When ordering images, nurse practitioners, PAs far more trigger-happy than docs: study

Non-physician clinicians who have the authority to order imaging exams are much more likely to do so than doctors presented with similar patient cases.

The finding emerged in a study of a Medicare claims conducted by the Harvey L. Neiman Health Policy Institute and published in JAMA Internal Medicine.

The analyzed data turned up a 34% disparity in imaging ordering between primary care physicians and advanced practice clinicians (APCs) such as nurse practitioners and physician assistants. The orders studied were those that followed patients’ office visits for evaluation and management.

The authors noted that expanding the use of APCs may help alleviate the shortage of primary care doctors—but the cost-effectiveness of that solution remains an iffy proposition. “While increased use of imaging appears modest for individual patients,” they wrote, “this increase may have ramifications on care and overall costs at the population level.”

Co-author Richard Duszak, MD, chief medical officer and senior research fellow of the Neiman Institute, expounded on that point for RadiologyBusiness.com.

“The conventional wisdom from a policy perspective is that we as a society need to get more of these folks because they’re overall cheaper for care,” he said, pointing out that Medicare pays APCs at 85% of the rate it pays physicians. “Medicare basically gets a 15% discount when a nurse practitioner or PA sees a patient rather than a physician. But if these folks are really using more imaging, then maybe those savings will get washed out.”

Stressing that that the study is a broad, “foundational” project—the first in a series of more narrowly focused investigations to come from the Neiman Institute—Duszak said the sample Medicare dataset provided no insights into why APCs are quicker to order exams than physicians.

“My suspicion is that, when confronted with a condition in which an individual is less familiar, he or she is likely to err on the side of caution and order more imaging,” he said. “We see that all the time in the non-APC arena. A gastroenterologist sees a patient with chest pain, or a cardiologist sees someone with abdominal pain—in those circumstances, those specialists may be a little less judicious with their utilization of imaging than somebody who sees a given condition day in and day out.”

Duszak said the next hypothesis to be tested at Nieman may draw from the premise that, “when people are practicing within their tight scope of practice, they’re going to be much more adherent toward best practice utilization than when they start getting out in the penumbra of their clinical skills. That would have important implications with regard to how we best use APCs. Maybe there’s a different role for them in rural versus urban institutions, where there’s less or more clinical backup.”

Answers to such questions, he said, may create an opportunity for greater use of clinical decision support software, with its helps for people looking to follow protocols and best practices—including APCs who presently may be over-ordering.

“I think that’s probably the easy fix,” he said, “assuming the use of clinical decision support truly is as effective as I am predicting” at significantly reducing inappropriate imaging use and its associated costs.

JAMA Internal Medicine has posted the study online (subscription or purchase required).