RESEARCH DOSSIER
Health Insurance Surprise Billing and Cost-Sharing Protections
Abstract. Surprise medical billing arises when the financial relationship among a patient, health insurer, medical facility, and treating clinician produces liability that the patient could not reasonably identify or control before receiving care. The problem is acute in emergencies, in facilities where participating and nonparticipating clinicians work together, and in complex episodes in which transport, professional services, and institutional care are billed separately. This monograph develops an international, source-critical account of the legal, economic, actuarial, and administrative mechanisms through which health insurance systems constrain these liabilities. It distinguishes provider balance billing from ordinary contractual cost sharing, billed charges from realized payments, provider network participation from the network status of a facility, and consumer protection from the rules governing payment between health plans and providers. The United States provides the principal empirical and regulatory case because its No Surprises Act, effective for relevant services from 2022, established a detailed combination of patient liability restrictions, statutory payment sequencing, and federal independent dispute resolution. The analysis integrates the underlying statute, administrative rules and guidance, the 2026 amendments to dispute operations, a 2026 appellate ruling concerning qualifying payment amount methodology, the Government Accountability Office's longitudinal review, public administrative dispute counts, federal enforcement records, and original claims-based economic studies. Additional comparisons examine contracted care in the Netherlands, Swiss compulsory health insurance, European cross-border reimbursement, the English National Health Service, and Australian private insurance disclosure duties. These systems differ in financing and legal architecture, so their instruments are compared by the financial exposure they regulate rather than treated as interchangeable legal solutions. The evidence establishes three analytically linked propositions. First, surprise billing is generated by incomplete patient control over the selection of charge-generating providers and by fragmentation of contracting and payment authority. Second, a legal cap on the patient's payment can substantially change the distribution of immediate liability while leaving the insurer–provider price and the longer-run premium effects to be determined by bargaining, arbitration, market structure, and enforcement. Third, administrative records of provider payment disputes, claims with an out-of-network indicator, patient invoices, actual collections, and household financial hardship measure different outcomes and require separate estimands. Original analytical contribution consists of the integrated mechanism reconstruction, measurement distinctions, and comparative interpretation, without proposing a new named theory, proprietary doctrine, or newly estimated causal effect.
Keywords: health insurance; surprise medical billing; balance billing; out-of-network care; No Surprises Act; cost sharing; emergency services; air ambulance; provider networks; qualifying payment amount; independent dispute resolution; insurer–provider bargaining; claims administration; financial protection; regulatory enforcement; comparative health law; consumer protection.
Introduction
A covered hospital visit can generate several independent invoices. A patient may choose an in-network hospital, be treated by an emergency physician assigned by the facility, receive anesthesia from another professional group, undergo imaging interpreted by a separate radiology practice, and be transported by an ambulance operated independently of the hospital. Each professional or organization may have a distinct insurance contract and billing right. The patient cannot reliably infer the contractual status of the whole episode from the network status of its most visible institution. The legal and economic problem begins at that point of fragmentation.
A surprise bill is a specific instance of uncertain consumer liability. Federal United States law and agency explanations describe it in relation to nonparticipating providers and unexpected excess charges; the No Surprises Act statute was enacted within the Consolidated Appropriations Act, 2021, and its main protections took effect in January 2022. The operative protections concern specified covered services, plans, facilities, and providers, rather than every medical charge. The CMS consumer rights account describes emergency services, certain nonemergency services at participating facilities, and covered air ambulance services, while identifying ground ambulance services as outside the federal surprise-billing prohibition. This legal specificity matters to the classification of every claim and to any empirical estimate of the law's effects.
Surprise billing has at least four economically distinct dimensions. The first is the contractual network relationship between payer and provider. The second is the information available to a patient before treatment. The third is the provider's legal ability to demand an amount exceeding ordinary in-network patient cost sharing. The fourth is the payment mechanism deciding the final amount owed by the insurer or plan. A rule that protects the patient on the third dimension may also affect bargaining on the fourth, yet that consequence cannot be read mechanically from the consumer's protection. In a system financed by premiums, taxes, or employer contributions, reallocating a payment obligation may shift costs among participants over time.
This monograph asks three central questions. Which combinations of network status, emergency circumstances, and provider conduct create an otherwise preventable out-of-network liability? How do laws and payment resolution systems divide the resulting obligation among patients, insurers, and providers, and what incentives follow? Which empirical measures can credibly establish the welfare, distributional, and insurance-market consequences of those arrangements across jurisdictions and periods?
The approach is documentary, comparative, and analytical. It uses the original statute, current administrative sources, federal reporting, peer-reviewed claims-based studies, and official comparative health-insurance materials. Historical incidence studies show the scale of the pre-reform exposure, while administrative datasets describe the implementation environment. A causal effect on consumer welfare would require a credible counterfactual, consistent observation of patient payments, and controls for changes in service mix, insurance enrollment, physician organization, and regulatory coverage. This dossier supplies a reproducible evaluation design and assesses published evidence; it has not obtained individual claims records or independently reestimated published regression coefficients.
The regulatory cutoff is November 16, 2026. The August 2026 TMA III notice documents an appellate development affecting some qualifying payment amount rules and confirms that the federal resolution system continued to operate. The same official notices document the start of batching changes in November 2026. A sound legal description must retain its as-of date, distinguish statutory protections from implementing calculations, and refrain from treating a pending or evolving implementation question as a settled universal rule.
International comparison is necessary because the United States' combination of private network bargaining and statutory out-of-network arbitration is only one way to organize financial protection. WHO's universal health coverage framework treats access to needed services without financial hardship as the central objective. It offers a common evaluative outcome while allowing different institutional pathways. For this dossier the central comparative question is whether an individual has predictable and enforceable exposure when care must be obtained under constraints that limit meaningful choice.
Foundations and Source-Critical History
The object of surprise billing
Health insurance cost sharing ordinarily consists of a deductible, copayment, coinsurance, or another clearly specified contribution by the insured toward a covered claim. Balance billing involves a provider seeking the difference between its charge and amounts recognized or paid through the plan and patient obligation. In a typical protected surprise-billing event, the provider seeks additional payment because it lacks an agreement with the patient's plan, even though the patient entered an otherwise covered episode expecting ordinary network protection. The federal consumer-protection explanation distinguishes an unexpectedly imposed balance from the cost sharing that remains payable under the person's plan.
The economic question is not determined by whether a number on a medical bill appears large. A high deductible can generate substantial but contractually anticipated liability. An out-of-network physician can generate a small prohibited balance bill. A service excluded from a contract may involve a major expense without fitting the specific legal definition of protected surprise billing. Likewise a medical bill can exceed the actual cash amount eventually collected. Regulatory classification therefore requires separate variables for coverage, network status, permitted cost sharing, billed balance, attempted collection, dispute status, and final payment.
Three obligations should be distinguished throughout. The medical provider's posted or submitted charge is an asserted price. The payer's allowed, adjudicated, negotiated, or statutorily determined amount is the basis for allocation of payment under a relevant rule. The consumer's enforceable liability is a legally and contractually constrained amount. These quantities may coincide in simple cases, but the analytical value of the distinction arises precisely when they differ.
Consider an illustrative episode in which a physician submits a $2,000 charge, the applicable amount recognized for calculating permitted consumer sharing is $600, and the patient's plan imposes 20 percent coinsurance after the relevant deductible has been met. In a case to which the described rule applies, $120 could be the patient share while the $1,400 difference between the charged amount and the recognized base is not automatically collectible from the patient. The final payment to the provider may be established separately by the plan's initial payment, agreement, or an applicable payment determination. These are constructed numbers, not observed claims or advice for a particular bill. The precise cost-sharing base and even the applicable rule depend on the statutory category, plan terms, jurisdiction, and current legal framework.
Insurance contracts, medical production, and payment fragmentation
Health care production is modular. Hospitals buy or contract for professional labor and diagnostic services while the insurance market often prices and contracts with those professionals separately. From the standpoint of an emergency patient, however, these services form a single clinical episode. The mismatch between clinical indivisibility and contractual separability is foundational to the exposure.
The underlying bargaining problem was described by Cooper, Scott Morton, and Shekita (2020) in their emergency billing study. They showed how an emergency physician may serve patients who cannot choose among professionals, giving that group a bargaining option outside ordinary network contracting. Their study of New York's arbitration intervention identified a change in the economic returns to staying out of network. The published peer-reviewed analysis and an earlier working-paper version report different estimates; this dossier uses the peer-reviewed publication when describing its principal result and treats working-paper figures as historical versions of the research rather than interchangeable estimates.
The phenomenon also arises in nonemergency hospital episodes. Cooper, Nguyen, Shekita, and Scott Morton (2020), studying a large insurer's 2015 claims, documented nonparticipating bills for anesthesiology, pathology, radiology, and assistant-surgeon services at participating hospitals. Their hospital-based physician study provides strong evidence that an insured patient's hospital choice does not fully determine provider participation. The published estimates refer to particular services and payer records, rather than an all-payer national rate.
A useful conceptual distinction follows from the evidence. An insurer–provider contract is negotiated ex ante, often for a population of prospective patients. An individual emergency encounter is allocated ex post, sometimes without patient selection of the professional. The physician's revenue from unexpected out-of-network encounters can affect the threat point in the ex ante negotiation. A surprise-billing prohibition changes that threat point and creates a policy channel through which consumer protections can influence provider contract prices. The direction and magnitude of resulting changes depend on arbitration design, market concentration, negotiation behavior, and the insurer's payment obligations.
Historical evidence and the shift to statutory protection
Pre-2022 research documented the practical scale of the risk. Garmon and Chartock (2017) estimated that in 2014 approximately 20 percent of inpatient admissions originating in an emergency department, 14 percent of outpatient emergency department visits, and 9 percent of elective inpatient admissions were likely to lead to surprise bills. The underlying Health Affairs study measured likely exposure, not an independently verified rate of successfully collected excess payments.
Sun, Mello, Moshfegh, and Baker (2019) evaluated 5,457,981 inpatient admissions and 13,579,006 emergency visits at in-network hospitals in a large commercial claims dataset from 2010–2016. Their claims-based analysis found that the share of emergency visits associated with an out-of-network bill increased from 32.3 percent to 42.8 percent; mean potential liability rose from $220 to $628 in 2018 dollars. For inpatient admissions the corresponding shares increased from 26.3 percent to 42.0 percent, while mean potential liability increased from $804 to $2,040. These are historical, study-specific measures of potential financial exposure, not a claim that these amounts were universally collected or remain current in 2026.
The published literature employs different denominators. A hospital admission, an emergency episode, an individual professional claim, and a member-month are not equivalent units. The presence of one nonparticipating service among several claims for an episode inflates episode-level exposure relative to individual-service rates, all else equal. Differences between a single insurer's data and a multi-insurer dataset can reflect enrollee geography, contracting, plan design, and selection. Apparent discrepancies between studies therefore require reconciliation of the underlying definition before they become substantive disagreements.
The statutory response combined patient protection, provider notice requirements, payment calculation rules, and a dispute channel. The 2021 implementing materials make clear that patient protection and administrative administration are related parts of the same intervention. In 2026 the system continues to evolve through appellate litigation, revised procedural rules, operational data publication, and enforcement. Historical baseline observations can establish that the problem existed and may reveal plausible policy mechanisms; on their own they cannot establish the causal impact of the later national law.
Research design, evidence hierarchy, and units of analysis
The evidentiary hierarchy is defined by the proposition under review. Statutory scope is established from enacted law and binding legal instruments; practical implementation is established from current regulator notices and procedural sources, including applicability dates; claims incidence is established from methods and definitions of the original empirical study; recorded disputes are established from administrative records; and financial hardship is established from observed household expenditures or validated financial surveys. Expert interpretation is weighed against those underlying documents rather than substituted for them.
Legal propositions are indexed by jurisdiction, payer type, service, encounter date, claim date, and regulatory status. Economic propositions are indexed by the population, observed unit, payer arrangement, market definition, and estimand. For legal and economic comparison, those indexes must remain connected: a claim from 2020 cannot be subjected retrospectively to a 2026 operating rule, and an insurer–provider arbitration outcome cannot be relabeled as a patient's balance bill.
The necessary research archive includes source identification, access conditions, methodological limitations, version history, and the institutional location of the record. The methodological object is a linked sequence from encounter to claim adjudication, enforceable patient obligation, provider payment, and final collection. This sequence provides the foundation for the remaining chapters and for an auditable reconstruction of financial protection.
Network Status and Emergency Care
Network participation as a multidimensional attribute
A provider network is a set of contract-specific rights and obligations governing payment, access, and administrative treatment. A facility's participation tells a patient comparatively little about the contractual status of each independent clinician involved in care. A single emergency episode may involve a participating hospital, a nonparticipating emergency physician, a separately contracted laboratory, and an out-of-network transport service. Research must represent these positions at the relevant service, billing entity, payer, and date rather than assign a single episode-wide indicator merely because the patient entered an in-network building.
Network status is often time varying. Professional groups can enter and leave contracts; leased networks and third-party administrators may administer a product differently; provider mergers or group changes can cause claim routing to diverge from a directory; and a health plan may have multiple products with different networks. The federal provider requirements accordingly have to be read together with plan participation, facility identity, and statutory service category. A patient who is told that a hospital “accepts” an insurer may reasonably interpret that statement as comprehensive financial protection, even where contractual documents give it a narrower meaning. That information asymmetry is one of the mechanisms that surprise-billing protection addresses.
Formally, let network participation for provider p, plan j, service s, and treatment date t be N(p,j,s,t), with N equal to one when an applicable participation agreement covers that claim and zero otherwise. A claim-level indicator can be constructed by joining professional billing records to the contract table through the appropriate organization identifiers and effective dates. The classification should retain an unresolved category when evidence is insufficient; treating missing identifiers as proof of nonparticipation biases incidence estimates. Similarly, a participating facility should be identified separately as F(f,j,t). An episode with F=1 and any N=0 may be at risk of surprise bills, but the actual legal protection also depends on the service, exceptions, the type of plan, and any valid notice-and-consent procedure.
This distinction improves interpretation of claims research. Cooper and colleagues' hospital-based physician findings measured specific professional claims, revealing out-of-network rates at in-network hospitals, but their sample does not identify every form of patient liability or every nonhospital encounter. A complete study of consumer risk requires linking institutional and professional claims to enrollment, provider relationships, legal eligibility, and collection outcomes. Administrative aggregation without these links risks counting a contractual dispute as a patient debt or a clinical episode twice.
Emergency services and post-stabilization care
Emergency care is the central case of constrained consumer choice. An individual in an emergency cannot be expected to negotiate network contracts or comparative professional fees. U.S. federal protection applies, for most covered private insurance arrangements, to emergency services delivered by nonparticipating emergency facilities or providers. The CMS explanation of insured patient rights identifies a patient's right to in-network-level cost sharing for covered emergency treatment even when the provider or facility lacks an ordinary participation agreement. The key protection concerns patient liability; the provider's final payment is determined by a separate legal or contractual process.
A law's emergency classification must be distinguished from retrospective claims coding. Regulatory definitions can rely on symptoms and the perspective of a prudent layperson, whereas utilization datasets commonly classify encounters from diagnosis, place of service, or facility billing codes. A patient presenting with potentially life-threatening symptoms may have a less alarming final diagnosis. Studies that retrospectively remove such episodes from the protected population can distort both denominator and risk interpretation. Emergency departments also deliver diagnostic and stabilization services before anyone can know whether admission will be necessary.
Post-stabilization services present a particularly important boundary. Once an emergency medical condition has stabilized, the necessity of ongoing care, availability of alternative participating providers, transport feasibility, and statutory consent conditions may become relevant. The No Surprises Act consumer summary specifies the conditions under which some post-stabilization services can fall outside the emergency protections. An empirical audit should document the clinical timeline and relevant notice-and-consent evidence, rather than infer consent from a routine signature or a generic financial-responsibility form. In the emergency phase, the consumer generally cannot waive the statutory protections by signing a form.
The conventional phrase “in-network cost sharing” requires precision. For a protected claim, a person's responsibility is calculated according to the applicable regulatory framework, and qualifying patient payments generally count toward the corresponding in-network deductible and out-of-pocket maximum. Those maximums limit specified covered in-network cost sharing under the plan; they do not erase premiums, excluded services, or every expense associated with obtaining care. The choice of the recognized amount for determining coinsurance is legally significant because an identical percentage can produce different dollar liability when applied to different bases.
Non-emergency facility-based treatment
Surprise billing also occurs when an insured patient schedules care at a participating hospital or ambulatory surgical center but encounters an independently contracted nonparticipating clinician. Such episodes contain a stronger appearance of consumer choice than emergency care, yet the relevant clinician may still be assigned without meaningful advance choice. Anesthesia, radiology, pathology, laboratory medicine, hospitalist services, and assistant surgery illustrate the distinction between choosing a facility and choosing every professional.
The U.S. key consumer protections extend to specified nonemergency services furnished by nonparticipating providers at certain participating facilities. The rule has a carefully delimited notice-and-consent exception, and certain ancillary services are excluded from the exception so that a provider cannot simply obtain a purported waiver. These exclusions reflect the limited substitutability and control that patients often have over the professional furnishing an ancillary service. The precise covered facilities and professional categories must be drawn from the governing instruments and current guidance rather than extrapolated to all outpatient locations.
The financial exposure is also shaped by facility contracting practices. Insurers and hospitals can reduce patients' uncertainty when contracts address the participation of clinical groups operating within the facility or when referral pathways include advance network checks. But a facility's promise alone cannot be assumed to create enforceable insurer participation for independent physicians. A complete intervention may therefore combine direct statutory limitations, contractual obligations for ancillary professional groups, accurate directories, patient-facing disclosures, claims processing, and enforcement.
A careful study distinguishes inadvertent out-of-network participation from a strategically chosen contracting position. Nonparticipation may reflect administrative delay, product misclassification, a disagreement over fees, or a deliberate bargaining strategy. Claims alone reveal the relationship's outcome but not the underlying intent. Interviews and contract documents can help reconstruct the mechanism; assertions about a provider's motives remain interpretive unless supported by relevant evidence.
Air and ground ambulance transport
Ambulance transport illustrates the importance of legal categorization. Federal U.S. surprise-billing protections extend to covered nonparticipating air ambulance services under the applicable rules. The CMS consumer rights materials distinguish air from ground ambulance services. The federal No Surprises Act's balance-billing prohibition does not generally cover ground ambulance charges. State law, separate contract terms, and public programs may establish different protections. The legal conclusion for an ambulance claim must therefore specify the transport mode, coverage and jurisdiction.
Air transport involves emergency geographic constraints and specialized aircraft capacity; the referring clinician, dispatch system, or emergency circumstance may effectively select the operator. Yet the cost structure differs from bedside professional care. Aircraft and crews have fixed readiness costs, long distances can influence actual operating expenses, and competition can be geographically limited. A single system-wide payment percentage consequently cannot be treated as an invariant measure of economically appropriate reimbursement.
The U.S. Government Accountability Office's 2026 pre- and post-Act review includes air ambulance among the four examined services. Its reported trends in network participation and payment must be interpreted by service type because distinct contractual practices and market structures create different response pathways. An increase in the share of contracted air ambulance claims, for example, is meaningful for payer–provider relationships but does not independently show how frequently households received illegal bills or the value of collected payments.
The exclusion of ground ambulances from the federal prohibition is a central boundary for financial-protection research. A patient may be strongly protected in the emergency department but face unexpected exposure for arriving there. This creates a possible discontinuity across components of the same medically necessary episode. The relevant empirical quantity is household exposure across the entire care pathway, while the legal classification is service-specific. Research that analyzes the hospital claim alone could incorrectly conclude that emergency financial risk has been eliminated.
Consent, directories, and continuity of care
Information rules protect people only when the information is timely, accurate, comprehensible, and actionable. Network directories may become outdated, particularly when practices change affiliations or plans contract with a physician group under a new entity. An accurate directory that names an in-network facility still leaves uncertainty about a separate anesthesiology group. Preoperative consultations and digital eligibility checks can reduce exposure in elective care, but urgent cases often make any active verification impracticable.
Notice and consent have an administrative and a substantive dimension. A signed document may prove that a form was presented, while meaningful consent requires an understandable disclosure, permitted circumstances, sufficient lead time, and a practicable ability to choose a participating alternative. A defensible audit therefore preserves the language used, time of delivery, patient condition, alternatives identified, and the professional service to which the form applied. It should also distinguish the patient acknowledging the cost of a noncovered elective service from a legally valid waiver of a particular statutory surprise-billing protection.
Continuity provisions may matter when a long-standing treating provider becomes nonparticipating during active treatment. The protection of a patient mid-course can require transitional payment and cost-sharing arrangements under the relevant law. Financial safeguarding here overlaps with clinical continuity: an individual facing chemotherapy, pregnancy care, or other scheduled interventions may face substantial medical cost from abrupt provider substitution. Patient welfare encompasses treatment continuity and access costs as well as the face value of a prohibited balance bill.
An integrated classification system needs at least four independent judgments: whether the benefit covers the medical service; whether the individual provider and facility participate in the specific plan; whether the statutory or contractual protection applies to the encounter; and which payment and cost-sharing amounts follow. The result is a more accurate distinction between a genuine surprise-billing event, ordinary network-based cost sharing, an excluded benefit, and an unresolved administrative error. Those distinctions become decisive when analyzing arbitration statistics and consumer outcomes.
Payment Dispute Mechanisms
Separation of patient liability and provider remuneration
The core institutional advance of statutory surprise-billing protection is the separation of the patient's payment duty from the price negotiation between an insurer and a provider. Where the law applies, the patient pays only the permitted in-network-equivalent cost sharing and the provider is restricted from demanding an additional balance. The parties with access to contracts, utilization records, price schedules, and institutional bargaining resources then resolve payment between themselves. The legal protection can be effective for the patient even before the ultimate provider payment is determined.
Under U.S. federal law, a sequence of rules governs payment if no applicable state law or agreement instead determines the amount. The framework involves a timely initial payment or denial, a period of open negotiation, and the possibility of a federal independent dispute resolution (IDR) process. The CMS explanation of federal IDR explains a 30-business-day open negotiation window and a subsequent four-business-day period to initiate the federal process. These are the regulatory steps for eligible claims; the underlying scope depends on the governing plan, state-law interaction, and service date.
A fundamental accounting identity clarifies what may change. For a covered service, let A denote the recognized allowed or determined amount, C the permitted patient cost sharing, and P the payer-paid component. If no other statutory adjustment or contractual transfer intervenes, the total recognized payment may be written A = C + P. An independent provider–payer dispute can alter the total amount owed to the provider while the patient's protected liability remains constrained by the separate cost-sharing rules. The identity describes an allocation after the relevant amounts are defined; it does not claim that every jurisdiction computes A or C in the same manner.
The presence of an insurer–provider dispute is thus consistent with strong patient protection. Conversely, the absence of a formal dispute says little about enforcement. Parties can accept an initial payment, resolve the issue privately, or decide that the amount at stake does not justify the administrative cost. Some providers may improperly bill patients despite a legal prohibition. Researchers need separate data sources for provider-payment conflicts and patient-facing conduct.
Patient cost sharing and qualifying payment amounts
The No Surprises Act uses the qualifying payment amount (QPA) as a central benchmark in parts of its federal framework. Broadly, for many services it is based on median contracted rates in a defined market and base period, subject to specified methodology and adjustments. It is crucial to distinguish the QPA from the provider's charged amount, from a universal market-clearing price, and from the amount ultimately selected in IDR. Its role varies with the service category, applicable law, and precise payment calculation.
For the patient, ordinary plan coinsurance can be applied to the relevant legal cost-sharing amount, with protections analogous to in-network treatment. For the interparty dispute, the QPA is one of the relevant statutory factors and is not simply synonymous with the final arbitrated price. In an economic model, the patient's price elasticity may depend on the allowed cost-sharing base while the insurer's spending depends on the total provider payment. The two can move differently. Treating QPA as both variables without investigating the governing rule creates measurement error.
Administrative calculation is important because apparently technical definitions can have major financial effects. Contracted rates must be grouped by service, provider specialty, insurance market, and geographical or other regulatory categories; plan data may be missing or heterogeneous. A benchmark computed from such rates can change if the calculation pools or excludes particular contract types. The official No Surprises Act notices report that the Fifth Circuit's August 2026 en banc ruling in TMA III addressed the validity of some QPA calculation rules and related guidance. The court's ruling must be distinguished from the continuing statutory prohibition of prohibited balance bills and from the continuing operation of IDR.
The empirical implication is that a payment trend cannot be interpreted without version control. A researcher comparing outcomes across 2022–2026 should identify the benchmark rules actually governing each episode and dispute, document agency responses to judicial decisions, and distinguish contractual changes from regulatory changes. A current general guidance page can describe operational practice but does not erase the historical legal regime in force for older services.
Open negotiation and independent dispute resolution
In a qualifying dispute, negotiation provides the parties a chance to agree before using formal arbitration. If negotiation fails, a certified IDR entity evaluates eligible competing offers under federal requirements. The CMS IDR process description outlines how the process separates eligibility, selection of an entity, submissions, and payment determinations. The process is commonly described as baseball-style arbitration because the decision generally selects one of the parties' offers rather than drafting an intermediate price.
This design changes bargaining incentives. When each side knows that a more extreme offer may become less likely to be selected, it may moderate its demands. However, the precise effect depends on the information available to the adjudicator, relevant statutory criteria, expectations about the other party's offer, and the cost of participating. When either side systematically anticipates a favorable outcome, arbitration can encourage filing rather than settlement. The resulting administrative workload becomes an element of overall insurance-system cost.
Volume figures must be interpreted with their units intact. The CMS IDR reports record 750,022 disputes initiated and 784,352 closed during August–September 2026; the closed total includes 664,305 payment determinations, 101,698 ineligibility decisions, and 18,349 other closures. Since system launch through September 30, 2026, the agency reports 7,798,616 initiated disputes and 7,304,875 closed. These figures measure administrative cases as defined by the federal program. They do not directly measure distinct patients, distinct encounters, prohibited patient invoices, or successful consumer recoveries, and batching rules alter the relationship between a dispute and its underlying items.
The fact that more cases were closed than initiated in a reporting interval reflects processing of the prior backlog, not negative demand. Measures of administrative throughput should separately report new filings, resolved determinations, withdrawals and ineligibility, the outstanding stock, processing times, and the composition of participating entities. Comparison of raw counts across different procedural regimes is inappropriate when eligibility and aggregation rules change.
The 2026 operational reforms and litigation
The Departments introduced revised operational requirements in 2026. The official federal IDR final-rule summary describes an effort to improve dispute eligibility screening, information exchange, consolidation, operational efficiency, and payment procedures. The CMS notices page further records implementation changes applicable to open negotiation periods beginning November 1, 2026. The ability to batch multiple eligible items into a single dispute changes both administrative cost and the interpretation of dispute counts.
Changes to an administrative interface are economically significant. If parties must document the applicable QPA, network status, correct payer, and claims identifiers in a standardized form, the data burden shifts toward verifiable records. That can reduce some processing delays and duplicate cases, although the effect on actual resolution speed requires observation. A measure that reduces filings by combining claims into batches may increase average dollars or items per docket without changing underlying medical utilization or provider out-of-network incidence.
Litigation further complicates comparison. The August 2026 TMA III notice describes a judicial ruling affecting QPA-related requirements, and confirms continued operations of federal IDR. It would be incorrect to infer that the balance-billing prohibition ended because a calculation regulation was invalidated. Equally, a historical payment estimate cannot automatically be treated as a valid current statutory calculation merely because it appeared in an earlier FAQ. Legal hierarchy requires distinguishing the Act, agency regulations, nonbinding guidance, court decisions, and the dates they apply.
The stability of a consumer's immediate protection can coexist with legal uncertainty for provider and plan payments. For insurance science, this illustrates the importance of decomposing “legal risk” into a claim's protection eligibility, a monetary calculation rule, a procedural right to contest payment, and the enforceability of an award. Each component has a different date and documentary source.
Economic incentives and bargaining effects
Before strong consumer protection, nonparticipating providers with high probability of treating insured patients could use the prospect of high balance bills as an outside option in network negotiations. An insurer seeking a comprehensive network might then accept higher negotiated reimbursement to avoid out-of-network exposure. Theoretical conclusions depend on the provider's ability to bill and collect, patient sensitivity to surprise bills, insurer reputational costs, and competition among providers.
Cooper, Scott Morton, and Shekita's emergency-care bargaining evidence places the outside option at the center of the empirical explanation for emergency physician prices. A statutory prohibition changes that option, but does not automatically determine the new equilibrium. An arbitration benchmark may be higher or lower than the previous negotiated rate, and parties respond to expected arbitration outcomes. A consumer-protection reform can therefore influence both out-of-network incidence and in-network negotiated prices through bargaining channels.
Let B be the expected collectible balance from a nonparticipating encounter before a prohibition, R the expected remuneration through an alternative payment mechanism, K the expected administrative cost, and Q the expected patient volume attributable to the insurer. A stylized provider outside option can be described as Q times expected net receipts per encounter, with costs and legal risk separately accounted for. Replacing potential B with regulated R changes the outside option, while dispute fees and information requirements influence K. This formulation is an analytical decomposition of ordinary bargaining incentives rather than a new model or an empirical estimate. Its value lies in specifying which data would test each causal channel.
Insurers likewise weigh network adequacy, premium effects, patient access, and the costs of formal disputes. A low payment rule may induce providers to join networks, exit them, change service allocation, or reduce participation in a specific product; the response depends on market alternatives and clinical capacity. A high award expectation can reduce an insurer's ability to steer care and may translate into future premiums. Whether either effect occurs, and for which population, is an empirical question requiring controlled comparisons.
The final incidence of a payment is distributive. A provider's additional remuneration may be absorbed by an insurer's margin in the short run, passed through to employers and households in premiums, financed through public contributions, or offset by changes to network design. A patient facing a prohibited bill may receive an immediate and substantial benefit even if some aggregate cost is later transmitted through premiums. The analysis must report both channels with their respective time scales and causal evidence.
Consumer Financial Outcomes
Potential balance bills, paid bills, and household harm
Financial exposure is a sequence of events, rather than one scalar quantity. A nonparticipating provider may submit a charge, the insurer may recognize an allowed amount, the provider may seek a residual balance, the patient may dispute that balance, the bill may be reduced or canceled, and a final payment or debt collection may or may not occur. Administrative and research records may capture only selected stages. A credible assessment of financial protection must assign a distinct measure to each.
A potential surprise bill is the balance that could be demanded under an observed service and payment arrangement. A sent bill records a provider's attempted demand. A collected payment establishes a cash outflow from the patient. A resulting debt may indicate persistent liability or an unlawful collection practice. Medical debt's effects on credit, care seeking, and household liquidity can be more durable than the initial invoice. Even the difference between billed and ultimately paid amounts may generate transaction costs through phone calls, appeals, confusion, or delayed treatment. These outcomes should be observed and described separately.
Sun and colleagues' historical hospital-claims study documented increasing potential out-of-network liabilities during 2010–2016. It is compelling evidence of exposure and contractual fragmentation, but the reported amounts should not be presented as actual cash collected from all affected households. The distinction is equally important when interpreting older studies of emergency claims. Garmon and Chartock's estimate of likely surprise bills identifies exposure among admissions, while household financial hardship requires additional measurement outside claims records.
To evaluate a protection's value, one can measure expected household medical payment conditional on a service episode, the upper tail of that payment's distribution, the frequency of disputed invoices, and the uncertainty surrounding expected obligations before treatment. A policy may have a small effect on the mean payment but substantially reduce catastrophic upper-tail exposure. Conversely, a reduction in reported balances may conceal unpaid collection attempts unless the data include invoice, adjudication, and collections stages. Consumer welfare depends partly on the ability to anticipate and insure against financial demands, not only on the average realized charge.
The WHO universal health coverage framework places financial protection alongside effective access to needed health services. This is an appropriate high-level outcome standard for international comparison, but general measures of catastrophic health expenditure include pharmaceuticals, copayments, transport, lost income, and many costs unrelated to surprise bills. They cannot substitute for identifying the particular residual provider charge regulated by a surprise-billing law.
Evidence before and after statutory intervention
Two broad groups of evidence are available: pre-reform incidence studies and post-reform administrative or claims evaluations. They answer different questions. Historical research establishes baseline institutional conditions and counterfactual plausibility, subject to sampling limitations. Studies covering the post-2022 period can identify changes in provider participation, billing, or remuneration; causal interpretation additionally requires separating the effects of the No Surprises Act from other simultaneous changes.
The Government Accountability Office examined selected services in its February 2026 provider-participation and payments review. Its analysis covers 2019–2023, thereby spanning the federal protections' January 2022 implementation. Among the four selected specialties—emergency medicine, anesthesiology, radiology, and air ambulance—it reports that the share of services provided in network rose after implementation for three. This pattern is consistent with a changing bargaining environment, but the agency also identifies limits of causal attribution and variation among groups. It does not establish a single effect on all providers or all claims.
An increase in network participation is valuable to examine because it may reduce uncertainty faced by consumers and may reduce formal dispute volume for covered services. Yet higher participation does not itself identify the final negotiated rate, whether the rate represents a price increase or decrease, or how much of any cost change reaches premiums. It may also reflect changes in business identifiers, specialty composition, plan mix, or negotiated contracts unrelated to the statute. For those reasons, the outcome belongs in the network and payment dimension of an evaluation and should be correlated with separately measured consumer outcomes.
The evidence from New York's earlier reform provides a quasi-experimental point of comparison. The published emergency physician study examines an arbitration-based intervention in a particular state and payment market. Its results help clarify how a payment rule can change the incentive to remain out of network. They do not supply a national post-2022 impact estimate for every service or certify that an identical mechanism would operate in every jurisdiction.
The most defensible causal study of national consumer financial protection would link comparable encounters in treatment and control settings, or leverage statutory exemptions and staggered state protections that provide credible variation. Such designs must address changes in coding, provider organizations, network contracts, and the composition of episodes. Difference-in-differences approaches require evidence of comparable pre-intervention trends; event studies can reveal anticipatory behavior and delayed effects; regression discontinuities at contract or threshold boundaries require a credible nonmanipulation assumption. None of these designs can be validated by one overall administrative count.
Administrative compliance and enforcement
The U.S. administrative system captures formal complaints and reviews possible violations. In a December 2025 enforcement and complaint report, CMS reported that 39,999 complaints concerning federal health insurance market-reform requirements had been received from January 2022 through December 2025, most involving provisions of the No Surprises Act. The report identified 15,145 closed complaints, including 2,086 closed with a violation found, 7,838 closed without one, and 24,854 open. The agency reported approximately $30.05 million in monetary relief to consumers and providers together. The combined relief figure must never be represented as an amount returned exclusively to patients.
Complaint counts are a selected outcome. Some patients lack information about their rights, some obtain an informal correction, some do not pursue a complaint, and some report a grievance that is later found to fall outside the statutory prohibition. Administrative data therefore measure the volume and adjudication of reported problems, agency processing, and documented relief. Their denominator is not the national number of affected patients or eligible encounters. The ratio of violations to closed complaints is not a population rate of illegal billing.
Enforcement also operates through plan and provider practices that are often less visible than formal complaints. Automated claims reprocessing may correct erroneous cost-sharing calculations. Financial counselors may direct patients to the appropriate resolution pathway. Payers may identify systematic overbilling from patterns of claims and appeals. These mechanisms can contribute to compliance while remaining difficult to quantify from a federal complaints dashboard.
Patient-provider disputes involving a good faith estimate are another legally distinct category. For certain uninsured or self-pay patients, federal requirements establish a patient-provider payment dispute process when a bill is substantially higher than the estimate, with a generally relevant threshold of at least $400 above it. This process should not be merged analytically with federal insurer–provider IDR. The two have different parties, documentation, timing, and source populations. A study that labels both simply “surprise-billing arbitration” loses information about the protection mechanism.
Distributive effects and second-order costs
The distributive consequences of surprise billing depend on income, insurance product, location, health condition, and the availability of alternative providers. People requiring emergency or highly specialized services have fewer practical choices; patients with low financial reserves can experience serious hardship from relatively small unexpected balances. Limited proficiency in the language used for billing, limited access to digital records, disability, and discontinuous insurance enrollment can increase administrative burdens. Researchers should assess these factors through observed data rather than presume uniform risk.
Protection from an excess provider charge has a direct redistributive effect when it reduces the amount collectible from the patient. It may also reduce time costs and uncertainty. The ultimate insurer payment is allocated through a different market process. Greater provider remuneration can be financed from premiums, public expenditure, or other insurance-system components; lower remuneration can affect provider revenue, network composition, and investment. A net welfare assessment requires measuring those effects over time, recognizing that they may be experienced by different populations.
The OECD's financial-hardship indicators offer a method of comparing high out-of-pocket burdens relative to household resources. Such an indicator can be used as a broader dependent variable, with careful interpretation. A reduction in balance bills could improve high-cost outcomes for a relatively small exposed population while barely moving a national catastrophic-expenditure statistic. Conversely, a country could have no U.S.-style surprise bills and still show considerable hardship from ordinary patient charges, exclusions, medicines, dental care, or long-term care.
The dynamic effect on access deserves equal weight. If provider participation and payer reimbursement change, the geographic distribution of contracted clinicians may also change. Access outcomes include distance to an available clinician, appointment waiting time, emergency capacity, post-acute follow-up, and continuity of treatment. Financial protection should be assessed in the context of both enforceable patient liability and access to medically appropriate care. A reduction in a narrow billed-balance statistic does not alone establish a comprehensive improvement in health-system performance.
Data, Measurement, and Validation
Claim-level reconstruction and eligibility
A valid administrative evaluation reconstructs the legally relevant episode from raw records. The minimum source tables are: beneficiary enrollment and plan product; date-stamped coverage terms; institutional claims; professional claims; provider and facility identifiers; network agreements; initial billed charges; payer adjudication; cost-sharing transactions; payment determinations; patient invoices and collections where accessible; complaints; and applicable legal and regulatory versions.
The join must recognize that one clinical episode can generate multiple claims with distinct claim identifiers and adjustment sequences. A resubmitted claim should not be mistaken for a new treated patient; a previously denied service may later be paid after an appeal; the same encounter may appear in facility and professional files. Episode grouping should be clinically defined, with separate crosswalks for inpatient admissions, emergency visits, surgery, diagnostic services, and ambulance transport. Such grouping must be reproducible from explicit dates, provider codes, and claim relationships.
Eligibility requires a decision tree. First identify the applicable coverage arrangement. Then determine the service date, jurisdiction, service setting, whether the facility and provider are participating under the relevant plan, whether the encounter meets an emergency or protected nonemergency definition, and whether any legally permitted notice-and-consent exception applies. Next identify the correct patient cost-sharing amount and insurer–provider payment procedure. The CMS provider resource collection supplies scope and operational guidance, but claim-level classification must use the actual statutory and regulatory provisions applicable to the encounter.
Some patients covered by Medicare, Medicaid, Veterans Affairs health programs, TRICARE, and related public arrangements are protected from various balance-billing practices under different legal structures. They should not simply be treated as untreated controls for a federal private-plan intervention. Employer-sponsored self-funded plans, fully insured state-regulated products, and public purchasing programs likewise differ in legal authority. Researchers should encode these differences rather than collapse them into a binary “insured” variable.
Data access is a practical limitation. Provider network files can be confidential or imperfect, actual patient collections may be unavailable, and some public federal reports aggregate services or disputing entities. A claims dataset that lacks invoice and collections data can estimate apparent exposure but not actual patient payment. The missing component must be declared; it should not be imputed silently from the provider's submitted charges.
Defining financial and behavioral estimands
The research estimand determines what an outcome means. Let i index a clinically defined episode, t time, and G a relevant group of covered encounters. Four distinct estimands are useful.
The exposure estimand is Pr(S_i=1 | G,t), where S indicates that an episode includes a legally relevant nonparticipating provider and a potential unprotected balance demand. The attempt estimand is Pr(B_i>0 | G,t), where B is the amount of an excess balance actually billed to the patient after normal payer adjudication. The collection estimand is E(C_i | G,t), where C is the patient amount actually collected that exceeds legally permitted cost sharing. The administrative estimand is the number of disputes per eligible billed service or per distinct eligible episode, with attention to changes in batching and claim definitions.
These estimands are related but none is a direct substitute for another. An intervention can cause S to fall by increasing network participation; it can cause B to fall by forbidding invoices even if providers stay nonparticipating; or it can cause C to fall by improving dispute resolution and enforcement after the invoice is issued. A patient may also experience financial distress because of ordinary deductible liability even when C is zero. Any publication that reports a single undifferentiated “surprise billing rate” should state its observable construction.
For price outcomes, the relevant estimands include initial payer offers, negotiated rates, QPAs calculated under applicable methodology, IDR awards, realized total allowed payments, and patient cost-sharing bases. These variables should be analyzed per comparable service and adjusted for case mix, site of care, severity, contract structure, and provider type. A higher IDR award is not necessarily a higher amount ultimately collected from a patient. A change in the median award among arbitrated cases may reflect changing selection into arbitration rather than a change in the population distribution of provider payments.
An informative results table should report at least one patient-centered and one system-centered outcome. The patient-centered measure might be unlawful collected balances per 1,000 eligible episodes and its distribution across income strata. The system-centered measure might be total recognized payments per service, adjusted for clinical case mix, and the share of eligible episodes requiring formal arbitration. Both require a consistent denominator and uncertainty intervals. Reporting them separately prevents a financially beneficial consumer protection from being misdescribed as proof of lower total medical spending.
Causal inference and selection effects
The observed mean difference in a measured outcome before and after enactment is not itself the causal effect of legislation. During the period examined, population enrollment, health-service use, pandemic disruptions, hospital consolidation, contract negotiations, plan benefit design, billing software, and claims coding can all change. The counterfactual is what the same market and population would have experienced without the intervention. Different identifying strategies answer different counterfactual questions.
For difference-in-differences, a basic estimand compares the change from before to after in a group newly exposed to protection with the contemporaneous change in a group whose protection did not change. The design requires sufficiently comparable trends absent the intervention and careful treatment of policy spillovers. Because federal law may supersede or complement state protections and apply to self-funded plans differently, a state's prior law cannot be coded as a simple treatment indicator without a plan-level legal classification. A treatment effect estimated among emergency physician claims may not generalize to ground ambulance invoices, which lie outside the principal federal prohibition.
An event study can test for pre-trends and study implementation dynamics, but it should avoid treating anticipation as a violation of its assumptions when parties reasonably changed contracts in advance. An instrumental-variable design would require a source of exogenous policy exposure that affects the financial outcome only through the relevant legal channel. Structural models of payer–provider bargaining can simulate payment and network responses, but their results depend on the modeled disagreement outcomes and demand assumptions. Simulations are informative mechanism analyses; they do not automatically provide observed post-reform effects.
The GAO 2026 longitudinal review is a useful example of cautious empirical interpretation: the agency documents trends before and after federal implementation and consults stakeholders, while identifying what available claims and interviews cannot establish. Scientific review should preserve these limitations when generalizing from the agency's findings. Pre-2022 studies such as Sun and colleagues' claims dataset analysis provide a historical baseline but cannot serve as a matched causal counterfactual for 2026 without additional design work.
Selection into formal arbitration is particularly strong. Only a subset of eligible, contested claims reaches the IDR program, and each actor's filing propensity depends on fee amounts, claim size, knowledge, expected success, batching rules, and administrative capacity. As a result, an award distribution is conditional on filing, eligibility, and determination. Its median can move because new categories of claims are being filed, even if the underlying distribution of actual health-service prices remains unchanged.
Validation, sensitivity, privacy, and reproducibility
A defensible source-to-claim audit begins with exact document version and date. For every legal statement, record the law or agency instrument, jurisdiction, effective period, plan scope, and any subsequent judicial decision. For every empirical estimate, record dataset years, enrollment universe, payer type, unit of analysis, numerator, denominator, confidence measure, identification strategy, and study-version history. The federal IDR reporting series and 2025 complaint report must remain separate administrative datasets.
At least five sensitivity checks are indispensable: alternative episode grouping rules; exclusions of adjusted or duplicate claims; alternative specialty and facility identification; changes in the observation of actual patient collections; and a legal-applicability audit for state-regulated versus federally regulated products. Further checks may include income-weighted burdens, geographic market definitions, hospital and insurer fixed effects, time-varying plan benefits, and comparison of billed charges with realized payments. The estimation code should preserve the difference between billed, allowed, awarded, and collected amounts throughout.
Privacy and confidentiality are major practical concerns. Medical claims may contain diagnoses, dates, provider details, and other information that can identify individuals even after obvious names are removed. A reproducible scientific study can publish its classification rules, variable dictionary, code, disclosure-safe aggregates, and source-version ledger while retaining identifiable records in secure authorized facilities. A claim to reproducibility should specify which results can be independently recomputed from public data and which require a controlled data-access agreement.
A final scientific QA should then test whether the prose is supported by the cited primary material. An official description of a federal process supports its procedural scope and current operation; it does not prove that the law reduced medical debt by a particular amount. A published claims analysis supports an estimated rate for its sample and period; it does not establish the legal outcome of every claim. A comparative national health-system source supports that country's rules and institutional design, with no presumption that its terminology maps directly onto U.S. benefit categories.
The objective of this validation discipline is to preserve the distinction between measured outcomes, documented institutional requirements, causal inference, and interpretive synthesis. It is the central condition under which evidence from heterogeneous insurance systems can inform a coherent assessment of surprise billing and cost-sharing protection.
International Synthesis and Scientific Conclusions
United States
The United States is the most revealing case for this study because the legal instrument addresses a precise mismatch between patient choice and provider contracting while maintaining decentralized insurer–provider bargaining. The No Surprises Act enacted text creates direct consumer protections for specified emergency, facility-based, and air ambulance services, and an interparty process for determining many disputed payments. It is the coexistence of these mechanisms that makes a simple before-and-after measure of out-of-network invoices insufficient to evaluate the law.
The CMS provider requirements specify regulated conduct, while the federal IDR guidance specifies the procedural channel for eligible disputes. Laws applying to insured products can coexist with relevant state regimes; self-funded employer plans raise distinct federal-preemption and state-law questions. The applicable payment route should be determined for the claim's product and jurisdiction, not merely inferred from a national headline. Public beneficiaries also have protections under other systems that do not share the same private-plan procedural design.
The 2026 record provides a useful division of evidence. The Government Accountability Office's longitudinal claims review describes network and payment changes in selected specialties from 2019 through 2023. The CMS dispute reporting describes the administrative volume and outcomes of formal interparty proceedings. The complaint and enforcement record describes reported compliance issues, agency case closure, and monetary relief involving both patients and providers. The sources are jointly informative only when their different populations, units, and selection mechanisms are preserved.
The U.S. experience demonstrates that enforceable consumer liability protection can be instituted independently of final provider price determination. Its strongest transferable lesson is institutional: a patient who cannot choose the contractual counterparties should not have to perform that negotiation after treatment. Whether the specific arbitration benchmark optimizes premiums, network access, and clinical capacity remains a continuing research question, especially given the 2026 legal and operational changes.
Netherlands
The Netherlands organizes compulsory basic health insurance through regulated competition among insurers and health care providers. Consumer exposure depends on the insurance product, contracted provider network, covered benefits, and required contributions. The Dutch government's description of insurance policies distinguishes in-kind policies, under which insurers pay contracted providers directly, from combination policies that may reimburse contracted or noncontracted care differently. A person selecting a noncontracted provider can face partial reimbursement or upfront payment, depending on the policy.
This institutional arrangement has some of the same information problems as U.S. network insurance, but it should not be described as simply another implementation of the No Surprises Act. The operative protections derive from Dutch statutory health insurance, insurer duties, contractual reimbursement terms, and relevant emergency or continuity provisions. The U.S. classification “balance bill prohibited under federal IDR” does not transfer automatically.
The Netherlands illustrates a separate mechanism for reducing uncertainty: making benefit coverage and provider-contract networks legible to consumers before they choose a plan or care location. The government's policy guidance explicitly asks prospective insured persons to consider whether preferred providers have an insurer contract and how much noncontracted treatment may cost. This transparency is valuable when a choice exists, while an emergency or facility-assigned specialist can still create an information disadvantage.
A comparative study should distinguish the statutory deductible and lawful noncontracted-care reimbursement gap from a prohibited surprise provider balance. A patient paying a published deductible is experiencing cost sharing; a patient who is unexpectedly sent to a clinician outside a meaningful choice set may be exposed to a different coordination problem. The appropriate Dutch estimand is therefore an encounter-level unexpected residual obligation relative to the applicable product, with contract information and any statutory safeguards included.
Switzerland
Swiss compulsory health insurance combines standardized statutory benefit obligations with managed reimbursement, provider tariffs, and patient contributions. The Federal Office of Public Health's benefits and tariffs guidance explains that service remuneration is organized through agreed or government-regulated tariffs. It also differentiates payment by the insurer directly to a provider from payment by a patient who later seeks reimbursement. These payment channels alter the cash-flow experience of a patient even when the final insured benefit is similar.
The distinction between the tiers garant and tiers payant arrangements is particularly instructive. In a reimbursement-first arrangement, a patient can receive and initially bear an invoice before obtaining the insurer's contribution; in a direct-payment arrangement, the insurer settles with the provider and seeks the patient's lawful share. An identical statutory benefit can therefore generate different short-term liquidity requirements. Research focused only on net reimbursement may miss a temporary but consequential financing burden.
Hospital choice creates another important comparison. The Swiss benefits and tariffs explanation states that treatment at a hospital outside the canton of residence may result in a difference borne by the patient when the chosen hospital's tariff exceeds the applicable home-canton benchmark. It also provides for full compulsory coverage where extra-cantonal hospital treatment is medically necessary because of emergency circumstances or unavailability of the treatment in the home canton. The source's “full costs” is a description of the covered hospital treatment under the governing rules; general statutory deductibles and other patient contributions require separate consideration.
This is a strong example of treating constrained medical choice differently from elective choice. It is institutionally unlike U.S. federal insurer–provider baseball arbitration, yet both arrangements recognize that clinically necessary emergency care raises distinctive financial-protection concerns. The comparative unit should be the patient's permitted residual obligation and access to medically necessary treatment, not an asserted equivalence between nationally different rate-setting systems.
European cross-border healthcare
Cross-border care makes the legal role of the payer's home system particularly visible. The European Union's Directive 2011/24/EU addresses patients' rights in cross-border healthcare, including conditions and limits of reimbursement. Its Article 7 generally links reimbursable costs to the level of cost that would have been assumed by the patient's Member State of affiliation for equivalent domestic benefits, subject to the directive's qualifications and relevant arrangements. That reimbursement is not an unlimited guarantee against the difference between a foreign provider's charge and the home system's recognized amount.
The directive is only part of a broader European legal architecture. Emergency access during a temporary stay may also operate through social-security coordination instruments and European Health Insurance Card arrangements with different applicable rules. Planned treatment may be subject to prior authorization in specified circumstances, while care under the cross-border directive has its own conditions. A comparison that merges these routes could incorrectly label a legally expected reimbursement limitation a prohibited surprise bill.
The economically relevant mechanism is allocation of tariff risk across borders. A patient may know that a service is covered at home but face a foreign charge based on another country's price schedule. The financial consequence depends on eligibility, provider type, public versus private care, prior authorization, and which legal pathway applies. A study needs the patient's entitlement document, the place and type of care, the invoiced amount, the home-country reimbursement amount, and the applicable regulatory route.
This case expands the scope of surprise-billing research beyond domestic network participation. The patient's uncertainty may arise from the interaction of two public payment regimes rather than from a single insurer's contracting with a private emergency physician. The comparison supports an international definition focused on unpredictable and constrained-choice residual liability while preserving the different legal mechanisms that create or restrict it.
England
England's National Health Service is organized chiefly through public entitlement and provider payment arrangements rather than a general market in which insured patients must verify participation of each hospital-based professional. For persons entitled to NHS-funded care without charge at the point of service, the U.S. out-of-network balance-billing mechanism is therefore not the relevant default problem. Ordinary NHS services are financed through another institutional pathway.
The boundaries of entitlement nevertheless matter. The Department of Health and Social Care's May 2026 overseas visitor charging guidance sets out when some overseas visitors may be charged for relevant NHS services, exemptions, and the requirement that urgent or immediately necessary treatment must proceed without delay because of payment questions. In qualifying circumstances charges can still be recovered after treatment. Thus a duty to provide urgent care and a right to receive it free of charge are analytically different propositions.
The exemption for accident and emergency services at the relevant settings does not mean every subsequent hospital admission or follow-on procedure is necessarily free to a chargeable visitor. Conversely, a visitor entitled under an international arrangement or another exemption cannot automatically be charged merely because that person is not ordinarily resident. The England-only guidance also does not purport to govern the devolved charging regimes of Scotland, Wales, or Northern Ireland.
England illustrates an entitlement-based way of controlling unexpected financial liability. A policy evaluation of its system should focus on the proper classification of patients and services, clarity of exemptions, billing compliance, and continuity of necessary treatment. The unit of analysis is entitlement and NHS charging, rather than network-negotiated private insurer payments. Comparing final household burdens is possible; equating the underlying contractual mechanisms would be misleading.
Australia
Australian private health insurance offers a further perspective because of its explicit emphasis on policy disclosure, insurer reimbursement arrangements, and informed financial consent. The Australian Government's policyholder disclosure requirements state that insurers must inform policyholders about relevant policy coverage, changes, and applicable out-of-pocket costs for treatment or services covered by the policy. This disclosure function helps patients decide whether to consent to medical treatment with known financial consequences.
Australia also has a public Medicare financing structure and a distinct private insurance market. Out-of-pocket gaps can arise from hospital and medical professional fee arrangements, exclusions, insurer agreements, and benefits. An informed financial-consent requirement improves the patient's ability to anticipate certain charges, but a disclosure rule does not necessarily cap every amount payable in every emergency or private-hospital situation.
The analytical significance is the distinction between protection by payment limitation and protection by informed choice. A statutory cap on a balance bill acts after the patient encounters a nonparticipating provider, whereas well-designed price disclosure can alter elective choices beforehand. In emergencies, information may arrive too late to be actionable. A complete comparative evaluation therefore assesses the availability of meaningful alternatives along with the quality of information provided.
Because the Australian material cited here concerns insurer disclosure duties, this chapter makes no claim of a comprehensive Australian statutory analog to U.S. federal IDR or a precise national rate of surprise medical billing. Those further questions require a separate service-specific legal and empirical study with original Australian administrative and insurer claims data.
Comparative scientific interpretation
These jurisdictions show several distinct institutional ways to manage unpredictable patient exposure. The United States limits certain balance bills while allocating disputed payments through insurer–provider processes. The Netherlands relies in important part on compulsory insurance products, contracted networks, and policy-specific reimbursement. Switzerland combines compulsory benefits, tariff governance, reimbursement channels, and special treatment of medically necessary extra-cantonal hospital care. European cross-border healthcare uses entitlement and reimbursement rules tied to the responsible state. England relies on NHS entitlement and regulated charging. Australia imposes significant insurer information duties alongside its public and private financing arrangements.
A useful comparison must begin with a common outcome rather than a common label. For a clinically necessary episode, ask: How much may the patient lawfully be required to pay? Which participants can the patient realistically choose? Which obligation is disclosed in advance? How does a patient challenge an incorrect demand? How is provider remuneration set when the patient is protected? What cost remains for the household at the time of treatment, while awaiting reimbursement, and ultimately after settlement?
The answers reveal three broad causal channels. First, payment rules determine the upper bounds and timing of patients' immediate liability. Second, provider contracting and public tariffs determine the division of remuneration between purchasers and providers and shape long-term premiums, taxes, and access. Third, administrative implementation determines whether legal entitlements are realized in billing and collections. These channels are analytically separate but interact in actual medical care.
The international evidence also demonstrates why financial protection cannot be reduced to the existence of a statutory out-of-pocket cap. Deductibles, reimbursements, network restrictions, uncovered services, transport charges, and the timing of payments can all alter the experience of a household. The OECD financial-hardship framework and the WHO universal coverage definition therefore serve as outcome frameworks rather than as proofs that national legal regimes are identical.
The strongest comparative conclusion is that constrained choice requires a specifically designed financial-protection mechanism. Information about network status is valuable where the patient can select among providers; a regulated patient liability rule is particularly valuable where the patient cannot choose the provider, the service is clinically urgent, or a contract fragment is hidden within an otherwise participating episode. Payment governance and enforcement must then preserve the promised protection without confusing interparty remuneration with household liability.
Conclusion
Surprise billing is a problem of fragmented authority over medical prices, clinical services, and financial obligations. A single care episode brings together actors that negotiate and adjudicate payments under separate contracts, although the patient often experiences those activities as one necessary service. The risk becomes especially acute when the patient cannot choose an emergency physician, an ancillary professional, or an ambulance operator. Legal systems address that exposure by defining protected circumstances, identifying the permissible patient share, regulating provider conduct, and assigning remaining payment decisions to the responsible institutions.
The reviewed primary evidence establishes a clear history of substantial potential out-of-network exposure in U.S. commercial insurance before the federal No Surprises Act. The Act changed the legal framework for specified privately insured encounters beginning in January 2022. Subsequent evaluations found changes in provider network participation, while administrative reports documented a large and evolving volume of formal payment disputes. By November 2026 the federal framework had also experienced consequential judicial and regulatory developments. Claims about individual patient relief, causal effects on negotiated rates, and system-wide premium consequences must be grounded in the specific data and legal version relevant to each proposition.
The principal analytical result is a separation of three outcomes that are frequently conflated: the patient's legally enforceable liability; the provider's ultimate remuneration; and the insurance system's subsequent allocation of expenditure and access. Legal protection can constrain the first while the second remains disputed and the third develops over time. That result follows from the comparative reconstruction of existing institutions and source evidence, rather than from a proposed new theoretical system.
A second conclusion concerns measurement. Counts of arbitrated cases cannot be read as counts of surprise-billed patients. Potential billed balances cannot be read as actual collected payments. Complaint findings cannot be transformed into population prevalence. Network participation cannot stand in for verified reductions in household debt. Each source has a distinct denominator, period, population, selection mechanism, and claim to scientific inference. An evaluation becomes credible when it records those distinctions, connects actual encounters to applicable legal protections, and distinguishes observed changes from identified causal effects.
A third conclusion is international. The precise U.S. insurance-network problem has no exact universal counterpart because compulsory insurance, tariff regulation, state reimbursement, and direct public service provision distribute financial authority differently. Nevertheless, an intelligible comparative standard exists: a person's exposure to unexpected, enforceable medical charges when clinical necessity and institutional arrangements limit informed choice. The Netherlands, Switzerland, European cross-border law, England, and Australia clarify how network disclosure, tariff administration, medical necessity, public entitlement, and financial consent alter that exposure.
Future research should prioritize linked claims–invoice–collections data, credible causal designs for the 2022 federal intervention, provider-contract and premium pass-through studies, the effects of changing 2026 dispute procedures, state and federal enforcement heterogeneity, and the consequences for clinically meaningful access. These priorities derive directly from identified evidence gaps and are proposed as research questions. The normative purpose of surprise-billing protection is fulfilled only when the recorded billing process implements the rights that the governing legal instrument promises.
Technical Appendix: Research and Compliance Audit Protocol
Legal applicability reconstruction
For each proposed study define the jurisdiction, insurer or plan type, effective service date, clinical setting, provider participation, facility participation, transport mode, statutory exemption, and any qualifying notice-and-consent evidence. Record the governing statutory text, applicable rule or agency notice, date of any judicial decision affecting calculation, and payer/provider jurisdictional assignment. Retain source versions corresponding to the service dates. Do not conflate an enacted statute with guidance describing a later implementation.
Claims and episode reconstruction
Specify the data custodian, coverage universe, inclusion dates, service and diagnosis classification, linked claim adjustment keys, provider identifiers, network-contract effective dates, facility and professional joins, and eligible-episode denominator. Define missing-data handling and a reproducible duplicate-suppression rule. Preserve line-item results separately from episode-level aggregation. In studies of arbitration, capture the claim count represented by each batched dispute and the procedural rules in force.
Financial quantities
Maintain separately the submitted charge, initial payer payment, plan cost-sharing base, legally permitted patient obligation, assessed or disputed provider balance, final adjudicated provider payment, invoice delivered to the patient, amount collected, refunds, and outstanding debt. Identify the timing of each cash flow. The relationship between patient payment and insurer-paid amounts must be interpreted under the applicable contractual and statutory framework, rather than imposed as an invariant across countries.
Scientific inference and uncertainty
For an intervention study define treatment, comparison group, pre-treatment period, post-treatment period, target population, outcome, estimand, treatment assignment mechanism, and primary identification assumptions. Report uncertainty measures, sensitivity to episode definitions, coding and payer mix, parallel pre-trends where relevant, provider consolidation, changing plan types, and exposure spillovers. Separate descriptive findings from counterfactual inference. When original data are inaccessible, report only evidence-supported published results.
Source and editorial integrity
Each external research URL appearing in the manuscript must correspond to a source entry in EVIDENCE.json with a claim-specific verification statement and access limitation, and to a full bibliographic entry in References. In-text hyperlinks use meaningful contextual phrases; internal source identifiers remain within the evidence ledger. The authoritative archive preserves the exact manuscript and evidence SHA-256, approved H1 and canonical URL, individual research decision, date, editor, and Git commit. Passing a structural source-link checker provides evidence of technical integrity, while substantive scholarly and editorial review remains a separate documented decision.
References
Australian Government Department of Health, Disability and Ageing. (2025). What insurers must tell policy holders. https://www.health.gov.au/topics/private-health-insurance/operating-rules-for-private-health-insurers-and-providers/what-insurers-must-tell-policy-holders.
Centers for Medicare & Medicaid Services. (2026). About independent dispute resolution. https://www.cms.gov/initiatives/no-surprise-billing/overview/engaging-idr/about-independent-dispute-resolution.
Centers for Medicare & Medicaid Services. (2025). Complaint data and enforcement report on health insurance market reforms: December 2025. https://www.cms.gov/files/document/december-2025-complaint-data-enforcement-report.pdf.
Centers for Medicare & Medicaid Services. (2026). Federal independent dispute resolution operations final rule. https://www.cms.gov/newsroom/fact-sheets/federal-independent-dispute-resolution-operations-final-rule.
Centers for Medicare & Medicaid Services. (2026). Independent dispute resolution reports: Through September 30, 2026. https://www.cms.gov/initiatives/no-surprise-billing/overview/policies-resources/independent-dispute-resolution-reports.
Centers for Medicare & Medicaid Services. (2026). Know your rights when using health insurance. https://www.cms.gov/initiatives/your-patient-rights/medical-bill-rights/know-your-rights/insurance.
Centers for Medicare & Medicaid Services. (2026). No Surprises Act: Overview of key consumer protections. https://www.cms.gov/files/document/nsa-keyprotections.pdf.
Centers for Medicare & Medicaid Services. (2026). No Surprises Act notices. https://www.cms.gov/initiatives/no-surprise-billing/overview/notices.
Centers for Medicare & Medicaid Services. (2026). Providers: Payment resolution with patients. https://www.cms.gov/initiatives/no-surprise-billing/overview/policies-resources/providers-payment-resolution-patients.
Centers for Medicare & Medicaid Services. (2026). Provider requirements and resources. https://www.cms.gov/initiatives/no-surprise-billing/overview/policies-resources/provider-requirements-resources.
Cooper, Zack, Fiona Scott Morton, and Nathan Shekita. (2020). Surprise! Out-of-network billing for emergency care in the United States. Journal of Political Economy, 128(9), 3626–3677. https://doi.org/10.1086/708819. https://isps.yale.edu/research/publications/isps17-22.
Cooper, Zack, Hao Nguyen, Nathan Shekita, and Fiona Scott Morton. (2020). Out-of-network billing and negotiated payments for hospital-based physicians. Health Affairs, 39(1), 24–32. https://doi.org/10.1377/hlthaff.2019.00507. https://pubmed.ncbi.nlm.nih.gov/31841351/.
Department of Health and Social Care (United Kingdom). (2026). Charging overseas visitors in England: Guidance for providers of NHS services. https://www.gov.uk/government/publications/nhs-cost-recovery-overseas-visitors/charging-overseas-visitors-in-england-guidance-for-providers-of-nhs-services.
European Parliament and Council of the European Union. (2011). Directive 2011/24/EU on the application of patients’ rights in cross-border healthcare. https://eur-lex.europa.eu/eli/dir/2011/24/oj/eng.
Federal Office of Public Health (Switzerland). (n.d.). Health insurance: Benefits and tariffs. https://www.bag.admin.ch/en/health-insurance-benefits-and-tariffs.
Garmon, Christopher, and Benjamin Chartock. (2017). One in five inpatient emergency department cases may lead to surprise bills. Health Affairs, 36(1), 177–181. https://doi.org/10.1377/hlthaff.2016.0970. https://pubmed.ncbi.nlm.nih.gov/27974361/.
Government of the Netherlands. (n.d.). More Q&As about health insurance in the Netherlands. https://www.government.nl/documents/question-and-answer/health-insurance/more-qas-about-health-insurance-in-the-netherlands.
Organisation for Economic Co-operation and Development. (2025). Health at a Glance 2025: Financial hardship and out-of-pocket expenditure. https://www.oecd.org/en/publications/health-at-a-glance-2025_8f9e3f98-en/full-report/financial-hardship-and-out-of-pocket-expenditure_4160e6d2.html.
Sun, Eric C., Michelle M. Mello, Jasmin Moshfegh, and Laurence C. Baker. (2019). Assessment of out-of-network billing for privately insured patients receiving care in in-network hospitals. JAMA Internal Medicine, 179(11), 1543–1550. https://doi.org/10.1001/jamainternmed.2019.3451. https://pubmed.ncbi.nlm.nih.gov/31403651/.
United States Congress. (2020). Consolidated Appropriations Act, 2021, Pub. L. 116-260, Div. BB, Title I (No Surprises Act). https://www.govinfo.gov/app/details/PLAW-116publ260.
United States Government Accountability Office. (2026). Private health insurance: Provider participation and payments for selected services before and after the No Surprises Act (GAO-26-107169). https://files.gao.gov/reports/GAO-26-107169/index.html.
World Health Organization. (2025). Universal health coverage (UHC). https://www.who.int/news-room/fact-sheets/detail/universal-health-coverage-%28UHC%29.