ACR Prepares ‘Appropriateness Guidelines’ for Radiologists’ Use

ORLANDO—Why is there such a wide variation in the way radiologists practice across the United States? “It’s a puzzle to me, why similar patients in different locations receive different treatments,” Philip N. Cascade, MD, said as he posed the question at the annual meeting of the American Society for Therapeutic Radiology and Oncology.

The work of the American College of Radiology (ACR) Task Force on Appropriateness Criteria, chaired by Dr Cascade, of the University of Michigan, may help reduce such regional variations by providing diagnostic and treatment recommendations reached through consensus by panels of experts.

The task force is comprised of 10 consensus panels—eight diagnostic and two therapeutic. The task force’s numbers are fluid, changing as needed, but currently it is comprised of 118 diagnostic radiologists, 78 radio-oncologists, and 37 medical specialists. While seeming a daunting task with so many cooks looking at the brew, Dr. Cascade is adamant that such diversity is the only way to get a clear picture for developing criteria that will be readily accepted.

The Methodology

The ACR developed its own methodology based on eight principles developed by the Agency for Health Care Policy and Research (AHCPR) for producing credible guidelines: validity (scientific basis), multidisciplinary (diversity of representation), reproducibility, clarity, clinical applicability, clinical flexibility, documentation, and scheduled reviews.

The intention, Dr. Cascade said, is to rely on scientific evidence as much as possible, but professional judgments and group consensus enter into the mix as well. “It turns out that even when you use sophisticated metaanalysis techniques, the science isn’t always clear,” he said. A good example of this is the controversy surrounding screening mam-mography between the ages of 40 and 49. “There would be no controversy if the science were clear,” he said.

The science is assessed by the panel leaders who fill out an evidence table based on a review of the literature. This table includes, for each reference, the study design, the number of patients in the study, the purpose of the study, the study results, and the strength of the recommendations.

“Each topic leader presents a summation of the major literature, and that’s how we address each topic,” Dr. Cascade said. “And at any time, the panelists can ask for the full articles and do the evidence tables on their own.”

Panel members then fill out a Worksheet Appropriateness Questionnaire (WAQ) on which they vote on the possibilities presented in the evidence table, using a scale of 9 to 1, with 9 being the most appropriate and 1 the least.

There are three rounds of voting. With each round, the panelist receives a spreadsheet showing how the previous voting went. It is anonymous, showing only how many voted one way or the other. The idea is to try to head toward a unified consensus, considered to be 80% agreement. “When there is no consensus, it just means the science is lacking and there’s a wide variation of opinion,” Dr. Cascade said.

When the voting is finished, the agreed upon criteria are published in a document that is four to six pages long. It includes a face sheet listing the potential tests or treatments and their appropriateness scores, as well as a narrative describing the assessment of the literature and the rationale behind the conclusions. The major references are also included.

Dr. Cascade stressed that the final treatment decision is always between the patient and the doctor. The published criteria include fine print explaining that they are guidelines and not a standard or what must be done. They are not to be used for reimbursement purposes for a third party, but are guidelines for patients and doctors.

The implementation of the guidelines Dr. Cascade believes, depends on their broad dissemination, “not just to physicians, but to the medical specialties as well,” he said. He also emphasized his personal opinion that the guidelines should be made available in lay language and distributed to patients, politicians, medical residents, and others who may need to understand them.