Commentary (Kosecoff/Michelson): Current Status of Practice Guidelines in Oncology

It is somewhat bittersweet that an article about the need for cancer guidelines is being written in 1995. On the one hand, it is heartening that Dr. Winn has patiently organized a coherent framework for creating and implementing guidelines. On the other hand, it is quite bitter medicine to realize, as Dr. Winn points out, that even today, "follow-up studies.... have not been able to demonstrate significant changes in physician behavior."

Years ago, oncologists proved that people with diseases such as acute leukemia and Hodgkin's disease benefited greatly when they were treated strictly according to up-to-date protocols. Furthermore, deviations from these protocols produced worse outcomes. National and international trials have demonstrated repeatedly the need for standardization and use of sophisticated protocols for the management of patients with cancer. Thus, the groundwork was laid for greatly improving care of cancer patients through the adherence to protocols and/or guidelines. Based on this historical tradition and given the stakes in treating cancer patients, one would have predicted that oncology would distinguish itself as the branch of medicine that would provide leadership in the creation and use of guidelines. It is surprising, therefore, that oncologists have not embraced and promoted guidelines.

We welcome and heartily endorse the renewed interest oncologists are showing in guidelines and strongly urge that guideline development and implementation activities proceed with greater intensity. Fortunately, a number of developments will accelerate the guideline process. In recent years, there have been significant advances in guideline development methodologies, which focus on the value of evidence, use of expert judgment, and careful implementation. In addition, the advent of managed care has made the use of guidelines easier and more important. Managed care organizations have the ability to create structures and incentives that will promote the use of guidelines. They also need guidelines to provide specificity for the services patients will be provided. Finally, managed care companies must demonstrate to payors that they are providing consistent access to and high quality of care.

Implementation Must be Addressed

Once guidelines are developed, the focus shifts to ensuring their adoption. The literature and experience, up to now, indicate that changing physician behavior is hard. The use of guidelines depends at least as much on whether the delivery system is receptive to the behavioral change as on whether the guidelines are perceived to be valid and scientifically based. If, for instance, a guideline requires the use of an expensive drug to prevent nausea and nobody will pay for it, then changing physician behavior to prescribe this drug will be difficult, even if the drug produces the right outcomes.

Similarly, if a guideline requires physicians to spend more time explaining the outcomes of various cancer therapies to their patients and to perform a quantitative assessment of their preferences, these tasks will also be difficult to accomplish unless the system allocates sufficient time and reimbursement for them. The economic, organizational, and other characteristics of the health-care system must be supportive of the behavioral change that the guidelines are trying to produce. In short, implementation-that is, conforming real world practice to the guideline-must be addressed with at least as much intellectual intensity and creativity as the development of the guideline itself.

Two Red Herrings to Avoid

Cancer care can be improved if we commit ourselves to developing sound guidelines and implementing them in a professional manner. However, it is important to avoid two red herrings.

First, guidelines cannot be permitted to impede progress in medicine. On the contrary, as long as there is a commitment to updating and revising guidelines, the guideline development process itself will identify areas for which a lack of knowledge prevents experts from reaching a consensus on what should be done to a patient. These are the very areas that should be addressed in new and better clinical studies.

Second, even when guidelines are produced and regularly updated using a scientific process, there still must be opportunity for physicians to deviate from them. These deviations, however, should be justified by clinical theory or strong patient preferences. The argument that "breast cancer in New York is different from breast cancer in California" is not a sufficient reason to depart from a recommended protocol, particularly when the potential value of many cancer treatments is equivocal or unknown and the risks are real. On the other hand, analyses of justifiable deviations from guidelines will result in new clinical hypotheses about cancer care. These new hypotheses, in turn, can lead to new scientific knowledge and clinical practices that eventually will produce new and better guidelines.

Guideline development and implementation in oncology offers the best hope for improving care to patients now and in the future. Of all the many changes underway in the health-care delivery system, guidelines offer the best opportunity to improve clinical care, contain costs, and provide greater insights into clinical efficacy.

References:

Brook RH, Chassin MR, Park RE, et al: A method for the detailed assessment of the appropriateness of medical technologies. Int J Technol Assess Health Care 2(1):53-63, 1986.

Fink A, Kosecoff J, Chassin M, et al: Consensus methods: Characteristics and guidelines for use. Am J Publ Health 74:979-983, 1984.

Kosecoff J, Chassin MR, Fink A, et al: Obtaining clinical data on the appropriateness of medical care in community practice. JAMA 258(18):2538-2542, 1987.

Kosecoff J, Kanouse DE, Rogers WH, et al: Effects of the National Institutes of Health Consensus Development Program on physician practice. JAMA 258(19):2708-2713, 1987.