Health Affairs: Non-rad MRImore studies, more costs
 |
Since the 1990s, more non-radiologists have acquired CT, PET and MRI scanners for their practices, enabling those physicians to self-refer and bill payors directly for scans. Moreover, with imaging orders and costs rising dramatically and reports indicating that non-radiologists may be responsible for 20 percent of in-office MRI and up to 40 percent of imaging overall (according to a presentation at RSNA 2010 by Ramin Khorasani, MD, MPH of Brigham and Women's Hospital in Boston), Medicare has begun scrutinizing and tightening regulations and reimbursements for imaging.
"More imaging may benefit some patients, but these incentives have also raised serious questions about the potential for inappropriate use," commented Laurence C. Baker, PhD, lead author and chief of health services research at Stanford University in Palo Alto, Calif.
"During the period studied here [1998 through 2005], reimbursements could greatly exceed the costs of acquiring and operating the equipment, making the services potentially quite profitable." Baker analyzed a 20 percent random sample of Medicare claims between 1999 and 2005 to compare MRI rates and costs between orthopedists and neurologists who did not bill for MRI and those who acquired MRI scanners and billed for the exams within the study period.
For orthopedics, 11,844 orthopedists with 1,129,660 episodes of care were analyzed. A total of 459,231 episodes were recorded among 5,993 neurologists. Of the 11,844 orthopedists studied, 3,535 began billing for MRI during the study, while 706 out of 5,993 neurologists began billing for MRI.
Imaging rates increased for both specialties over the study period, from 46 to 72 images per 100,000 claims among orthopedists and from 48 to 83 procedures per 100,000 claims among neurologists. The average increase in imaging volume for physicians who acquired in-house MRI was significantly larger. Orthopedists ordered 38 percent more MRIs after acquiring the scanners, while neurologists ordered nearly 20 percent more.
Baker also observed that even before the billing physicians acquired MRI they ordered more scans than traditional orthopedists and neurologists. Still, Baker said that it is "important to note that the trends for both acquiring and non-acquiring doctors were very similar in the pre-billing period. There is no evidence that billing doctors were increasing their use faster than traditional users in the pre-billing period."
"But at the time of acquisition of MRI, there was a distinct (and strongly statistically significant) increase," Baker continued. "In the first quarter after they began billing, orthopedists jumped from using about 20 more MRIs per 1,000 episodes than traditional users to using about 45 more. Neurologists jumped by more, increasing their rate by about 60 to nearly 140 procedures per 1,000 episodes more than traditional users." These heightened imaging rates for billing physicians held steady throughout the entire study period.
After adjusting for imaging trends over time, physician characteristics and case-mix, physician and outpatient spending for billing orthopedists rose by an average of $25 per case, a 2 percent cost increase. "Similarly, average ninety-day physician and outpatient spending for neurologists increased by about $91 after physicians began billing for MRI—about a 6 percent increase in spending, driven by spending for MRI and physician procedures." According to Baker, though, these cost increases were not only the direct result of greater MRI volume. "Acquisition of MRI generated increased spending not only for MRI but for other procedures, indicating that more MRI leads to expanded treatments for patients."
While acknowledging that his study did not investigate the reasons behind the increased imaging among billing physicians, Baker indicated that the incentives for profits played a substantial role in the spike of imaging orders. "One of the important features of the time period over which this study took place was the generous reimbursement for in-office imaging built into the Medicare fee schedule. This, along with generous reimbursement by private payors, was probably an important factor driving physicians' acquisition of the technology."
Proponents of in-office imaging and billing often cite the convenience, to patients and physicians, of self-referral. "The finding, however, that much of the observed increase in MRI use did not take place on the day of the initial visit seems to diminish the strength of the argument that convenience was the central driver [of study increases.]"
Nonetheless, Baker stressed the importance of further studies to examine the effects on quality of care, efficiency and patient access to care before proposing sweeping changes to Medicare reimbursement and regulation.
Still, Baker concluded that the "change [in MRI rates for self-referring physicians] appears primarily to reflect doctors' recommending MRI for patients for whom they would not have recommended the service before they acquired the ability to bill. ... Results show that beginning to bill for MRI was associated with substantial increases in the likelihood of patients' receiving MRI procedures."