On March 1, 2016, the U.S. Supreme Court held in Gobeille v. Liberty Mutual Insurance Co. that a Vermont law requiring comprehensive claims information to be conveyed to the state’s all-payer claims database was preempted by the federal ERISA statute and could not be applied to self-insured employers or their third-party administrators. Almost immediately, commentators on both sides of the political aisle objected to the ruling, which makes it seem as if nobody won the case. To my way of looking at it, a better conclusion is that nobody lost.
It is important to understand that the vote in Gobeille wasn’t close. Justice Kennedy (writing for the Court) and five other justices found the Vermont law preempted because data collection and reporting are core ERISA functions, and Justice Scalia would have agreed with them had he lived to see the opinion released. Only Justices Ginsburg and Sotomayor dissented, somewhat uncharacteristically defending states’ rights and resisting federal uniformity.
Taking into account both ERISA’s purposes and prior judicial interpretations of the statute, there were good reasons for the Court to resolve Gobeille as it did. My earlier post on the case, written shortly after the oral arguments last fall, explained the legal and political underpinnings and accurately predicted the outcome. It also noted that Congress very rarely opens the can of worms that is ERISA, so that Supreme Court rulings tend to be dispositive albeit incremental. Rather than repeat that analysis here, let me comment on the decision’s implications for transparency.
Several of these objectives are relevant to all-payer claims databases, which have been enacted in nearly 20 states but are uneven in their scope, accessibility, and purpose. Some states with databases are primarily interested in backstopping their regulators and in monitoring public expenditures, particularly Medicaid. With respect to private competition, information about claims and payments based on billing codes tends to be more understandable to corporate payers than to individual consumers, must contain prices actually paid to be useful, and if too current and too easily accessible to providers can facilitate collusion and stabilize prices rather than boost competition. Database laws that don’t include prices, or that grant data access only to government entities or researchers, have little likelihood of furthering competitive goals.
With respect to enhancing provider performance and creating a “learning health care system,” existing databases are far from perfect. Since the 1960s, U.S. health care providers have mainly collected the information they needed to collect in order to get paid. Unfortunately, such claims data presents a skewed version of reality. One can make a strong argument that the critical information for a “learning system” is clinical data, particularly the subset that can be used by providers to deliver care more cheaply, quickly, and reliably. This information may be more accurate and more useable if providers themselves, rather than the payers and employers who challenged Vermont’s law, are responsible for collecting and reporting it. Similar observations can be made about information that bears on the social determinants of health, very little of which is reflected in paid insurance claims.
What does this have to do with the Gobeille decision? Our patchwork system of health insurance has always involved a variety of overseers: HHS/CMS for Medicare and some of Medicaid; the U.S. Department of Labor (DOL) for employee benefit plans; and states for Medicaid and non-ERISA insurance coverage. Self-insured employee benefit plans became widespread in large part because ERISA prevented states from regulating them, and the Department of Labor was not a strong presence in health coverage before the ACA. Therefore, a court decision holding a state law preempted by ERISA meant that nothing replaced it. That was the so-called “ERISA vacuum,” and if Gobeille had been decided in such an environment the result would indeed have been to chill important data experimentation at the state level.
Things have changed. The ACA made a real effort to link the previous silos of insurance oversight into a more coordinated whole, and the Gobeille decision reinforces that commitment. Today, one can expect the now more engaged U.S. Department of Labor and the newly insurance-savvy HHS to work together, and with states, to systematize the collection of claims data (and hopefully clinical and population health data) and to learn from it. Because of the Supreme Court’s ERISA ruling, Vermont and other states can’t go it alone, with the risks and costs that such variability might create. But they can expect to work collaboratively with the feds on approaches that will serve both regional and national goals. Out of many, perhaps we are becoming one.
No comments:
Post a Comment