Study shows prior authorization curbs rapid imaging growth

A study combining the work of two commercial health insurance plans and a Medicare Advantage managed care program indicates that imaging prior authorization dramatically slows the use of high-tech imaging in the short run, but its impact decreases over time.

Commercial healthcare insurers have increasingly turned to prior authorization from radiology benefit management firms to address rising imaging costs. Despite strong physician opposition, more than 100 million people were subject to prior authorization for MR, CT, PET and nuclear cardiology, mainly covered by private plans in 2008.

Following a controversial recommendation from the Government Accountability Office, the Obama Administration endorsed the use of prior authorization for advanced imaging covered by the Medicare Physician Fee Schedule. Some 37 million elderly and handicapped people participate in outpatient Medicare Part B.  

The study by Jean M. Mitchell, Ph.D., a professor of public policy at Georgetown University, is one of the first scientific inquiries to measure how well prior authorization in radiology, often called radiology benefit management (RBM), works. In collaboration with R. Robert LaGalia, CFO of National Imaging Associates, Mitchell learned that it works quite well, especially when it first starts.

"RBMs appear to be a viable approach for controlling the rapidly escalating use of advanced imaging," she said in an interview.

Mitchell found a profound first-year change in the imaging utilization pattern in case studies covering a commercial healthcare insurer in several western states, a commercial carrier from a Midwestern state, and an east coast Medicare Advantage managed care plan.

Each contracted with NIA, a subsidiary of Magellan Health Services, for mandatory imaging prior authorization programs implemented from early 2004 to February 2005 after at least one year of double-digit utilization growth for MRI, CT, PET and other advanced imaging applications.

For the western states insurer, CT and MR use rates dropped 12% and 10% respectively in the 12 months after a radiology benefit management program was initiated. CT, which peaked at 60.9 procedures per 1000 enrollees in 2004 fell to 53.8 procedures per 1000 in 2005. MR use dropped from 46.6 to 41.9 scans per 1000 in the same period.

The experience with PET was skewed by relatively low procedural demand. Its use rose 50% in the year before RBM implementation, but still involved less than one scan per 1000 lives that year. In the two years, after implementation, it rose another 31% but was still only 1.1 scans per 1000.

This pattern was repeated after the RBM program was introduced by the Midwestern insurer in early 2005. The demand for PET increased 75% to 1.3 procedures per 1000 before implementation. It rose 15% in the first year after RBM was implemented before jumping 63% to 2.2 procedures per 1000 in the program's second year.

Nuclear cardiology utilization from applications often performed in a cardiologist's office experienced a complete reversal in the Midwestern case study. Prior authorization was credited for cutting the use rate by 22% in 2005, a year after demand rose 23% without regulation. The second year of RBM coverage, the demand declined another 2%.

Again in the Midwest, CT and MR both experienced double digital percentage increases the year before RBM implementation. Demand for CT and MR fell 9% and 15% respectively in the year after RBM was adopted. But most of that benefit was lost in the second year when CT and MRI use rose 11% and 8% respectively.

The same pattern emerged from Mitchell's analysis of the Medicare Advantage plan, although the demand for PET was up to 9 times higher for this generally older population than for the two conventional health plans. The CT and MR utilization rates for the Advantage program were more than 3 times and 1.2 times higher, again partially because of morbidity associated with older age, Mitchell said.

By introducing prior authorization, however, the contracted Medicare program slowed its 53% annual growth rate for PET to a 17% rise in the first year and a 12% increase in the second year after implementation.

For CT, the 13% increase seen in the year before RBM entrance was erased with a 14% decline in year after its introduction. MRI/MRA demand rose 11% in the year before mandatory prior authorization. It fell 8% in the first year of RBM involvement, though nearly all of that gain was lost in the second year.

Mitchell stressed that the second year results were mixed. The RBM efforts lost some of their sting as self-referring physicians learned how to manipulate the diagnostic codes to gain approvals, she said. In the backlash against prior authorization, some physicians were allowed to drop out of the programs. New applications, such a breast MRI, and increasing equipment inventories also fueled growth.

The study did not consider the impact of prior authorization on provider costs, efficiency, and healthcare delivery. It relied on a single RBM vendor for all three case studies. It did not include a control group for statistical comparisons and did not control for some variables that affected imaging demand during the trial.

Results were reported in the June 2009 Medical Care Research and Review Journal (Med Care Res Rev 2009; 66(3): 339-351).