There is no shortage of horrific anecdotes about surprise medical bills—cases where patients are unexpectedly billed at highly inflated prices by providers who do not accept their insurance. With public opinion firmly in support, Congress is poised to enact legislation to address this issue. In this article, we argue that a commonly suggested arbitration‐​based system, such as the one endorsed by Ike Brannon and David Kemp (see p. 40), is not the best option for solving this very real problem.

The problem of surprise medical bills is not limited to a few shocking anecdotes. Studies show that 14–20% of emergency department (ED) visits may result in a surprise bill, with that number exceeding 40% in some states. (Recent research using data from Oputm finds even higher rates of surprise billing.) Data from a national insurer show that, at the median hospital, only 1% of ED visits at in‐​network hospitals generated out‐​of‐​network bills. At just 15% of hospitals, however, more than 80% of ED visits generated a similar bill. Even among elective admissions—where patients presumably have control over where they receive care—nearly 10% of patients are at risk of receiving a surprise bill and being “balance‐​billed” when their insurer refuses to cover it. Some sectors of health care (such as air ambulances and at least one ED staffing company) appear to be built on a business model of sending surprise bills.

Physicians who are not chosen directly by patients exploit these dynamics to set artificially high prices for their services. These providers will only join an insurer’s network if the amount they receive is worth more than the right to balance‐​bill patients by remaining out‐​of‐​network. A recent report shows that the physicians least likely to be chosen by a patient (i.e., ED physicians, anesthesiologists, and radiologists) set their list prices roughly twice as high as similar physicians who are more likely to be chosen by patients. This strategy allows these providers to make more when they choose to be in‐​network, and increase the size of any resulting balance bill when they are out‐​of‐​network.

The public is understandably concerned about the problem. In a poll conducted by the Kaiser Family Foundation in August 2018, respondents listed “unexpected medical bills” as their top concern—ranking well ahead of prescription drug costs and health insurance premiums. In response, state policymakers have enacted various strategies for dealing with the problem of surprise medical bills. Congress is poised to follow them. The list of policy options is mercifully short: arbitration, contract‐​based strategies, and rate‐​setting.

The IDR idea / Brannon and Kemp argue that arbitration (or an Independent Dispute Resolution [IDR]) represents the best solution because it “keeps the price‐​setting ability in the hands of the market participants … without creating counterproductive unintended consequences.” They also highlight the value of IDR in appropriately balancing negotiations between market actors: an in‐​network guarantee or benchmarking would “lead to an imbalance in the negotiations between insurers and providers. Biasing the system toward either side leads to unintended negative consequences, such as increased insurance premiums or reduced access to care.”

We agree with Brannon and Kemp that the problem of surprise medical billing is best solved by requiring market actors to privately negotiate market prices. But an IDR does not accomplish this, nor does it prevent surprise medical bills. Instead, arbitration is an ex post dispute resolution system that represents a non‐​transparent version of the price‐​setting approach that Brannon and Kemp properly reject. Arbiters, not market actors, ultimately determine payment rates. In addition, Brannon and Kemp’s suggestion that reform should strive to balance incentives and not disrupt the status quo is inapt. Current bargaining dynamics are responsible for the problem of surprise medical billing. Reform should fix these problems by changing everyone’s incentives, rather than reward those who have engaged in strategic behavior in the past. Finally, although Brannon and Kemp suggest that narrow networks are responsible for the rise of surprise billing, the evidence indicates otherwise. We examine each of these issues below.

We begin by briefly reviewing the circumstances that can give rise to surprise bills. Providers decide whether to contract with a particular insurer (i.e., to be in‐​network) or not (i.e., to be out‐​of‐​network). Providers who are in‐​network agree to accept a contracted price, while out‐​of‐​network providers set their own prices and can seek to collect from the patient the difference between the total bill and the amount paid by the insurer.

Patients who knowingly choose to receive care from an out‐​of‐​network provider have no basis for complaint when they receive a balance bill. But in many situations, patients reasonably believe their physician is in‐​network or they have no control over whether a provider is in‐​network or out‐​of‐​network. Thus, surprise medical bills are likely to arise in three distinct settings:

  • Patient receives care at an ED in an in‐​network hospital, but one or more clinicians involved in the treatment (e.g., emergency medicine physicians, ancillary physicians, or other specialists working in the ED) are out‐​of‐​network.
  • Patient receives elective care at an in‐​network hospital, but an ancillary physician (e.g., an anesthesiologist) is out‐​of‐​network.
  • Patient is taken to an out‐​of‐​network facility in an emergency or is transported by an out‐​of‐​network ambulance.

So how should we address this problem? Brannon and Kemp suggest that IDR will address these dynamics and that the other two available solutions (contract‐​based strategies and rate‐​setting) are far inferior. We respectfully disagree.

Missed opportunity / The first problem with IDR is that it does not prevent surprise medical bills from being sent. Instead, it provides an after‐​the‐​fact mechanism for resolving disputes. This represents a missed opportunity. Second, the arbiter must ultimately set a price. Arbiters are generally instructed to pick a “reasonable” price, often choosing between prices proposed by the insurer and the provider. But arbiters must still develop a decision rule to decide these cases. Enacted and proposed legislation typically specify some parameters for determining a “reasonable” price (e.g., the in‐​network median payment rate, the 80th percentile of billed charges in New York). This process turns arbiters into implicit rate setters who conduct their work in a completely non‐​transparent way. For example, New York’s IDR system was introduced about five years ago and we have little evidence on the size of rulings.

Current bargaining dynamics are responsible for surprise billing. Reform should fix these problems by changing everyone’s incentives, rather than reward those who have engaged in strategic behavior in the past.

The lack of transparency associated with IDR is particularly worrisome because a number of proposed or implemented arbitration guidelines are based on billed charges. For example, as noted above, New York state law instructs arbiters to consider the 80th percentile of charges in an area. Ongoing work confirms that arbiters are following this standard in deciding the disputes that come before them. Because providers can set their charges as high as they see fit and IDR ensures they will be paid at that level as long as they are out‐​of‐​network, it is not surprising that most observers expect New York’s IDR‐​based approach will result in higher health care costs. Indeed, Brannon and Kemp cite a Georgetown University report in which providers freely admit that, as a result of New York’s IDR‐​based approach, they now receive “higher reimbursements to be in‐​network than they had prior to the law.” This problem is not limited to New York; Alaska has a similar provision, and a similar charge‐​based bill pending in the U.S. House of Representatives (the Protecting People From Surprise Medical Bills Act) has 71 co‐​sponsors.

We also respectfully disagree with Brannon and Kemp’s focus on the importance of ensuring balanced negotiations. Policymakers should focus on correcting the market failures that give rise to surprise medical bills. Some providers face effectively no tradeoff between prices and volume of their services. If patients cannot avoid certain providers, those providers can charge prices that dramatically exceed market rates without any adverse consequences. The data suggest that some (but certainly not all) providers do just that.

This emphasis on striking the appropriate balance appears to partially motivate Brannon and Kemp’s suggestion that narrow insurance networks are a substantial contributor to the problem of surprise medical bills. This argument is intuitively plausible, but the data indicate network breadth is unlikely to be driving this phenomenon. A recent study found similar rates of surprise billing among those with employer‐​sponsored insurance (19%) as those with marketplace plans, which generally have much narrower networks (22%). In addition, rates of surprise medical bills are similar across many types of insurance (HMOs, PPOs, and HDHPs), even though HMOs have much narrower networks. Other studies indicate that surprise bills are most common when patients do not control which provider they see—consistent with the three scenarios outlined above. Taken together, these findings suggest that narrow networks are unlikely to be driving the majority of surprise billing.

Contract‐​based alternative / So what should we do instead? In our view, a contract‐​based solution, which would require all providers at an in‐​network hospital to either contract with the same insurers as the hospital or secure payment from the hospital (who will bundle those costs as part of their in‐​network facility fee), will outperform IDR. A contract‐​based approach entirely eliminates the sending of surprise medical bills at in‐​network settings and puts the burden of negotiating market prices on those closest to the situation. A contract‐​based approach requires nothing from the vast majority of providers that do not engage in surprise billing, and it eliminates the need for policymakers to impute a market price or create and fund a dispute resolution system to do the same.

In fairness, Brannon and Kemp have a legitimate concern that a contract‐​based solution may allow insurers “to further reduce their pre‐​negotiated rates, possibly to prices below the actual costs of services.” But this supposed problem will sort itself out when the surprise bills originate from in‐​network facilities. If insurers insist on rates that are below the cost of services, physicians will look to hospitals to make up the difference—and hospitals will build that amount into their negotiations with insurers over facility fees. In short order, we would arrive at a natural market rate that did not reflect the ability of some providers to send surprise medical bills. Without that ability, we should expect rates to be lower than the status quo, albeit not below the cost of providing the services in question.

To be sure, a contract‐​based approach will not work in situations where patients are taken to an out‐​of‐​network facility in an emergency or are transported by an out‐​of‐​network ambulance to an in‐​network facility. We note that the legal system has developed strategies for handling such circumstances. Under admiralty law, courts will not enforce a bill for marine salvage that exceeds the market value for the services in question. Knowing this, everyone uses a standard form contract and disputes over billing are uncommon. What does it say about the medical profession that its billing practices would not pass muster if brought before a court handling a dispute over marine salvage?

Conclusion

Out‐​of‐​network balance bills are unique to health care. When you take your car to a body shop, the painter who repaints the door panel does not send you an inflated, separate bill and then balance‐​bill you when your insurance refuses to pay it in full. This is not because we have an elaborate arbitration system to adjudicate door panel repair bills; it is because the market demands all‐​in pricing.

In health care, normal market forces have failed to prevent surprise medical bills. Although a well‐​designed IDR system can help resolve such disputes, design details matter greatly in how effective this approach will be in arriving at market prices. A contract‐​based approach is likely to outperform IDR.

Readings

  • “Assessment of Out‐​of‐​Network Billing for Privately Insured Patients Receiving Care in In‐​Network Hospitals,” by Eric C. Sun, Michelle M. Mello, Jasmin Moshfegh, and Lawrence C. Baker. JAMA Internal Medicine (online), August 12, 2019.
  • “Consumers’ Responses to Surprise Medical Bills in Elective Situations,” by Benjamin Chartock, Christopher Garmon, and Sarah Schutz. Health Affairs 38(3): 425–430 (2019).
  • “New York’s 2014 Law to Protect Consumers from Surprise Out‐​of‐​Network Bills Mostly Working as Intended: Results of a Case Study,” by Sabrina Corlette and Olivia Hoppe. Robert Wood Johnson Foundation, 2019.
  • “One in Five Inpatient Emergency Department Cases May Lead to Surprise Bills,” by Christopher Garmon and Benjamin Chartock. Health Affairs 36(1): 177–181 (2017).
  • “Solving Surprise Medical Billing,” by Benedic Ippolito and David A. Hyman. AEI Economic Perspectives, 2019.
  • “State Approaches to Mitigating Surprise Out‐​of‐​Network Billing,” by Loren Adler, Matthew Fielder, Paul B. Ginsburg, et al. USC–Brookings Schaeffer Initiative for Health Policy white paper, 2019.
  • “Surprise! Out‐​of‐​Network Billing for Emergency Care in the United States,” by Zach Cooper, Fiona Scott Morton, and Nathan Shekita. National Bureau of Economic Research Working Paper no. 23623, 2017.