updated continuously by expert panels that include more than 500 physicians. The fourth volume, to be unveiled next spring, will increase the NCCN guideline coverage to about 95% of all cancers. [See page 3 for an interview with Dr. McGivney and Dr. Rodger Winn on the upcoming NCCN meeting.]

“We believe that physician experts must control the guideline process, especially where a consensus approach is needed due to controversial and/or insufficient evidence,” Dr. McGivney said. It is critical for patient care that these guidelines are driven by the medical profession, he added, “because we need our member institutions to follow them to make cancer care consistent, effective, and cost effective. The experts also play a continuing role in the re-evaluation process when new evidence is brought to light.”

While the NCCN is working to quickly create basic guidelines for all cancers, as well as convert them into patient guides, the American Society of Clinical Oncology (ASCO) has proceeded at a slower pace. ASCO uses a longer process for creating guidelines that relies on a stricter evidence-based methodology.

“While we have made important progress, there are also many pitfalls in guideline development and implementation,” said Mark Somerfield, PhD, director of ASCO’s Health Services Research Department. “Our challenge today is to determine how effective guidelines actually are in improving the quality of cancer care in this country,” he added.

ASCO currently has issued four evidence-based practice guidelines—on breast cancer surveillance, treatment of unresectable non-small-cell lung cancer (NSCLC), use of tumor markers in breast and colorectal cancer, and use of hematopoietic colony-stimulating factors (CSFs). Ten additional guidelines are in development, and the Society is also modifying the breast cancer surveillance and advanced NSCLC treatment guidelines for a lay audience.

The ASCO review process uses several criteria to select a guideline topic: the cost of a procedure, importance of the subject, perception of high practice variation, and availability of evidence. When a guideline idea passes that review, a committee is formed to create the guideline.

Use of any ASCO guideline is optional among its members, and when the ASCO guideline for CSFs was first released, physicians were somewhat slow to apply it to their practices, Dr. Somerfield said. It was not until its use was emphasized to physicians that significant cost savings were realized with use of the guideline.

“It has been shown that there was a decrease in overall CSF cost when physicians were encouraged to implement the ASCO guideline through a mechanism of accountability,” Dr. Somerfield said. “We are now looking into the reasons for the slow acceptance of the CSF guideline by physicians.”

Currently, ASCO is asking its members and other payers and providers to evaluate its guidelines in terms of quality and satisfaction, and to assess whether the guidelines are leading to changes in physician practices. “Our ultimate goal is to learn if these guidelines impact the quality of patient care,” Dr. Somerfield commented.

To ensure that guidelines promote the highest standard of care for cancer patients, academic centers must play a role in their development, dissemination, and implementation, said Rodger J. Winn, MD, chief of the Section of Community Oncology in the Department of Clinical Investigations, M.D. Anderson Cancer Center, and representative for the American Association of Cancer Institutes (AACI). “I believe that our guidelines must promote the gold standard of care based upon the scientific evidence,” said Dr. Winn, who is also chair of the guidelines committee for the NCCN.

Dr. Winn believes that the most critical role for academic centers is to prevent bias in practice guidelines. Experts from cancer institutes offer multiple treatment approaches, including regional considerations, that must be assessed before a national guideline can be created, he said.

Academic centers also can examine existing guidelines and demonstrate that they are reproducible; that is, that a second group of experts agrees with the recommendations of a previously published practice guideline.

For example, the AACI looked at the same data that the NCCN and ASCO used in developing their guidelines for the treatment of non-small-cell lung cancer, and found that both groups used the same endpoints and came up with very similar treatment recommendations.

“We can show that these guidelines meet the criteria of quality care, but the academic centers should not stop there,” Dr. Winn said. “We cannot forget the need for clinical trials to further the scientific evidence. Academic centers play a pivotal role in coordinating care among community physicians, hospitals, and managed care companies, but their focus must always be on the addition of knowledge.”