Payers' price transparency data still not user-friendly, say researchers

Insurers have largely met price transparency goals that went into effect Jan. 1, but consumers gaining access to the data is one thing—their ability to use the information is another.

At present, there's still work to do to make those data usable, according to an article in Health Affairs.

“In light of the data’s limitations that increase information processing costs and impede the potential usefulness of this data source, policymakers should consider requiring insurers to provide more standardized information for plan and provider identification purposes as well as drug dosage unit details,” the article said.

David Cutler, Ph.D., an internationally known healthcare economist, echoed those sentiments.

“It seems likely to me that insurers are complying in the least helpful way possible,” Cutler told Fierce Healthcare in an email. “It is possible that some company will digest all this and make it useable to consumers; I hope so. If not, and even if one does, there needs to be additional regulation to make it useable.”

Sophia Tripoli, director of the Center for Affordable Whole-Person Care at the consumer health advocacy organization Families USA, said she couldn’t agree more.

“People want and deserve to have clear, accurate and accessible information about the cost of medical services before they go to the doctor or hospital,” she told Fierce Healthcare in an email. “As families are struggling with crippling medical debt, policymakers must grapple with what is driving our nation’s affordability crisis—big healthcare corporations buying up local competition so they can keep prices high and keep wages low.”

Under the Transparency in Coverage Final Rule, the Centers for Medicare & Medicaid Services (CMS) mandated that insurers and employers who self-insure must provide beneficiaries with prices for 500 services as well as establish online tools to help workers compare benefits packages.

Beginning Jan. 1, 2024, payers will need to do the same for all services that they cover.

Researchers at Johns Hopkins Bloomberg School of Public Health looked at just how well insurers have complied so far. They noted that such data have the potential to help researchers study the price mechanisms of the healthcare market, policymakers identify ways to lower healthcare costs and employers and employees who want to comparison shop for the benefits that best fit their needs.

Johns Hopkins researchers reviewed data from 240 insurers and—because most insurers offer more than one benefits package—317,987 health plans, unique products with their own plan identification number. Of those, 298,806 are employer-sponsored health plans, including self-insured plans.

Major national players represent a majority of the 240 health insurance plans, and include subsidiaries and sub-subsidiaries. Also included are data from 19,069 individual plans and small group plans sold on the Affordable Care Act marketplaces.

“The volume of in-network pricing data is substantially greater than that of out-of-network pricing data because the former contains prices negotiated between a health plan and various in-network health care providers for a given procedure,” the article states.

The Johns Hopkins researchers note that CMS placed the same mandate on hospitals, but so far only about a quarter of hospitals have met the requirements, according to an analysis by the group Patient Rights Advocate.

Hospitals and insurers have responded much differently to the CMS transparency mandate from the beginning. Many hospitals ignored the Jan. 1, 2021 ,CMS deadline for the Hospital Price Transparency Rule, saying that the regulations were too onerous, especially in the midst of the COVID-19 pandemic.

That’s an argument that the American Hospital Association made in court, in a case that it eventually lost.

Insurers, on the other hand, released a deluge of data on their deadline of July 1, 2022, so much so that no one—not even computer experts—could even begin to decipher it all. Kosali Simon, Ph.D., a nationally known healthcare economist at Indiana University, told Fierce Healthcare that “even to download one tiny, tiny part of it is a giant task, let alone open the file or understand the data.”

About the Health Affairs article, Kosali tells Fierce Healthcare that she too would like to see more information about medication. Kosali: “more standardized information on drug dosage unit—there’s no requirement that the price reported be for a high or low dose of that medication, and in reality the price likely varies by dosage.”

Insurers have much more to lose for noncompliance and “the substantial noncompliance penalty—$100 per day per impacted individual—has likely contributed to high compliance among insurers relative to hospitals,” the Health Affairs article said.

The data examined by the Johns Hopkins researchers contain 599,204 procedure codes as well as 56 billion in-network negotiated prices, 96% of which are professional fees and 4% are institutional fees.

“These prices can also be grouped based on their price-setting type, namely negotiation-based (92%), fee schedule-based (6%), percentage-off-charge-based (2%), and per diem and derived price (fewer than 1%),” the article states.

The insurer price transparency effort releases too many data in certain categories but not enough in others. For instance, the Health Affairs article said insurers don’t have to disclose plan size, service utilization, quality outcomes, benefit design and patient characteristics.

In addition, insurers identify contracted employers using either an employer identification number (EIN) or a national provider identifier (NPI). But they do so without disclosing why one was chosen over the other.

The Johns Hopkins researchers argued that “creates barriers for users interested in merging the insurer price transparency data with external datasets that contain provider characteristics. Since NPI and EIN could refer to different levels of provider organizations … standardized provider identification would facilitate the merging and thus decrease users’ data processing cost.”