This analysis is part of the USC-Brookings Schaeffer Initiative for Health Policy, which is a partnership between Economic Studies at Brookings and the University of Southern California Schaeffer Center for Health Policy & Economics. The Initiative aims to inform the national health care debate with rigorous, evidence-based analysis leading to practical recommendations using the collaborative strengths of USC and Brookings. This post originally appeared in Health Affairs on July 16, 2019.
Fellow and Associate Director - USC-Brookings Schaeffer Initiative for Health Policy
Research Scientist - USC Schaeffer Center for Health Policy and Economics
Nonresident Senior Fellow - Economic Studies, USC-Brookings Schaeffer Initiative for Health Policy
Senior Fellow - USC Schaeffer Center for Health Policy and Economics
Nonresident Senior Fellow - Economic Studies, USC-Brookings Schaeffer Initiative for Health Policy
Co-director - USC Schaeffer Center for Health Policy & Economics
Nonresident Fellow - Economic Studies, USC-Brookings Schaeffer Initiative for Health Policy
As the Senate and House appear to be coalescing around a compromise proposal that would end surprise out-of-network billing and reduce patient costs, Representative Ruiz (D-CA-36) recently introduced H.R. 3502, an alternative approach to dealing with surprise out-of-network billing that has attracted 52 cosponsors in the House as of July 12, 2019.
The new legislation introduced by Rep. Ruiz would, like other proposals, prohibit balance billing for surprise out-of-network services and limit patient cost-sharing to standard in-network amounts. More distinctively, the bill proposes an arbitration process to determine how much the insurer must pay to the out-of-network provider. Although provider-payer dispute resolution is a feature of some state-based surprise billing laws, the Ruiz bill raises great concern by instructing arbiters to consider the 80th percentile of billed charges for a service in a given market, an extremely high rate that is largely unconstrained by any market forces (See Table 1 and Figure 2).
A provider’s billed charge is a list price that he or she sets unilaterally. Predictably, then, the H.R. 3502 approach is favored by provider trade groups as it would likely result in large revenue increases for emergency and ancillary services, paid for by commercially-insured patients and taxpayers. As a result, H.R. 3502 would almost assuredly result in significantly higher premiums, patient costs, and federal deficits. And this marked increase would come on top of the already inflated premiums people are paying due to today’s very high private insurer payments to emergency and ancillary clinicians that likely stem from the ability to leverage the threat of surprise billing patients.
By contrast, the compromise approach originally proposed by the House Energy and Commerce Committee and recently approved by the Health, Education, Labor, and Pensions (HELP) Committee in the Senate would set payment for surprise out-of-network bills at the relevant median in-network payment rate under that insurance plan. This policy is estimated to reduce commercial insurance premiums by 1 percent on average nationwide and decrease deficits by $25 billion over ten years, according to the Congressional Budget Office (CBO).
Eliminating surprise out-of-network billing is a laudable goal, but a solution shouldn’t simultaneously create another problem by inflating costs and shifting them to purchasers of health insurance (employers and consumers) and taxpayers.
The rest of this blog provides more detail on the legislation’s payment mechanism and why it would be so inflationary.
The Value of The Out-Of-Network Option
Surprise out-of-network billing is not just an occasional glitch; it results from a systemic market failure. Patients typically choose which hospital they go to and which primary physician treats them in non-emergency settings, but they do not have choice over the anesthesiologist, radiologist, or pathologist whose services are critical to their care. Yet these physicians and the hospitals they work at contract independently with health plans, leaving open the likelihood of discordant network status, where the hospital might be in-network while the ancillary physician is out-of-network with the patient’s health plan. A similar dynamic occurs for emergency services, for which patients typically lack choice over the physicians who treat them and many times are unable even to choose which hospital to go to, given the nature of emergency care.
Emergency and ancillary physicians, therefore, receive a flow of patients regardless of whether they contract with a health plan. Unlike their peer physicians, then, these facility-based specialists have an out-of-network billing option that provides leverage in negotiations with insurers, as they can credibly threaten to balance bill a health plan’s enrollees unless they are paid enough to forego this option. As a result, emergency and ancillary physicians charge much higher amounts than their peer physicians relative to what Medicare pays for those same physician services. Further, there is a non-trivial minority of these specialists with especially exorbitant charges, who appear to be most aggressively leveraging the threat of surprise billing patients.
Table 1: Ratio of Charges to Medicare Rates by Physician Type, 2016
The value of this out-of-network option is important not just for the size of surprise balance bills received by patients, but in determining payment amounts for in-network services – this is why surprise billing legislation can have significant impacts on insurance premiums and, in turn, federal deficits (because employer-sponsored insurance premiums are exempt from taxation). It also likely explains why emergency and ancillary physicians negotiate in-network payment rates from private insurers at much higher multiples of Medicare rates, on average, than their peer physicians (See Figure 1).
Figure 1: Average Contracted Payment Relative To Medicare Rates For Selected Specialties
Source: Stead and Merrick 2018; Trish, Ginsburg, Gascue, and Joyce 2017; MedPAC 2017
Note: Anesthesiologist comparison based on relative mean conversion factors in 2018. Emergency physician comparison based on relative mean payment rates for CPT code 99285 in 2012. For radiologists, 200% represents mean commerical payment for CT Head/Brain scans relative to the Medicare rate (CPT code 70450). All physicians comparison based on data from commercial PPO claims for one large national insurer.
Similarly, the out-of-network option helps explain why physician staffing companies such as Envision and TeamHealth that appear to more aggressively threaten to surprise bill patients negotiate in-network payment rates notably above even the high specialty averages – as high as five times Medicare rates according to a paper analyzing a large national insurer’s data from 2012-15.
How Insurer Payment For Surprise Bills Is Determined Under the Proposal
H.R. 3502 would set up an arbitration process to determine how much a health plan must pay to an out-of-network provider for services subject to the law, if the two parties are unable to come to an agreement on their own. This process would follow what is termed “final-offer” or “baseball-style” arbitration rules, in which the insurer and out-of-network provider make a final offer for the rate for a given service, and the arbiter must select one or the other as the winning amount. The theory behind the approach is that the possibility of the arbiter choosing the other party’s bid incentivizes each side to submit a reasonable offer.
Arbitration appears to have some political appeal because it has the feel of being less prescriptive, but in reality it mainly just punts the out-of-network rate setting decision to arbiters, which makes the process less transparent and adds administrative costs to the system – $1 billion, according to the leaked CBO estimate. There is no magic to the arbitration process and there is no reason to think that arbiters are better equipped than legislators or regulators to make payment rate determinations. Indeed, what ultimately matters is the guidance arbiters are given on how they should choose between competing bids.
Therefore, the most worrying piece of Ruiz’s proposed legislation is its explicit guidance that arbiters should consider “the 80th percentile of charges for comparable items and services for the specialty involved in the geographical area in which the item or service was furnished” when choosing between bids. Providers’ full charges, or list prices, tend to be exceptionally high because they are generally unconstrained by normal market forces. As discussed in greater detail in the next section, directing providers to consider the 80th percentile of charges could yield significant increases in commercial insurance premiums, patient costs, and federal deficits. Although the legislation also includes “commercially reasonable rates” as a factor for consideration in arbitration, that term is left undefined and so there is significant risk that arbiters will default to the charge-based benchmark.
Referencing The 80th Percentile Of Charges Would Increase Spending
Under H.R. 3502, if arbiters indeed set the 80th percentile of charges as the rule of thumb and hence the de facto out-of-network rate setting amount, emergency and ancillary clinicians would see a large increase in the value of their out-of-network billing option. Today, while going out-of-network can be lucrative for these specialists, it also comes with significant costs: the time and money spent collecting from patients; the need to compensate the hospital they work at for; the reputational harm from allowing surprise billing; and most physicians find it distasteful to surprise bill patients and would rather not do so. But by mandating extremely high health plan payments and requiring the insurer to treat the service as in-network for purposes of patient cost-sharing, H.R. 3502 would eliminate these obstacles to going out-of-network. Instead of receiving large surprise bills, the high costs of payment at the 80thpercentile of charges would be spread among all enrollees in a patient’s health plan through higher premiums.
Not only is the 80th percentile of charges in a market higher than the list price that four-fifths of clinicians charge for a given service, but this new law would require that the patient’s health plan pay that full amount (minus standard in-network patient cost-sharing amounts). This differs notably from today, where anecdotally it appears that emergency physicians are only able to collect something on the order of 40 percent of total dollar value of out-of-network balance bills. Given that one would expect diminishing marginal collections from increasing charges even further under today’s system, it seems very likely that emergency or ancillary physicians would collect substantially more under the Ruiz bill than they do now out-of-network.
Facility-based emergency, ancillary, and similar clinicians would have every reason to refuse all network contracts that do not to pay them something near the 80th percentile of charges. The choice would be quite simple for many providers: stay in-network at current contracted rates or go out-of-network and automatically collect the 80th percentile of charges, which can be several times higher than average in-network rates.
Figure 2: Median In-Network Rates Vs. 80th Percentile Of Charges
Source: Fair Health, 2018-19. Data are for New York state.
Moreover, because provider charges face little market restraint, specialists affected by the law would have strong incentives to continue increasing their charges over time, especially staffing companies or physician groups with large market share who can more easily affect the 80thpercentile in a market. (For example, a physician group providing more than 20 percent of a certain service in a market can unilaterally shift the 80th percentile of charges in their market higher and higher.)
Without much constraint on either charges or contracted rates, emergency and ancillary physicians, as well as other providers that patients don’t choose and facility rates for emergency services, could see their per service payments for commercially-insured patients increase several-fold as a result of H.R. 3502.
Evidence From New York
Advocates for Ruiz’s approach often point to a working paper from Yale researchers (Cooper, Scott Morton, and Shekita) estimating that a similar law in New York reduced in-network payment rates for emergency physicians, as evidence that their legislation will not lead to large increases in health care costs. However, that study is not as informative for predicting the effects of federal legislation as advocates make it out to be.
First and foremost, the study only analyzes data from one insurer for one year post-enactment of New York’s law. The law’s full effect is not likely to be felt that quickly because, importantly, the 80th percentile of charges is only one of many factors that New York provides as arbitration guidance, and thus it likely was unclear to providers and insurers at the outset that this would subsequently become (as it has) arbiters’ rule of thumb for determining the winner. Stakeholder conversations we’ve had reinforce that many provider groups and insurers did not initially understand that arbiters would generally select whichever offer is closer to the 80thpercentile of charges. This uncertainty, however, should dissipate over time. Indeed, the number of cases going to arbitration in New York has increased markedly in recent years, which is both the opposite of the intended effect of the binding arbitration model and potential evidence that providers are catching on that the arbiters’ rule of thumb is favorable to them (See Figure 3).
Figure 3: Summary of NY Arbitration Cases
Notes: “Total IDR cases” represents the total number of eligible bills for which a decision was rendered or are still in process of being adjudicated. 336 cases in 2018 were still in process according to the most recently published data as of 10/25/2018.
Source: NY Department of Financial Services
There are also a number of important factors specific to the New York law that blunt its expected inflationary impact compared to a similar federal proposal, like H.R. 3502.
- Self-insured health plans, which comprise roughly 60 percent of commercial insurance enrollment, were not subject to New York’s law. Not only does this substantially reduce the portion of the market affected, but it makes it much less likely that the law would prompt providers to decline insurance participation for the market segments that are covered, since doing so would still require them to bill and collect from patients directly for the majority of the market not covered by the law.
- New York exempts balance bills for emergency services under $600 (plus inflation since 2014).
- Health maintenance organizations (HMOs) in New York were previously already subject to a hold harmless provision for surprise out-of-network bills, which further blunts the added effect of New York’s newer arbitration law.
Therefore, given these considerations and that the Ruiz legislation dramatically improves the out-of-network billing option for affected providers, the preliminary evidence on the effects of the New York law does little to assuage concerns that H.R. 3502 would result in a large increase in spending on emergency and ancillary services – and in turn premiums, patient costs, and federal deficits – at least over the longer run.
Additional resources on surprise billing are available here.
The Initiative is a partnership between the Economic Studies program at Brookings and the USC Schaeffer Center for Health Policy & Economics, and aims to inform the national health care debate with rigorous, evidence-based analysis leading to practical recommendations using the collaborative strengths of USC and Brookings.