Public CMS databases allow geographic glimpses of imaging use

Public data made available by the Centers for Medicare and Medicaid Services (CMS) paints an informative picture of the wide variation in imaging utilization across the country.

According to a study published online Feb. 5 in Radiology, authored by Ivan K. Ip, MD, MPH, with Brigham and Women’s Hospital in Boston, and colleagues, this information can be helpful in assisting imaging quality improvement initiatives.

In their research, they sought to examine geographic trends in providers’ use of diagnostic imaging (specifically CT and MRI) to identify targets for quality improvement initiatives for imaging referral in various hospital referral regions (HRR).

Using two CMS datasets, which became available to the public in April 2014, the authors found all claims for beneficiaries enrolled in Medicare’s fee-for-service program and selected those who had undergone an imaging procedure. The researchers assessed the provider’s use of imaging for each HRR and created an imaging referral index (IRI) across HRRs.

For 34 million Medicare beneficiaries in 2012, there were a combined 124 million unique diagnostic imaging services that totaled an estimated $5.6 billion.

They found that CT utilization intensity ranged from 330.4 studies per 1,000 beneficiaries to 684.0. For MRI, the numbers ranged from 105.7 studies per 1,000 to 256.3. According to the authors, head CT and lumbar spine MRI were the most common imaging exams performed in the cohort.

High-impact HRRs for CT, which fall into the highest deciles for both adjusted intensity and payment, included Baltimore and Jacksonville, Fla. There were 11 high-impact HRRs for MRI: Fort Myers, Fort Lauderdale, Orlando, and Jacksonville, Fla; Dallas; Palm Springs–Rancho Mirage, Calif; Las Vegas; Washington, DC; New Brunswick, N.J.; East Long Island, N.Y.; and Phoenix.

Ip and colleagues noted that previous studies have focused on utilization of imaging performed on patients who resided in a region and did not take into account where the exams were performed.

“However, because many interventions to optimize imaging use focus on providers, it is necessary to identify regional variation in providers’ use of imaging to identify potential targets for performance improvement initiatives,” the authors wrote.

In studying patient referral patterns across HRRs, the study team found substantial regional variation in provider’s use and payment of diagnostic CT and MRI procedures and believe that wide variations across HRRs provide opportunities for future outcome or comparative effectiveness research for clinical scenarios.

“Although national campaigns and federal policy changes may enable broad system-level improvement, our results suggest that more targeted interventions, focusing on providers that order and/or perform the most common procedures in the highest impact regions, may more efficiently promote needed change,” Ip and team concluded.

They urged policymakers to consider more targeted intervention campaigns that may prove both more cost-efficient and more effective than what the authors called “one-size fits all national campaigns” to reduce waste and improve medical imaging quality.