Devices & Diagnostics, Health IT

Latest advance of UDI initiative to keep Big Data geeks busy

CMS is supporting the incorporation of FDA’s unique device identifiers in insurance claims forms, a boon to medical researchers who would receive a bounty of data about medical device safety and effectiveness.

healthcare big data

In case you didn’t think number crunchers are taking over the world in the age of Big Data, artificial intelligence and “deep learning,” the feds just handed them another victory, this time in the realm of medtech.

The Centers for Medicare & Medicaid Services (CMS) has come out in support of incorporating unique device identifiers in insurance claims forms. This would be a boon to medical researchers, and provide a bounty of data about medical device safety and effectiveness, according to medtech consultant Jay Crowley, who spearheaded the ongoing unique device identification (UDI) drive during his tenure at the Food and Drug Administration.

“If as a society we want better insight into the performance of implants, this is a step we need to take so we can react more quickly,” Crowley said during a phone interview.

Insurance claims are a unique and powerful data source because they provide a plethora of longitudinal information about a patient’s interaction with the healthcare system, Crowley said. Because the ubiquitous documents are standardized (private insurers typically follow CMS’s lead when it comes to claims forms) they can be easily aggregated across health plans.

Back in 2013 FDA laid the groundwork for unique device identification, which is designed to enable monitoring of individual devices via unique labels that are affixed to the device itself and electronically tracked. Inclusion of the UDI codes on insurance claims forms is necessary to reap the benefits of the initiative.

On July 13, Department of Health & Human Services Secretary Sylvia Burwell announced that incorporating UDI information in insurance claims forms would enable product evaluation and identification of safety concerns, provide patient data that would encourage innovation, and help payers and providers calculate and compare costs and outcomes of various device models. She estimated the implementation will cost CMS $300 million.


A mock UDI label

But CMS was previously against the incorporation of UDI into insurance claims forms, which makes Burwell’s stance all the more significant.

In a February 2015 letter to senators, former CMS administrator Marilyn Tavenner (who opposed the recent development), highlighted the complexity of the job ahead.

Tavenner wrote:

“Considering just high-risk implantable medical devices, 300,000 UDIs would need to be cross-walked to device dependent Medicare Severity Diagnosis Related Groups (MS-DRGs), Ambulatory Payment Classifications (APCs) and procedure codes for related physicians’ services. While the FDA’s Global Unique Device Identification Database (GUDID) could be used to verify DIs (device identification), claims edits would be needed to associate DIs with each different payment code used in every payment system by each payer.”

Others agree that the project could take as long as five years, at the very least.

“CMS updates the claims form on infrequent basis,” said Ben Moscovitch, an officer at the Pew Charitable Trusts’ Medical Device Initiative, during a phone interview. “The next update will be in approximately 2021. If we miss this window to add claims forms we may have to wait until 2030. That’s too long to wait to generate better data on device safety and performance.”

The wait might not be that long, however.

Due to lobbying from Sens. Elizabeth Warren (D-MA) and Chuck Grassley (R-IA), the FDA, advocacy groups like Pew, and medical associations like AMGA , patients now likely won’t have to wait until 2030 to see tangible benefits from UDI.

Also on July 13 when Burwell announced her support, CMS acting administrator Andrew Slavitt and FDA commissioner Dr. Robert Califf sent a letter to the chair of a standards committee asking that the UDI information be included in the next version of the claims form.

Currently, the claims forms indicate which type of procedure a patient received, but not the specific type of device used. So, a doctor or researcher can infer that a patient received a stent if they see that the procedure performed was percutaneous coronary intervention, but they cannot determine which type of stent the patient received, not to mention even more detailed information, such as the stent’s size.

But Crowley said certain UDI information will be excluded from claims forms. The UDI codes to be incorporated will enable identification of devices to the model level (for example, Boston Scientific’s Synergy stent or Abbott’s Xience stent), but it won’t enable tracking of a single Synergy stent or other individual implant due to technical and patient privacy issues.

That individual-patient level information would greatly improve the effectiveness of medical device recalls, but those will be incorporated into electronic health records instead, Crowley said.

The model-level data to be incorporate into insurance claims form will demonstrate the benefits (or lack thereof) of device across millions of patients, and enable the comparative analysis of competing implants. The data could also be used to accelerate insurance reimbursement, something that would delight industry.

Studies emanating from UDI data in claims forms will require intenseness statistical analysis, making this development another win for the number crunchers and Big Data geeks.

Photo: Flickr user Infocux Technologies