Penalties Loom for a Third of Outpatient CT Installed Base

Before January 2016, an estimated third of the outpatient installed CT base will need to be replaced, if providers wish to avoid the 5% penalty that Medicare will levy on exams acquired with outdated CT technology. The penalties are associated with an underreported quality measure in the Protecting Access to Medicare Act of 2014, the same legislation that prevented the 24% physician pay cut and that requires physicians to consult appropriateness criteria prior to ordering advanced medical imaging.

Penalties for exams acquired using noncompliant technology will soar to 15% in 2017. They apply to providers billing under the Medicare Physician Fee Schedule and the Hospital Outpatient Prospective Payment System.

The good news is that an estimated two-thirds of installed CT systems already meet (or can readily meet) the NEMA XR 29 (2013) radiation-dose standard with a software upgrade, according to Gail Rodriguez, executive director of the Medical Imaging & Technology Alliance (MITA), a division of NEMA. The association estimates that 40% of the two-thirds already meet the standard and 60% will require a software upgrade.

Rodriguez can’t address the cost of upgrading, but she suggests that vendors are not likely to charge providers unreasonable sums. “I know that the companies made commitments to the FDA, and it is very much in their interest to make these things available as straightforwardly as possible,”
she reports.

Nonetheless, the cost of upgrades could add up, especially for hospital outpatient departments and freestanding providers with multiple systems to upgrade. In addition, providers (and others) generally begrudge the expense of software upgrades—at any cost. John Boone, PhD, vice chair of radiology at the University of California–Davis Medical Center and president-elect of the American Association of Physicists in Medicine (AAPM), reports that his institution paid a deeply discounted $70,000 per system to purchase iterative-reconstruction software, a more advanced dose-control feature than XR 29 (2013) requires. He reports that this was more expensive than it should be, but probably the best way to control dose.

Rodriguez describes the evolutionary development of the standard, the tremendous effort on the part of all industry stakeholders to address rising concerns about radiation dose, and MITA’s role in the provision to apply penalties to exams acquired on outdated equipment. “We’ve been working with the FDA, physicists, and physicians—the entire industry—for a long time to try to get at how to reduce dose as much as possible,” she reports. “For many imaging folks, this is one logical step in a series of steps we’ve tried to take over the past four years.”

In 2010, the FDA launched the Initiative to Reduce Unnecessary Radiation Exposure from Medical Imaging (following a series of widely reported overexposure incidents), in a big-tent effort that included the ACR®, the AAPM, and MITA. “They are driving us hard, trying to make sure that the equipment is easy to use and understand—and that it has as many dose-reduction, dose-reporting, and dose-recording features as possible,” Rodriguez says.

The XR 29 (2013) Standard

Rodriguez refers to XR 29 (2013) as a base standard in that it represents an incremental change over the previous three standards (introduced during the past four years). “These are not burdensome requirements: NEMA XR 25 CT Dose Check, alarm systems, and that sort of thing,” she notes. “Only very old equipment will have difficulty being (or be unable to be) upgraded. It’s really just the old, antiquated equipment that has to go.”

Many 64-slice CT systems either came with the dose-saving features or are upgradable to automatic exposure control with tube-current modulation. They have iterative-reconstruction software as well, Boone says.

“Earlier scanners (such as the 16-slice or four-slice systems), in general, do not have an upgrade path, as far as I know,” he adds. “The fact of the matter is that big companies do not develop new software for legacy systems.”

The XR 29 (2013) standard includes four components: two product features and two standards features. The product features are automatic exposure control and reference protocols. Rodriguez describes automatic exposure control as less a dose-reduction feature than a dose-optimization feature. It automates radiation-dose adjustments based on body mass and greatly reduces the number of images that must be acquired, she says.

Reference protocols are developed by physicists and clinicians and are loaded into the machine by the vendor. Rodriguez is uncertain whether those protocols will be standardized across vendors. The two standards-based features are DICOM Radiation Dose Structured Report, which makes the recording of dose information in a standardized electronic format possible, and NEMA XR 25 CT Dose Check, which automates dose notifications and dose alerts to warn operators and physicians when dose exceeds established thresholds.

The Physicist’s Viewpoint

In general, Boone believes, the quality measure will improve patient care by making the operators of CT systems more engaged with machine operation at both technologist and radiologist levels. Because many systems are in private-practice, office, and IDTF settings, the physicist might visit just once a year to conduct standard, annual quality-assurance testing, which is not sufficient for the time-consuming task of protocol evaluation, he says.

Boone advocates expanded technologist education and testing in the United States1 and asserts that efforts to standardize technology are an important initial step. “In many cases, CT technologists are not certified in CT, and those who are need a better understanding of the features of scanners that have an impact on radiation dose,” he says. 

When technologists have to operate CT systems from different vendors, the task is even more difficult—and in some instances, fundamentally confusing. “In many cases, the parameters that are used to adjust the automatic exposure control on the systems are completely different and hard to understand,” he says. “For example, with company A, you turn the parameter up to reduce dose; with company B, you turn the parameter down to reduce dose.”

According to Rodriguez, the push back from the provider community on the new requirement has been negligible. “Providers understand if they have the latest, greatest, and safest machine, they should be paid more for it,” she says. “I think that fits very nicely with Medicare and its priorities to reimburse differentially for higher-quality care.”

Likewise, the FDA has asked vendors to drive adoption of dose-reduction features, Rodriguez reports. “Medicare wants to pay for quality, so it made sense,” she says. “All of those trends coincided.”

Boone observes that many people detest regulations, but points out that in California, where law requires that the dose-measurement data (volumetric CT dose index and dose–length product) be recorded in the interpretive report, regulation has had an unintended benefit. The five University of California medical centers are collaborating in an initiative to standardize and optimize CT dose, across the system, so that patients receive the lowest possible dose to produce the necessary medical benefit.

“We now have an enormous database, which allows us to have a better understanding of our doses to patients,” he says. “We use these data to compare protocols across different CT scanners, and even to compare them with those of other institutions with the same scanners.” Averaged over tens of thousands of individuals, the information has been extremely instructive in formulating protocols, he reports.

Editor’s note: This article expands on a news report that first appeared at


  1. Boone JM, Hendee WR, McNitt-Gray MF, Seltzer SE. Radiation exposure from CT scans: how to close our knowledge gaps, monitor and safeguard exposure—proceedings and recommendations of the Radiation Dose Summit, sponsored by NIBIB, February 24-25, 2011. Radiology. 2012;265(2):544-554.