On March 1, 2016, the Supreme Court substantially undermined the growing state effort to understand and control rising health care costs. The 6-2 opinion by Justice Kennedy in Gobeille v. Liberty Mutual Insurance Co. held that the Employee Retirement Income Security Act (ERISA) invalidates state all-payer claims database (APCD) reporting requirements for self-funded employee health plans, depriving states of essential information on health care utilization, pricing, and quality in the state.
In Gobeille, the majority concluded that ERISA, the federal statute governing employee benefits, trumps Vermont’s ability to include self-insured employer plans in a requirement that all health plans report data to the state’s APCD. The Court reasoned that Vermont’s APCD requirements are inconsistent with ERISA’s aim of providing a “single uniform national scheme for the administration of ERISA plans without interference from laws of the several States even when those laws, to a large extent, impose parallel requirements.”
In this post, we offer some thoughts on the implications of the case for state and federal efforts to control health care costs, and we discuss how states may access critical data from self-insured employer plans moving forward.
What Are All-Payer Claims Databases And What Is At Stake?
APCDs are large-scale, state-run databases that collect health care claims data and provider data from all payers in the state, including private insurers, public payers, dental insurers, prescription drug plans, state employee health plans, and others. Eighteen states have or are implementing similar databases, which are viewed as important tools in health care cost control. In addition to health care price, quality, and utilization data for each patient encounter, some APCDs gather data on patient demographics, diagnosis, services rendered, charges, payments, and procedure codes.
Without self-funded health plans’ data, APCDs will be deprived of a significant amount of information about private health insurance prices and services. Nationally, 63 percent of workers with employer-based health insurance are in self-funded plans. In Vermont, the ruling eliminates data from 20 percent of the total population from inclusion in the APCD. Moreover, those who get their insurance through employers tend to be healthier than those covered by public payers, so removing the majority of those with employer-based coverage from the database can skew the data and undermine the accuracy of any policy analysis performed using the data.
Without full private health care price and market data, states are in the dark if they want to address the growing problems of increasing health care costs, consolidation, and price variation. Recently, Zack Cooper, Stuart Craig, Martin Gaynor, and John Van Reenen released a landmark study on national private health care price variation that was completely at odds with Medicare data on health care spending. We cannot even begin to understand, let alone manage, the market dynamics driving a huge portion of our health care spending if we cannot access private price and claim data.
Immediate reactions in the press and on social media inaccurately characterized the Gobeille ruling as a win for insurers. But some health insurers and self-insured employers, particularly smaller ones with less market power, can benefit significantly from greater price transparency and regulatory efforts to reign in the pricing power of dominant health care systems. Typically, the opacity of healthcare prices tends to help providers and insurers with a great deal of market power secure higher negotiated prices and competitive advantage.
The Implications Of The Ruling For Health Care Cost Control Policy
Unfortunately for states and health care consumers, Gobeille involved a collision of two highly technical and complicated legal and policy regimes: ERISA preemption analysis and health care price transparency. This complexity obscured the importance and implications of the decision for health care cost containment and quality improvement.
The majority characterized APCDs as state-mandated information reporting requirements for health plans, conflating them with the “reporting, disclosure, and recordkeeping . . . central to, and an essential part of, the uniform system of plan administration contemplated by ERISA.” But the reporting and recordkeeping requirements in ERISA primarily relate to retirement benefits plans and their solvency, not employee health plans and their claims.
APCDs involve a reporting of data of completely different scale, scope, substance, and purpose than the summary-level reporting ERISA plans make to the Department of Labor regarding plan finances. By contrast, APCDs gather data on private health care prices, at a claim-specific level (meaning, on a per-encounter basis), and pair that price data with information on outcomes, referral patterns, which type of services were used, and quality. The Court’s interpretation that ERISA preemption encompasses any state reporting requirement related to an employee benefit plan represents a troubling expansion of ERISA preemption.
The impact of this expansion may seem unremarkable unless one knows that claim-level data on price and quality is almost completely inaccessible otherwise. Because of contractual confidentiality provisions and invocations of trade secret protection, without a state law requiring disclosure to an APCD states and other entities have no ability to comprehensively examine health care prices, demand, and quality across providers. That is why APCDs are often touted as promoting health care price transparency.
APCDs’ functions, however, go far beyond consumer price transparency. They form the basis for all state regulatory efforts to control health care costs by enabling states to monitor rising health care prices and their drivers, the effects of consolidation and other market changes on private prices and quality, and the ability of payment reforms to curb overutilization and improve quality. Public health initiatives and health services research will also suffer from the loss of significant portions of health data from APCDs.
While Justices Ginsburg and Sotomayor, as well as the federal government, emphasized the important distinction between APCD data and the data ERISA plans generally report to federal regulators about their own finances and solvency, the majority did not. Instead, the majority performed a rote analysis of ERISA preemption, applying the very “uncritical literalism” it professed to reject, with sweeping and potentially devastating implications for state health care cost control efforts.
Now that the Supreme Court has crippled state APCDs, states have three less-than-ideal options for obtaining the claim-specific price, quality, and utilization data needed to pursue health care cost control or quality improvement.
Option 1 – Use More Limited APCD Data with Voluntary Submission by Self-Funded Plans
First, states could continue to require APCD data from all other types of payers—fully insured employee benefit plans, public payers, and individual and small group plans within and outside the Exchanges—and encourage self-insured employee health plans to submit information on a voluntary basis.
Virginia has implemented an entirely voluntary APCD, and four other states have created voluntary data collection initiatives similar to APCDs. While these voluntary databases don’t have perfect data, they do have some data. Because self-insured employer plans cover nearly two-thirds of individuals with employer sponsored insurance, the missing data may compromise policy analysis significantly. However, for states that have sunk substantial resources implementing their APCDs (and are legislatively mandated to do so), it may be worth proceeding even with a shrunken data set. Perhaps self-funded plans can be convinced that it is in their interest to voluntarily submit data to state APCDs to drive reforms that improve their bargaining power against dominant providers.
Option 2 – Get the Missing Data from Providers
Second, the states could avoid ERISA preemption by requiring reporting from providers instead of health plans. Providers would surely balk at this request—they have less built-in data generation and reporting capabilities than health plans. This is a less efficient and more expensive (and possibly ineffective) way of getting the data, but hospitals, physicians, and other providers could be required to supply data that cannot be wrung out of self-funded employee health plans.
Option 3 – Pursue a Federal Mandate for Claims Data from Self-Funded Plans
Finally, the Court suggests that states should ask the federal government, particularly the Department of Labor but perhaps also the Department of Health and Human Services (HHS), to mandate collection of the APCD-type of data from self-funded ERISA plans. The Court, particularly Justice Breyer, seemed to believe this was a tidy and uncomplicated solution to the ERISA preemption problem.
This solution is actually harder than it sounds. No federal agency, whether the Department of Labor or HHS, currently collects anything like APCD, claim-level price and quality data. Even if one of these agencies agreed to collect plan data, to be effective it would have to be willing to gather the kind of timely, granular, and locality-specific data mandated by APCDs. Statistical or summary data would have little value to the type of analysis needed to assess, for example, whether the prices charged by a large health system jumped when they acquired a physician group. Justice Ginsburg noted in her dissent, “It is unsettling . . . to leave the States dependent on a federal agency’s grace, i.e., the Department of Labor’s willingness to take on a chore divorced from ERISA’s objectives.”
Nevertheless, a federal claim data requirement could be structured to replace the missing APCD data, especially if the primary data collection function was delegated to the states. Furthermore, a federal requirement could standardize data reporting requirements in a way that would facilitate data analysis and comparison between states. The best path forward now may be for the Department of Labor—in conjunction with HHS and states with successful and robust APCDs such as Colorado, New Hampshire, and Massachusetts—to develop a standard set of federally mandated claims data that all ERISA plans must submit to state APCDs and the federal government.
The Need To Fix ERISA
At the end of the day, Gobeille should serve as a clarion call that ERISA is in dire need of revision or repeal. The scope of ERISA preemption has gone far beyond the statute’s original intent to the point of dramatically hindering fundamental state functions for the sake of promoting uniformity in employee benefit requirements—which ERISA often fails to provide in any case.
While waiting for Congress to act (which many see as unlikely), states need health care utilization, price, and quality data to inform their health care cost containment policies, so they have few options but to push forward and get this data by any means necessary. Unfortunately, the Supreme Court’s decision in Gobeille just made it significantly harder for the states to do this critical work.