The Economics of Oncology: Why Oncologists Should Get Involved in Practice Guideline Development

The environment for medical practice is changing very rapidly.Much greater accountability is being demanded of all health-careproviders, including oncologists. Issues related to the philosophyof care, financial management, and ethics are all being scrutinizedmuch more carefully than in the past.

The management of cancer patients constitutes a progressivelylarger portion of the health-care industry budget. Thus, withalmost any of the possible new reform models for health care,oncologists are going to be challenged much more than in the pastto provide information on the value of cancer care, asserted Dr.Paul N. Anderson at an ASCO Health Services Research Committeeprogram, "Principles of Practice Guidelines," held lastyear. "We need then, as cancer care providers, to becomevery visibly and dependably accountable for outcomes, appropriateness,necessity, effectiveness, and the cost of cancer management,"said Dr. Anderson of the Cancer Center of Colorado Springs.

One definition of value is benefits minus harms over cost. Althoughoncologists already have some good information regarding the benefitsof cancer treatment and are amassing more data related to harmsand toxicities, they are just beginning to get involved in thearea of cost-benefit and cost-effective analysis, Dr. Andersonnoted. He contends that each of these components in the valueequation will become progressively more important to clinicians'ability to take care of cancer patients.

Dr. Anderson described several expected changes in health carehe believes will have a significant impact on oncology practice.First, it is anticipated that budget restriction, particularlycapitation, will be a major mode of payment. Under capitationplans, cancer subspecialists, such as radiation therapists, medicaloncologists, and surgeons, will compete with one another for eachdollar allotted to cancer treatment. In addition, the variousscans, follow-up examinations, and laboratory procedures willall vie for those same limited dollars.

Second, in all of the reform plans, there is a movement towardhaving primary-care doctors provide more cancer care than before.The third major change will involve switching from a mode of cancercare delivered with more and more services to achieve what isfelt to be a higher level of quality. Rather, clinicians willhave to provide data on outcomes, benefits, and costs to theircapitated revenue program in order to justify their diagnosticand treatment approaches.

A Fourth Major Change

The fourth major change, according to Dr. Anderson, is that oncologistswill increasingly become integrated into multispecialty hospitalphysicians' organizations, vertically integrated organiza- tions,HMOs, or other types of managed care networks. Moreover, cancercare will be delivered as part of a benefits package rather thanas unlimited individual patient selection of treatment.

As one key response to all of these changes in health care, Dr.Anderson advised oncologists to become experts in the technologiesof health services management. In particular, he said, practitionersought to become well versed in four technologies that currentlyare very much in vogue:

1) Outcomes definition, management, and evaluation;

2) Technology assessment;

3) Guidelines development; and

4) Oncology-defined cost-benefit and cost-effectiveness analysis.

"For our patients and for the perpetuation of excellencein cancer care, we need to become skilled in these methodologies,"Dr. Anderson asserted.

There are many reasons why oncologists should become involvedin these areas, Dr. Anderson stated, most of which are self-evident."We need to be able to organize evidence and outcomes betterthan we have before. We need to be able to deal with insurancecompanies and HMOs that are developing guidelines across thiscountry and in a relatively inconsistent manner. We need to beable to develop guidelines that guarantee access of cancer patientsto cancer specialists, to clinical trials, and to off-label drugs,amongst other issues. We need to be able to provide our own health-careentities with information for marketing, patient satisfaction,and these days, 'report cards' for the outcomes of cancer care.We would like to be able to avoid the proprietary guidelines whichare not really clearly related to cancer data and outcomes."In addition to involvement in guideline development on the nationallevel, Dr. Anderson urged oncologists to get involved in regionaland, especially, local efforts so that they can affect the systemsin which they work.

The purposes of clinical practice guidelines are sometimes confusingand overlapping, said Dr. Anderson. "The overall purposeof clinical practice guidelines and outcome-related activitiesis to provide physicians and their patients with recommendationsfor medical management in order to improve medical care. A secondbenefit is to reduce inappropriate care."

While the goal of improving care is fairly straightforward, inoncology the goal of reducing inappropriate care is much moredifficult, Dr. Anderson conceded. Many cancer treatments are beingchallenged; the value of using second-, third-, or fourth-linetherapies is under particular scrutiny. Questions also are beingraised over whether patients with far-advanced, metastatic, orrefractory disease or those with a poor prognosis should be treatedat all, particularly when good outcome data are not available.Consequently, oncologists need to become more involved in decidingwhen, for whom, and how resources can be expended most helpfully.

Attributes of Good Guidelines

The Institute of Medicine has defined practice guidelines as systematicallydeveloped strategies to assist practitioners make decisions aboutappropriate patient care for specific clinical circumstances.Attributes related to the development of good guidelines havebeen worked out extensively by the American Medical Association,the Agency for Health Care Policy and Research, and several otherorganizations. Currently, at least 200 medical organizations havecreated approximately 2,000 to 3,000 guidelines of varying degreesof quality, Dr. Anderson noted. This number does not include thevarious proprietary guidelines.

Dr. Anderson enumerated a long list of criteria for good practiceguidelines. First, he said, the practice parameters should bedeveloped by or in conjunction with physician organizations, notjust insurance companies or government agencies. The sponsoringorganization must have clinical and scientific expertise in thecontent of the guideline, which is not always the case with proprietaryguidelines. Broad-based involvement of the physicians who willuse the parameters is very important as well.

"A guideline needs to be linked first, to evidence upon whichthe guideline is based and second, to the outcomes which it ishoped the guideline will produce. This requires very careful attentionto the methodology required to produce the guideline, but it isvery obvious in many situations that evidence-based methodologyis not used as widely as it should be."

Thus, asserted Dr. Anderson, the parameters must not only be comprehensiveand specific but also should be based not just on opinions, butrather on reliable methodology that integrates research findings,including evidence from the scientific literature, results ofclinical trials, large databases, and clinical expertise. Theevidence on which the parameters are based should be describedin the document, and the review and scoring process for evidenceshould be spelled out. The credentials and affiliations of theexperts should be noted. The parameters need to be peer reviewedand endorsed by relevant organizations.

Another important element that should be included is a discussionof specific clinical management strategies, said Dr. Anderson."Appropriateness should be defined: whether the procedureis appropriate, equivocal, or inappropriate with varying categoriesof patients. Risks and benefits should be discussed, and expectedoutcomes or desired outcomes noted. Economic outcomes also shouldbe estimated. This is not always happening in the guidelines thatyou and I are being exposed to around the country."

The parameters should be based on current information. They shouldbe revised, updated, or, if necessary, rewritten at least every3 years. They should include criteria for outcome and utilizationevaluation. In addition to the guideline itself, there needs tobe a plan for disseminating the information contained therein,and for educating clinicians about and implementing the guideline'srecommendations.

Finally, patients' rights and preferences need to be consideredwhen guidelines are developed, Dr. Anderson stressed. As the Agencyfor Health Care Policy and Research has stated very clearly, guidelinesshould be very sensitive to patients' rights to make decisionsbased on their preferences.

When developing guidelines, clinicians must be aware of the purposesfor which the guidelines will be used, Dr. Anderson warned, asthey are very often used for other purposes besides helping patients."Guidelines in the health care industry are being used asdecision support for practitioners, for medical education andorientation, as criteria for self-evaluation, and as indicatorsfor quality review. They are used to minimize inappropriate variation,hopefully to improve outcomes and improve medical practice. Frequently,guidelines are being used for reimbursement decisions and choicesrelated to resource utilization. We hope they are used to improvepatients' quality of life and to evaluate new technology. Guidelinesare also obviously being used to credential physicians and incost-benefit analyses related to resource utilization."

ASCO Guideline Activity

Dr. Anderson concluded by summarizing the current status of ASCO'spractice guideline activity. Before the formulation of ASCO'smost recent strategic plan, a survey of ASCO membership indicateda very high interest in the development of clinical practice guidelines.As a result, the Health Services Research Committee was formedby the Board of Trustees, with a subcommittee focused on guidelinesdevelopment, technology assessment, and outcomes research. Thefirst set of guidelines developed by the ASCO committee, relatedto growth factors, has already been published [J Clin Oncol12(11):2471, 1994]. Other topics under evaluation or in the earlystages of development include stem-cell support for high-dosechemotherapy and the use of tumor biomarkers.

"It is very important too keep in mind that despite the involvementof our national organization, guideline development is a slowand cumbersome process nationally. We feel it is important thatindividual oncologists, local hospitals and programs, and stateand regional organizations become experts in dealing with thesetypes of methodology," Dr. Anderson concluded.