Complicated Comparisons

Comparative effectiveness research shifts into high gear, but can radiology find its place?complicated comparisons

Does your patient have a life worth living? That's one of the many questions that comparative effectiveness research (CER) endeavors to answer, according to Constantine Gatsonis, Ph.D., professor of Biostatistics and Applied Mathematics at Brown University in Providence, R.I.


"The scope of CER goes beyond traditional, narrow clinical trials," says Gatsonis. "It looks at the overall effect on the patient, including the quality of life." And since patients live in the real world, CER must be conducted in real-world settings, says Scott Gazelle, M.D., M.P.H., Ph.D., professor of radiology at Harvard Medical School in Boston. "Many biomedical and clinical trials and outcomes research are done in settings that are not real world," notes Gazelle. "In other words, some trials might compare treatment X to nothing or a placebo, but the real question in CER is how treatment X compares to last year's treatment."

Comparisons like these are critical to high-tech specialties such as radiology. "Imaging is a rapidly growing part of health-care spending," says Gazelle. "We have failed to generate the sort of evidence that we needed long ago to demonstrate the value of imaging." This proof is necessary, not only to establish how radiology adds value to medicine, but also to show what technologies do not contribute to patient outcomes. "In radiology, we develop instruments and test how accurate they are, but now we need to determine which instruments (or tests) we need and which ones we do not," adds Gatsonis.

"We're at a time where everyone knows we have unsustainable health-care cost growth," says William T. Thorwarth Jr., M.D., FACR, from Catawba Radiological Associates Inc., in Hickory, N.C. "We need to perform comparative reviews to demonstrate that one method or treatment is substantially better [than another] and warrants potentially higher costs."

Finding Comparative Costs

However, most CER studies do not include a cost-effective analysis (CEA). "That's kind of the 800-pound gorilla in the room," says Thorwarth. "Groups want to be funded for CER, but they realize that there's this political side; no one wants to talk about the cost or to appear as if payers or the government will make decisions based on cost."

Gazelle elaborates, "All arguments I've heard to not include CEA in CER are illogical. If we want to think about which therapy and diagnostics we want to choose, we can't do that without thinking about cost." The question of costs is one reason used to change the term CER to "patient-centered outcomes research." "At a one-day program at the Brookings Institution on June 9, 2009, Sen. Max Baucus (D-MT) pitched the name change ... so that the public wouldn't think that we're comparting A versus B and that someone in Washington will decide what [health-care treatment options] you get based on cost," says Thorwarth. To view webcasts on CER from the Brookings Institution program, visit

Whether it's called patient-centered outcomes research or CER, Gatsonis believes that there will be an overall cost consideration in looking at the effectiveness of procedures. "People will analyze and tabulate cost associated with procedures," he says. "After all, if payers want to say 'I'll pay for this,' somehow they must have in mind the amount of effort and research required."

complicated comparison in text 1

Funding and Recent Studies

The government has so far expanded a modest effort to fund CER studies. "No one really paid attention to [CER] until the American Recovery and Reinvestment Act, which set aside money for CER," says Gatsonis. The Agency for Healthcare Research and Quality currently has nearly 50 CER studies in the works, and the Patient-Centered Outcomes Research Institute just announced its first round of grants in December 2011, he says.

Recent CER studies within radiology cover ankle injuries, follow-up mammography, imaging for low back pain, routine chest X-rays in intensive care units, and screening mammography. These studies are summarized in the June 2010 JACR article, "Addressing 'Waste' in Diagnostic Imaging: Some Implications of Comparative Effectiveness Research."1 Additionally, the New England Journal of Medicine published a CER study on vertebroplasty for osteoporotic spinal fractures in August 2009. The researchers found that improvements in pain from vertebroplasty — injecting cement into a collapsed vertebral body — were similar to those who received a simulated procedure without cement.2

Unfortunately, it seems that some expansive treatment methods get promoted prior to evidence that they produce improved patient outcomes that are worth the added cost. For example, consider the new proton beam treatment facilities being built by the Mayo Clinic in Minnesota and Arizona and detailed in a January 2012 New York Times article co-written by Steven D. Pearson, M.D., M.Sc., FRCP, founder and president of the Institute for Clinical and Economic Review (ICER) at Massachusetts General Hospital and Harvard Medical School criticizing the cost of new proton beam treatment facilities.

In the article, Pearson and co-author Ezekiel J. Emanuel, M.D., Ph.D., head of bioethics at The Clinical Center of the National Institutes of Health and a breast oncologist, write that "Medicare will pay around $50,000 for proton beam therapy for a patient with prostate cancer, roughly twice as much as it would if the patient received another type of radiation." For a more in-depth look at the value of and debate surrounding proton beam therapy, read the upcoming June 2012 ACR Bulletin.

complicated comparison in text 2

Challenges in Radiology

Although more research is likely needed in the case of proton beam therapy, it may not be easy to uncover in the first place. In imaging, one of the main challenges is the specialty's indirect connection to patient outcomes. "Because we're one piece in the evaluation and management of patients, how do we decide what our contribution is relative to the patients' subsequent outcome?" asks Thorwarth. "Indeed, linking the diagnostic test to patient outcomes is more difficult conceptually than linking therapy to outcome," agrees Gatsonis.

Sean R. Tunis, M.D., M.Sc., founder and director of the Center for Medical Technology Policy in Baltimore, has suggested developing a program to help strengthen this linkage. "His connection," says Thorwarth, who heard Tunis speak at the Brookings Institute program, "was to develop an infrastructure that would help us develop reproducible patterns [of treatment] that would show the contribution radiology provides to a patient outcome." Tunis lays out his plans in "Strategies to Improve Comparative Effectiveness Research Methods and Data Infrastructure," a discussion paper available at

Focusing on the Benefits

In addition to Tunis and his work at the Center for Medical Technology Policy, the ICER is also entrenched in CER. Due to concerns regarding cost as well as level and quality of available evidence, ICER has been asked to review a large number of radiology treatments and procedures, says ICER Chief Review Officer Daniel A. Ollendorf, M.P.H. Representatives at the organization use a transparent, user-friendly, and accessible system (the "integrated evidence rating") to rate the comparative clinical effectiveness and comparative economic value of various treatments and procedures.

Uniquely, ICER also works with all key stakeholders, including patients and patient advocates, during its reviews. For example, ICER reviewed the evidence on available treatments for prostate cancer, and, in concert with a group of employers, payers, and providers in the Boston area, used the findings to create an evidence-based website ( that provides those diagnosed with low-risk prostate cancer with objective and comparative information on treatment options.

While CER promises good things for patients and radiologists alike, it's still in its infancy with a number of challenges — and opportunities — that remain undiscovered. "We are sitting here, only several years out from when CER really became part of the vocabulary and part of the law, but I can tell you that CER would be relevant whether or not it was included in law," says Gatsonis. "We need to know how effective things are and how they compare to what's out there."

1. Elshaug A, et al. “Addressing ‘Waste’ in Diagnostic Imaging: Some Implications of Comparative Effectiveness Research.” JACR 2010;7(8):603–13.
2. Kallmes D, et al. “A Randomized Trial of Vertebroplasty for Osteoporotic Spinal Fractures.” New England Journal of Medicine 2009;361:569–79.

By Raina Keefer
Raina Keefer (This email address is being protected from spambots. You need JavaScript enabled to view it.) is a freelance writer.

Share this content

Submit to FacebookSubmit to Google PlusSubmit to TwitterSubmit to LinkedIn