CMS Medicare rules are the machinery that turns the broad promises of Title XVIII of the Social Security Act into the concrete numbers on a hospital’s remittance advice and the coverage decisions that determine whether a particular service is paid. Congress wrote the statute, but it left the payment formulas, the coverage policies, and the daily claims machinery to the agency renamed the Centers for Medicare and Medicaid Services in 2001. This guide traces that machinery end to end: how proposed and final rules set annual payment rates, how national and local coverage determinations decide which services count as reasonable and necessary, how manuals and guidance instruct the contractors that process claims, how the five-stage appeals ladder reviews denials, and how the statute’s rulemaking provision, 42 U.S.C. 1395hh, decides which of these documents are law and which are merely instruction.

The statute is only the beginning

Title XVIII of the Social Security Act, which created the federal health insurance program for the aged and disabled, is administered by the Centers for Medicare and Medicaid Services, an agency of the Department of Health and Human Services headquartered in Baltimore. The agency is the successor to the Health Care Financing Administration, established in 1977 to bring the federal health financing programs under one roof, and renamed the Centers for Medicare and Medicaid Services in 2001 to reflect its broader responsibilities across the programs it manages. The implementing regulations for Title XVIII are compiled in Title 42 of the Code of Federal Regulations, at 42 C.F.R., where the agency translates statutory commands into the working definitions, formulas, and payment amounts that govern the program from year to year. Readers who need a guide to reading citations in that form, including the difference between a statute section, a regulation part, and a manual chapter, can consult the series explainer on how to read a federal statute, how to read a federal statute, before proceeding.

The central argument of this article is that the working law of the program is found less in the statute books than in these agency documents. Congress enacts the authorization, defining who is eligible, what categories of services are covered, and what methods of payment the agency may use. The agency then produces the operative numbers, and the distance between authorization and operation is wide. The statute says that hospitals furnishing inpatient services shall be paid under a prospective payment system, but it does not say what a given admission was worth in a given fiscal year. That figure is fixed by the agency in an annual rule, after a cycle of proposal, public comment, and revision. The same pattern repeats across nearly every setting the program pays. Physicians, outpatient departments, skilled nursing facilities, home health agencies, inpatient rehabilitation facilities, and long term care hospitals all receive amounts that were set, or reset, in a rulemaking published in the Federal Register. A practitioner who reads only the statute therefore knows the architecture of the program without knowing its economics, while a practitioner who reads the annual rules knows the economics without always seeing the architecture. Understanding how the program functions requires holding both in view, and recognizing which questions each answers. The architecture itself, including the division between hospital insurance and supplementary medical insurance and the meaning of the statutory text being implemented here, is mapped in the series account of the two parts of the statute, the two parts of the Medicare statute.

The statute itself draws a boundary around what the agency may do, and that boundary shapes the character of the rulemaking that follows. Section 1801 of the Social Security Act, codified at 42 U.S.C. 1395, provides that nothing in Title XVIII shall be construed to authorize any federal officer or employee to exercise any supervision or control over the practice of medicine or the manner in which medical services are provided. Congress deliberately fenced the program away from the clinical encounter. The agency may establish conditions of payment, define what is covered, and set the amounts it will pay, but it may not direct how a physician examines a patient, chooses a therapy, or manages a case. This limitation, and the history of the 1965 enactment that produced it, is examined in the series account of the original legislation, the Social Security Amendments of 1965. It explains a persistent tension in the regulations. The agency writes rules that reach deep into hospital cost accounting, physician work measurement, and the clinical evidence for new technologies, yet every rule must be framed as a payment or coverage determination rather than an instruction to the treating professional. The distinction is legal, not merely rhetorical, and it constrains how far the agency can go in implementing what the statute authorizes.

The scale of administration makes the annual rulemaking necessary. The program pays hundreds of thousands of providers and suppliers through a network of Medicare administrative contractors that process claims, issue local payment policies, and conduct audits within their assigned jurisdictions. Quality improvement organizations review the appropriateness and quality of care. The agency’s responsibilities extend beyond this program: it also administers the joint federal-state program whose state plans operate under an entirely different, state-administered model, described in the companion account of the Medicaid statute, the companion account of the Medicaid statute. A program of this size cannot be run from statutory text alone, and Congress never intended it to be. The statute delegates to the Secretary broad authority to prescribe the regulations necessary to carry out the program, and the agency has exercised that authority in thousands of pages of rules that give concrete meaning to broad statutory phrases. The result is a layered system in which Congress sets the framework, the agency sets the numbers, contractors apply both to individual claims, and the whole edifice turns over on an annual calendar.

Four machines drive the system. The first is the annual payment rule cycle, in which the agency proposes and finalizes the formulas and rates that determine what providers are paid for the coming year. The second is the coverage apparatus, in which the agency decides whether particular items and services meet the statutory standard for payment at all. The third is the sub-regulatory layer of manuals, transmittals, and guidance, in which the agency answers the operational questions that the rules leave open. The fourth is the appeals ladder, in which denials are tested stage by stage, from contractor redetermination through federal court. Both of the first two depend on public processes that invite evidence and argument before the agency decides, and on statutory foundations that converted the program from open ended reimbursement to administered prices between 1983 and 1997. All four leave the clinical decision to the physician, as section 1801 requires, while governing the financial consequences that follow it.

The annual payment rule cycle

The most consequential lawmaking in the program each year is not a statute. It is a set of annual payment rules that the agency publishes in the Federal Register, each following a proposed-then-final calendar tied to the payment year it governs. The inpatient prospective payment system rule for hospital inpatient services is proposed in the spring, with the final rule published by August 1 to take effect for the federal fiscal year that begins October 1. The outpatient prospective payment system rule, which sets rates for hospital outpatient departments, runs on a calendar year cycle, finalized by November 1 for the payment year beginning January 1. The physician fee schedule rule is proposed in the summer and finalized by November 1 for a calendar year payment period beginning January 1. Post-acute care rules follow their own versions of the same pattern: skilled nursing facility rates, home health rates, inpatient rehabilitation facility rates, and long term care hospital rates are each refreshed through an annual rulemaking that updates the base amounts, classification weights, and adjustment factors for the year ahead.

Each of these rules is a substantial document, often running to hundreds of pages in the Federal Register. A single hospital inpatient rule may adjust the standardized amounts, update the wage index that reflects local labor costs, recalibrate the diagnosis-related group weights, modify the outlier payment policy, and introduce new reporting requirements for hospital quality programs. The outpatient rule may reassign thousands of individual services among payment groups and revise the packaging rules that determine which ancillary services are paid separately. The physician fee schedule rule revalues the relative value units assigned to thousands of procedures, resets the conversion factor that turns those units into dollars, and sets the geographic adjustment factors. Because the rules combine so many moving parts, the payment a provider receives in a given year is the product of dozens of agency decisions made in the months before that year began, layered onto statutory formulas that the agency fills in with fresh data.

Congress built the notice and comment requirement for these rules directly into Title XVIII. Section 1871(b) of the Social Security Act, codified at 42 U.S.C. 1395hh(b), requires the Secretary to provide for notice of a proposed rule or other statement of policy that establishes or changes a substantive legal standard governing the scope of benefits, payment for services, or eligibility to furnish or receive services, with an opportunity for public comment for not less than sixty days. The statute carves out rulings, determinations, and other interpretive rules from this requirement, but the annual payment rules are substantive in every sense, and the agency treats them as subject to the sixty day floor. In practice the comment periods often run longer than sixty days, drawing responses from hospital associations, physician specialty societies, device manufacturers, beneficiary advocates, and state agencies, each submitting data and argument about the proposed rates. The sixty day requirement in Title XVIII sits alongside the general notice and comment obligation of the Administrative Procedure Act, and the two together define the legal character of the annual rules. The Act requires notice of a proposed rulemaking and an opportunity for interested persons to participate, and the courts have read that requirement to demand that the agency disclose the data and methodology on which a rule rests so that commenters can respond intelligently. The Title XVIII provision adds a substantive trigger and a minimum period tailored to the program: when the agency establishes or changes a standard governing benefits, payment, or eligibility, the public gets at least sixty days. The combination means the annual payment rules are among the most procedurally protected agency actions in the federal government. They cannot be issued by memorandum, announced by press release, or imposed retroactively without the prescribed process. The discipline is deliberate. When a rule moves billions of dollars among providers, the statute insists that the movement be proposed in public, supported by published reasoning, and exposed to criticism before it becomes final.

The distributional analysis that accompanies each proposed rule is a distinctive feature of this process. The agency publishes tables showing the estimated effect of the proposal on different categories of providers: urban and rural hospitals, teaching and nonteaching hospitals, hospitals of different bed sizes, physician specialties, and post-acute provider types. These tables translate the abstract formulas into concrete consequences, and they are the pages that provider associations study most closely. A proposed change that is budget neutral in the aggregate may still produce significant gains for some categories and losses for others, and the tables make those effects visible. The visibility serves a purpose beyond advocacy. It forces the agency to own the distributional choices embedded in its methodology, and it gives commenters the information they need to argue that a different methodology would produce a fairer distribution. The final rule then explains how the agency weighed those arguments, completing a record that a reviewing court can examine if the rule is challenged.

How does a sentence in Title XVIII become a hospital payment rate?

Congress writes an authorization in Title XVIII, the agency publishes a proposed rule in the Federal Register, the public comments for at least sixty days under 42 U.S.C. 1395hh(b), the agency issues a final rule with the formulas and rates, and codified text at 42 C.F.R. fixes the numbers hospitals use.

The proposed rule is the moment when the agency shows its arithmetic. The preamble to a proposed inpatient rule walks through the data sources, the methodology for recalibrating the diagnosis-related group weights, the proposed standardized amounts, and the distributional analysis showing how the proposal would affect hospitals by region, bed size, and teaching status. Commenters respond with their own analyses, pointing out data errors, questioning methodological choices, and arguing for alternative treatments of particular costs. The agency is obligated to consider these comments, and the final rule responds to them in a comment-and-response section that has become one of the most important sources of the agency’s reasoning. A hospital finance officer who wants to understand why a particular rate moved in a particular year reads the final rule preamble, not the statute, because the preamble explains the choices the statute left to the agency.

The codified text that results from the rule is only part of the final product. The regulatory language that is added or amended in 42 C.F.R. is relatively compact, but the preamble that accompanies it carries the operational detail: the tables of rates, the formulas for adjustments, the explanation of data sources, and the agency’s responses to comments. The preamble is not itself regulation in the strict sense, yet it is the document that tells providers what the agency intends the regulation to mean and how it will be applied. Billing staff, compliance officers, and health lawyers read the preamble as working law, because the distinction between the codified text and the published explanation collapses in practice. The agency enforces the rule as the preamble describes it, contractors apply it as the preamble describes it, and auditors test compliance against it.

The consequence is that far more operative law is made in these annual documents than in any statute enacted in a typical year. Congress may pass a major health bill once in several years, and each such bill may alter the framework: adding a payment system, changing an update formula, or creating a new reporting program. The agency, by contrast, issues dozens of rules every year that set the actual amounts paid, adjust the weights and factors that distribute those amounts, and define the conditions under which payment is made. By the time a fiscal year begins, the statutory formulas have been populated with agency-chosen numbers, and it is those numbers that determine the revenue of hospitals, physicians, and post-acute providers across the country. A reader of the statute alone could not predict what any of them would be paid. A reader of the final rules could predict it almost exactly.

Locating these documents is a matter of knowing the two places where the agency publishes its working law. The Federal Register carries every proposed and final rule, with the preamble and the codified amendments in a single publication, organized by date of issuance. The Code of Federal Regulations, Title 42, carries the codified regulatory text organized by subject, showing the current version of each regulation as amended by successive rulemakings. The Federal Register answers the question of what the agency did in a particular rulemaking and why. The Code of Federal Regulations answers the question of what the regulation says as amended to date. Professionals who work with the program use both, moving between the historical record of rulemaking and the consolidated text of the regulations, and neither source requires anything more than knowing where to look.

The annual cycle also creates a rhythm of advocacy and adaptation that shapes the program continuously. Because the agency must propose rates before it finalizes them, every provider community has a standing opportunity to influence the numbers that will govern its revenue. Specialty societies commission studies to support their arguments about the relative value of their services. Hospital systems model the distributional effects of proposed changes on their facilities. Beneficiary groups and consumer advocates comment on provisions affecting access and cost sharing. The agency weighs this input against its budgetary constraints, its statutory directives, and its own policy goals, and the final rule reflects that weighing. The cycle then begins again, with the next year’s data, the next year’s proposed rule, and the next round of comments. The program’s payment policy is therefore never settled in any permanent sense; it is renegotiated annually through a public process that Congress designed for exactly that purpose.

What the annual rules actually adjust

The annual rules do more than refresh base rates. They recalibrate the machinery that distributes payment across providers, and the details of that machinery reveal how much discretion the statute leaves to the agency. In the inpatient rule, the agency updates the wage index, which adjusts payments for differences in local labor costs. The wage index is built from wage data that hospitals report on their cost reports, and it is revised each year as new data arrive. Hospitals may seek reclassification to a different labor market area when they can show that their labor costs resemble a neighboring area more closely than their own, and the annual rule adjudicates these requests under criteria set out in regulation. The wage index is one of the most contested features of the inpatient system, because small changes in its construction shift revenue among regions, and the comment record on the proposed rule each year carries detailed argument about data sources, occupational mix adjustments, and the treatment of rural floors.

The inpatient rule also maintains the outlier payment policy, which protects hospitals against the financial risk of extraordinarily costly cases. A case qualifies for an outlier payment when its costs exceed a threshold defined by the fixed loss amount plus the diagnosis-related group payment, and the rule sets that threshold each year so that total outlier payments remain near the target share of total inpatient spending established in regulation. Setting the threshold requires the agency to project the distribution of hospital costs for the coming year, and the projection is sensitive to changes in charging practices and cost report data. When the threshold is set too low, outlier spending exceeds the target; when it is set too high, hospitals that treat the costliest cases are undercompensated. The annual rulemaking is where the agency explains its projection methodology and where hospitals argue about its accuracy.

Many of the adjustments the agency makes must be budget neutral, meaning the agency must offset the cost of a policy change with offsetting changes elsewhere so that aggregate spending does not increase. Budget neutrality is a statutory requirement for particular kinds of changes, and it forces the agency to show its arithmetic in the proposed rule: the cost of the proposed change, the source of the offset, and the resulting adjustment factor. The requirement disciplines the rulemaking, because the agency cannot simply add a favorable adjustment without identifying who pays for it, and the distributional analysis in the rule shows the winners and losers explicitly. Commenters respond in kind, proposing alternative offsets and arguing that the agency’s estimates are wrong. The budget neutrality calculation is one of the places where the program’s payment policy most visibly resembles a negotiated settlement.

The agency does not set the annual updates in a vacuum. The Medicare Payment Advisory Commission, an independent body created by the Balanced Budget Act of 1997, reports to Congress each year with recommendations on the payment updates for hospitals, physicians, and post-acute providers. The Commission’s recommendations are advisory, and the agency is not bound by them, but the reports shape the debate by supplying an independent analysis of payment adequacy, access, and quality. The Commission has consistently argued that updates should reflect the efficient provider’s cost of furnishing care, and its recommendations provide a benchmark against which the agency’s proposed updates are measured in the comment process and in Congress.

The annual rules have also become the vehicle for quality reporting programs that link a portion of payment to the submission of quality data. Beginning in the 2000s, the inpatient rule established reporting requirements under which hospitals that fail to submit specified quality measures face a reduction in their annual update. The physician fee schedule rule carries parallel provisions for physician quality reporting. These programs do not yet determine the base rates, but they represent a new use of the rulemaking machinery: the agency conditions the full update on participation in measurement, using the payment system to build the data infrastructure for quality assessment. The statutory authority for these programs comes from later legislation, but their implementation runs entirely through the annual rules, with the measures, the reporting mechanisms, and the penalties specified in the Federal Register.

The physician fee schedule rule has its own distinctive update mechanics. The Balanced Budget Act of 1997 established the sustainable growth rate system, which tied the annual update to the conversion factor to a formula comparing actual physician spending with a target based on economic growth and other factors. When spending exceeded the target, the formula called for reductions in the update; when spending fell short, it called for increases. By 2010 the formula was producing calls for steep reductions that Congress found unacceptable, and Congress intervened year after year with short-term legislation overriding the scheduled cuts. The sustainable growth rate episode illustrates the limits of formula-driven payment policy: a statutory formula can produce results that the political system will not sustain, and the annual rulemaking then becomes the record of the gap between what the formula says and what Congress permits. The agency publishes the update the formula would produce, Congress passes a patch, and the final rule reflects the patched number.

From open-ended reimbursement to administered prices

The annual payment rules described above did not always exist, and the systems they update were each the product of a statutory decision to replace an open-ended standard with a fixed price. Before 1983, hospitals participating in the program were paid on a reasonable cost basis, meaning the agency reimbursed each hospital for the costs it incurred in treating program beneficiaries, subject to principles of cost finding and cost apportionment set out in regulation. The reasonable cost standard had a predictable defect: it gave hospitals little reason to control their costs, because higher costs produced higher payments. Spending on hospital inpatient care grew rapidly through the 1970s, and by the early 1980s the fiscal pressure had become a central concern of health policy. Congress responded with the Social Security Amendments of 1983, Public Law 98-21, which created the inpatient prospective payment system and phased it in during 1983.

The new system paid hospitals a fixed amount for each admission, determined by the diagnosis-related group to which the case was assigned, rather than by the costs the hospital actually incurred. Diagnosis-related groups classify inpatient cases by diagnosis, procedure, age, and other characteristics that predict resource use, and each group carries a weight reflecting its expected cost relative to the average case. The agency multiplied the weight by a standardized amount, adjusted for local wage levels and other factors, to produce the payment for the admission. A hospital that treated a case for less than the fixed payment kept the difference; a hospital that spent more absorbed the loss. The incentive structure was inverted in a single stroke. Hospitals that had been rewarded for higher costs were rewarded for efficiency instead, and the agency acquired a powerful tool for controlling the growth of inpatient spending: it could set the annual update to the standardized amounts and thereby control the rate of increase across the entire system.

The prospective payment system also changed the relationship between the agency and the hospital industry. Under reasonable cost reimbursement, the critical regulatory questions concerned cost accounting: which costs were allowable, how overhead was allocated among departments, and how program costs were separated from nonprogram costs. Under prospective payment, the critical questions concern classification and pricing: whether the diagnosis-related group assignment for a case is correct, whether the weights reflect current patterns of resource use, and whether the standardized amounts keep pace with input costs. The annual inpatient rule became the venue in which these questions were decided, and the comment process became the mechanism through which the industry argued about them. The statute created the system, but the agency has spent every year since refining its machinery.

Physicians were paid under a parallel open-ended standard for the first quarter century of the program. Before 1989, payment for physician services was based on customary, prevailing, and reasonable charges, a methodology that tied the program’s payments to what physicians had historically charged. The customary charge reflected what an individual physician had charged in the past, the prevailing charge reflected what physicians in the same locality typically charged, and the reasonable charge was the lowest of the actual, customary, and prevailing amounts. Like reasonable cost reimbursement for hospitals, the charge-based standard rewarded inflation: physicians who raised their charges could, over time, raise the amounts the program paid them. The result was a payment system that ratified existing billing patterns rather than measuring the resources actually required to furnish services.

The Omnibus Budget Reconciliation Act of 1989, Public Law 101-239, replaced the charge-based methodology with a fee schedule built on the resource-based relative value scale. The new scale measured three components of each service: the work of the physician, the practice expense of furnishing the service, and the malpractice expense associated with it. Each component received a relative value, the three were summed into a total relative value unit, and the unit was multiplied by a conversion factor to produce a dollar amount, with geographic adjustments applied for local cost differences. The fee schedule was implemented in 1992, and it remains the basis for physician payment in the program. The conversion of physician payment from charges to measured resources was the largest single repricing in the history of the program, and its effects redistributed revenue across specialties according to the measured intensity of their work rather than the history of their billing.

Why did Congress replace cost-based payment with fixed prices?

Open-ended cost reimbursement let hospital spending grow without an effective limit, so the Social Security Amendments of 1983 imposed diagnosis-related groups that pay a fixed amount per case, and the Omnibus Budget Reconciliation Act of 1989 extended the same logic to physicians through the resource-based relative value scale.

The intellectual foundation for the physician fee schedule came from a study at the Harvard School of Public Health, led by the economist William Hsiao and begun in 1985 with federal funding. The Hsiao study set out to measure the work involved in thousands of physician services by surveying physicians about the time, mental effort, technical skill, and stress associated with each procedure, then converting those judgments into relative values. The study produced a scale that valued cognitive services, such as evaluation and management visits, more generously relative to procedural services than the charge-based system had, and it supplied the empirical basis on which the agency constructed the work component of the fee schedule. The episode illustrates a recurring pattern in the program’s history: the agency funds or commissions research, the research supplies a measurement methodology, and the methodology becomes the basis for a payment system that redistributes billions of dollars according to the new measure. Payment policy in the program is applied health economics, and the Hsiao study is its most influential example.

The Balanced Budget Act of 1997 extended the prospective payment model to the settings that the 1983 and 1989 reforms had left on cost-based or charge-based systems. The 1997 law created prospective payment systems for skilled nursing facilities, home health agencies, and hospital outpatient departments, and it laid the groundwork for additional systems in the years that followed. Skilled nursing facilities moved to a per diem system based on patient classification groups that reflect expected resource use during the stay. Home health moved to a prospective episode payment for sixty day periods of care. Hospital outpatient departments moved from cost-based payment to the outpatient prospective payment system, which groups services into ambulatory payment classifications with fixed prices. Each conversion followed the same logic as the earlier reforms: an open-ended standard that rewarded higher volume or higher costs was replaced with an administered price that rewarded efficiency, and the agency was given the authority to set and update the price through rulemaking.

The cumulative effect of these conversions is that nearly every major payment system in the program is an administered price system. The agency sets the base rates, the classification weights, the conversion factors, and the adjustment factors through the annual rules, within parameters established by statute. Congress retains the power to change the parameters: it can alter the update formulas, create new adjustments, or establish entirely new payment systems, and it has done so repeatedly. But the day to day determination of what providers are paid is an agency function, carried out through the rulemaking calendar. The reasonable cost era and the customary prevailing reasonable era belong to the program’s history; the administered price era is its present and, as far as the statutory framework indicates, its future. Understanding the program therefore requires understanding not only what the payment systems are, but how the agency builds and maintains them, one rulemaking at a time.

The classification systems at the heart of these payment models have themselves been refined through rulemaking. The original diagnosis-related groups distinguished cases by diagnosis and procedure but captured differences in patient severity only crudely. For fiscal year 2008 the agency adopted Medicare Severity diagnosis-related groups, which split many of the original groups according to the presence of complications, comorbidities, and major complications or comorbidities, producing a finer classification that pays more for sicker patients and less for straightforward cases. The refinement was implemented through the annual inpatient rule, with the agency phasing in the new weights and publishing extensive analysis of the distributional effects. The episode shows how the agency uses the rulemaking calendar to improve the measurement on which payment rests, and how each improvement redistributes revenue among hospitals according to the severity of the patients they treat.

The capital costs of hospitals received parallel treatment. Operating costs moved to prospective payment in 1983, but capital costs, the costs of buildings and equipment, remained on a cost basis for several more years. The Omnibus Budget Reconciliation Act of 1990 directed the agency to establish a prospective payment system for capital, which the agency phased in beginning in fiscal year 1992. The capital system pays a federal rate per discharge, adjusted for local cost differences and other factors, and it has been updated through the annual inpatient rule ever since. The separation of capital from operating payment reflects a policy judgment that investment decisions should face a fixed price signal rather than open-ended reimbursement, and the annual rule is where the agency calibrates that signal.

The physician fee schedule has undergone its own refinements. The work component, built on the Hsiao study, was the first to become resource based. The practice expense component followed: the agency replaced the original charge-based practice expense values with resource-based values developed from surveys of practice costs, phased in over several years beginning in 1999 and completed in 2002. The malpractice component was refined in parallel. The statute also requires the agency to review the work relative values periodically, and the agency has conducted five-year reviews in which specialty societies nominate codes they believe are misvalued and the agency reexamines the underlying work. These reviews are among the most intensely lobbied proceedings in the program, because revaluing a widely billed code moves revenue across specialties, and the budget neutrality requirement means that increases for one specialty must be offset by decreases for others. The fee schedule is therefore not a fixed scale but a periodically rebalanced one, with the agency serving as the arbiter of relative value.

The post-acute systems created by the Balanced Budget Act of 1997 each rest on a patient classification system analogous to the diagnosis-related groups. Skilled nursing facilities are paid a per diem rate adjusted by resource utilization groups, which classify residents according to their clinical characteristics and expected resource use. Home health agencies are paid an episode rate adjusted by home health resource groups, which classify sixty day episodes by clinical, functional, and service utilization characteristics. Inpatient rehabilitation facilities are paid per discharge under a case-mix system based on patient assessment data. Long term care hospitals are paid under a system modeled on the inpatient prospective payment system, with long term care diagnosis-related groups. Each classification system is maintained through rulemaking: the agency recalibrates the weights, updates the base rates, and revises the classification criteria as new data become available. The post-acute rules are less visible than the hospital and physician rules, but they govern a large and growing share of program spending, and their annual updates follow the same proposed-then-final discipline.

Coverage policy and the reasonable and necessary standard

The payment rules determine how much the program pays for services it covers. A separate apparatus determines whether particular items and services are covered at all. The statutory standard for that determination is section 1862(a)(1)(A) of the Social Security Act, codified at 42 U.S.C. 1395y(a)(1)(A), which provides that no payment may be made for any expenses incurred for items or services that are not reasonable and necessary for the diagnosis or treatment of illness or injury or to improve the functioning of a malformed body member. The phrase reasonable and necessary is the most litigated standard in the program, and also the most consequential for beneficiaries, because it is the gate through which every new drug, device, procedure, and diagnostic test must pass before the program will pay for it.

The standard is deliberately general. Congress did not define reasonable and necessary in the statute, and it did not specify what evidence the agency must consider in applying the phrase to a particular technology. The generality is both a strength and a source of controversy. It gives the agency the flexibility to evaluate new technologies on their merits, using clinical evidence that did not exist when the statute was written. It also gives the agency wide discretion, because reasonable and necessary can be read to demand different levels of evidence depending on how the reader understands the words. A technology that is reasonable in the sense that it might help a patient is not necessarily reasonable in the sense that its benefits justify its costs, and the statute does not say which reading the agency should adopt. The agency has interpreted the standard primarily through evidence of clinical effectiveness, asking whether the item or service improves health outcomes for program beneficiaries, rather than through explicit consideration of cost. That interpretation has been contested by commentators who argue that reasonableness must include some assessment of value, and defended by others who argue that the statute directs the agency to consider only whether the service works. The debate continues, and the agency’s practice reflects a cautious approach: coverage decisions rest on published clinical evidence, technology assessments, and the judgment of medical experts, without a formal cost effectiveness threshold.

Coverage policy operates at two levels: national and local. National coverage determinations are made by the Secretary and apply uniformly across the program. The statutory process for national coverage determinations is set out in section 1862(l) of the Social Security Act, codified at 42 U.S.C. 1395y(l), which was added by the Medicare, Medicaid, and SCHIP Benefits Improvement and Protection Act of 2000. The 2000 law responded to longstanding complaints that the national coverage process was slow, opaque, and unpredictable, and it imposed deadlines and procedural requirements designed to make the process more transparent and more responsive to requests from manufacturers, providers, and beneficiaries.

Under the statutory process, the agency must publish a proposed national coverage determination within six months of receiving a formal request, extended to nine months when the agency commissions an external technology assessment or refers the matter to the Medicare Evidence Development and Coverage Advisory Committee. The proposed decision is published with an explanation of the evidence the agency considered and the reasoning that supports the proposed outcome. The agency then provides thirty days for public comment on the proposed decision, a period in which interested parties submit additional evidence, critique the agency’s analysis, and argue for a different result. After considering the comments, the agency issues a final determination that states the coverage decision and the conditions, if any, under which the item or service will be covered. The deadlines are statutory, which means the agency cannot defer a decision indefinitely, and the public comment requirement means the decision is made on a public record that explains what evidence was considered and why.

The outcomes of national coverage determinations take several forms. A determination may cover an item or service for all beneficiaries, deny coverage for all beneficiaries, or cover it under defined conditions, such as for particular diagnoses, in particular settings, or when furnished by particular types of providers. The agency may also issue a determination that coverage is appropriate only with evidence development, meaning the item or service is covered when furnished in the context of an approved clinical study or registry that collects data on its effectiveness. This last category reflects the agency’s effort to balance access to promising technologies with the need for better evidence: beneficiaries can receive the service, but only in settings that generate the data the agency needs to make a firmer decision later. The conditional forms of coverage are the most common, because few technologies present evidence that supports unrestricted coverage or outright denial, and the agency prefers to narrow its decisions to the populations and circumstances for which the evidence is strongest.

How does a national coverage determination bind the whole payment system?

An NCD is issued by the Secretary under section 1862(l) after a formal request, a proposed decision with six or nine month deadlines, and thirty days of public comment; once final, 42 C.F.R. 405.1060 makes it binding on contractors, quality improvement organizations, qualified independent contractors, administrative law judges, and the Appeals Council.

The binding effect is unusually strong. National coverage determinations bind all Medicare contractors, including the Medicare administrative contractors that process claims and issue local policies. They bind quality improvement organizations and qualified independent contractors, the entities that review claims for medical necessity. They bind administrative law judges who hear appeals of denied claims, and they bind the Medicare Appeals Council, the highest level of administrative appeal within the Department. An administrative law judge may not disregard a national coverage determination or set it aside, even if the judge believes the determination is wrong. The judge’s authority is limited to deciding whether the determination applies to the particular claim under review: whether the beneficiary’s condition, the provider’s documentation, and the circumstances of the service fit within the terms of the determination. If the determination applies, the judge must follow it. This rule makes the national coverage determination the most powerful instrument in the coverage apparatus, because it removes the service from the discretion of every adjudicator in the system and settles the coverage question nationally and finally, subject only to the review mechanisms described below.

The strength of the binding effect explains why the procedural requirements matter so much. Because a national coverage determination forecloses contrary decisions by contractors and adjudicators across the country, the statute requires that it be made through a public process with published reasoning, evidence review, and an opportunity for comment. A manufacturer whose device is denied coverage nationally has no alternative forum in which to relitigate the coverage question claim by claim; the determination controls every claim. The process is therefore the protection. The agency must show its work, consider the evidence submitted by interested parties, and explain its conclusion in a published decision. Aggrieved parties who believe the determination is invalid may seek review by the Departmental Appeals Board under the procedures at 42 C.F.R. Part 426, which provides a forum for challenging the validity of a national coverage determination outside the individual claims appeal process. The Board review examines whether the determination was properly issued, and it gives affected parties a path to challenge the determination itself rather than merely arguing about its application to a single claim.

The national coverage process also illustrates the relationship between the agency and the medical evidence base. The agency does not generate clinical evidence; it evaluates evidence generated by researchers, manufacturers, and professional societies. The quality of a coverage decision therefore depends on the quality of the available evidence, and the agency’s decisions often reveal the gaps in what is known about a technology’s effectiveness in the program’s beneficiary population. Technologies studied primarily in younger populations may lack evidence for the aged and disabled beneficiaries the program serves. Technologies with strong evidence for surrogate outcomes may lack evidence for outcomes that matter to patients, such as survival or functional status. The agency’s coverage memoranda discuss these gaps explicitly, and the conditions attached to coverage often reflect the limits of the evidence: coverage for the population studied, in the settings studied, with the follow up the studies measured. In this way the coverage process functions as a form of technology assessment, translating the medical literature into payment policy one determination at a time.

The evidence on which the agency relies comes from several institutional sources. The Agency for Healthcare Research and Quality conducts technology assessments at the agency’s request, producing systematic reviews of the clinical literature on the effectiveness of particular interventions. These assessments synthesize the published studies, grade the strength of the evidence, and identify the gaps, and the agency’s coverage memoranda draw on them heavily when they are available. The Medicare Evidence Development and Coverage Advisory Committee, a panel of clinicians, researchers, and other experts, advises the agency on the strength of the evidence for particular technologies and on the questions that further research should answer. The committee’s meetings are public, its deliberations are recorded, and its recommendations become part of the record on which the coverage decision rests. Manufacturers and professional societies submit their own evidence syntheses during the comment periods, and the agency’s final memoranda respond to the major submissions, explaining which studies the agency found persuasive and why.

The agency has developed a middle path for technologies whose evidence is promising but incomplete: coverage with evidence development. Under this approach, the agency covers the item or service on the condition that it is furnished in the context of an approved clinical study or a registry that collects specified data on patient outcomes. The coverage decision states the research questions the data must answer, the standards the study or registry must meet, and the conditions under which individual beneficiaries qualify. Data accumulate through the covered use, and the agency may later reconsider the determination in light of the new evidence, either broadening coverage, narrowing it, or withdrawing it. Coverage with evidence development reflects a view of the coverage process as iterative rather than final: the agency makes the best decision the current evidence supports, creates the conditions for better evidence, and revisits the decision when the evidence improves. It also reflects the practical reality that an outright denial may deprive beneficiaries of a beneficial technology while an unrestricted approval may commit the program to paying for a technology that does not work. The conditional form splits the difference, and its use has grown as the technologies coming before the agency have become more complex and more expensive.

The coverage memoranda the agency publishes for each national determination are themselves a significant body of reasoning. A memorandum reviews the technology, summarizes the evidence, discusses the comments received, and explains the agency’s conclusion in detail. The memoranda are longer and more discursive than the regulatory text of the payment rules, because they must engage with clinical science rather than accounting methodology, and they are read closely by manufacturers planning their evidence strategies, by professional societies developing clinical guidelines, and by contractors writing local determinations on related technologies. A well-reasoned memorandum can shape the development of evidence for years, by signaling to researchers what kinds of studies the agency will find persuasive in a future reconsideration.

Local coverage determinations and regional variation

Where no national coverage determination exists, coverage policy is made closer to the ground. The Medicare administrative contractors that process claims in each jurisdiction also issue local coverage determinations, which state whether and under what conditions particular items and services are covered for the beneficiaries in that contractor’s territory. Local coverage determinations apply the same statutory standard, reasonable and necessary under section 1862(a)(1)(A), but they do so through the judgment of the contractor’s medical directors, informed by the evidence available to the contractor and the patterns of practice in the region. The result is that coverage for the same service can differ from one part of the country to another, not because the statute differs, but because the contractors charged with applying it reach different conclusions.

The local coverage process follows its own notice and comment procedures, though less formal than the national process. A contractor developing a local coverage determination typically publishes a draft, holds a comment period, and consults with a contractor advisory committee composed of physicians from the jurisdiction who provide input on the clinical questions the draft raises. The contractor then issues a final determination with a summary of the comments received and the contractor’s responses. The determination sets out the covered indications, the documentation requirements, the coding guidance, and any limitations the contractor imposes. Providers in the jurisdiction are expected to know the local determinations that apply to the services they furnish, because claims that do not satisfy the stated conditions will be denied, and repeated denials can trigger audits and overpayment demands.

The reasons for regional variation are structural. Contractors are independent organizations with their own medical staff, their own review of the evidence, and their own experience with the providers in their jurisdictions. Two contractors examining the same body of literature may weigh it differently: one may find the evidence sufficient to support coverage for a broad set of indications, while the other may find it sufficient only for a narrow set, or insufficient altogether. Local practice patterns also matter. A service that is well established and routinely furnished with good outcomes in one region may be new and unproven in another, and the contractor’s medical directors will reflect that difference in their judgments. The reasonable and necessary standard is capacious enough to accommodate these differences, because it does not specify the quantum of evidence required or the method for weighing conflicting studies. Variation is therefore not a malfunction of the local system; it is the predictable product of decentralized decisionmaking under a general standard.

Why can coverage for the same service differ from one region to another?

Where no national coverage determination exists, each Medicare administrative contractor writes local coverage determinations for its own jurisdiction, applying the reasonable and necessary standard to local medical evidence; because contractors weigh evidence differently, a service covered routinely in one jurisdiction may face restrictions or denial in another.

The practical consequences of variation fall on providers and beneficiaries. A physician who practices in two jurisdictions, or a hospital system that operates facilities across contractor boundaries, must track the local determinations in each territory and conform its documentation and coding to each. A service that is covered without special documentation in one jurisdiction may require detailed records of failed conservative therapy in another. A supplier of durable medical equipment may find that the same device is covered under one contractor’s determination and denied under another’s, with no national determination to resolve the conflict. Beneficiaries, for their part, may discover that a treatment their physician recommends is covered when they live in one state and not covered when they move to another, a result that is difficult to explain to a patient who understands the program as a single national benefit. The program’s national character, which is so evident in the payment rules, gives way in coverage to a patchwork that reflects the judgment of the contractors Congress assigned to administer it.

The relationship between national and local determinations is hierarchical. A local coverage determination cannot contradict a national coverage determination; where the Secretary has spoken nationally, the contractors must follow. Local determinations operate only in the space the national determinations leave open, which is considerable, because the agency issues national determinations for only a fraction of the services the program pays for. The national process is resource intensive, with its statutory deadlines, evidence reviews, and advisory committee consultations, and the agency reserves it for technologies that raise significant questions of effectiveness, safety, or program cost. The vast majority of coverage questions never reach the national level. They are decided by contractors, claim by claim and determination by determination, in the ordinary course of claims administration. The local layer is therefore not a minor supplement to the national process; it is where most coverage policy is actually made.

The agency oversees the local determinations without fully standardizing them. Contractors must follow the agency’s instructions on the format and process for local determinations, and the agency reviews new and revised determinations for consistency with national policy. But the agency has generally allowed contractors to retain their discretion on the substance of coverage, accepting variation as the price of a system that can respond to local evidence and local practice. Proposals to consolidate or nationalize the local process have been debated for years, with supporters arguing that a single national standard would be fairer and more predictable, and opponents arguing that it would be slower, less responsive, and more vulnerable to the bottlenecks that already constrain the national process. The debate reflects a deeper question about the program’s design: whether uniformity or responsiveness is the more important value in coverage policy, and how much variation a national program can tolerate before it ceases to feel national.

Aggrieved parties have avenues for challenging coverage decisions at both levels. As noted above, the validity of a national coverage determination may be reviewed by the Departmental Appeals Board under 42 C.F.R. Part 426, a process that examines the determination itself rather than any individual claim. Local coverage determinations are subject to a parallel reconsideration process in which interested parties may ask the contractor to reconsider the determination, and the agency provides for review of the contractor’s decision. These review mechanisms do not eliminate variation, but they give affected parties a forum in which to argue that a determination is unsupported by the evidence or inconsistent with the statute. The existence of review is itself a check on the discretion the system grants to contractors, and the published decisions that result from review contribute to the body of reasoning that shapes future determinations.

How the four machines fit together

The payment rules and the coverage determinations are often described separately, but in practice they operate as a single system. A service must first be covered before it can be paid, and the coverage determination defines the population and the circumstances for which payment is available. The payment rule then determines the amount paid for the covered service, through the classification systems, weights, and rates that the annual rulemaking sets. A change in either machine alters the economics of the service. A national coverage determination that expands coverage to a new indication increases the volume of paid claims; a payment rule that revalues the relative value units for the service changes the revenue each claim produces. Providers and manufacturers understand this interaction, and they direct their advocacy accordingly, pressing the coverage process for broader indications and the payment process for higher values.

The section 1801 boundary runs through both machines. The agency sets payment amounts and coverage criteria, but it does not direct the practice of medicine, and the distinction shapes the form of every rule and determination. A payment rule may create strong financial incentives for one treatment over another, but it may not order the physician to choose the favored treatment. A coverage determination may decline to pay for a service the agency finds unsupported by the evidence, but it may not forbid the physician from furnishing it; the beneficiary may still receive the service, paying for it through other means. The agency governs the program’s money, not the physician’s judgment, and the regulations are drafted with that limitation in mind. The limitation does not make the agency’s decisions less consequential. Financial incentives and coverage conditions shape medical practice powerfully, even when they stop short of command. But the legal form matters, because it is the form Congress required, and it defines the outer limit of what the agency may do in implementing the statute.

The thesis with which this article began can be stated in full. The statute authorizes, and the agency operates. Congress wrote the eligibility rules, the benefit categories, the payment system frameworks, and the coverage standard, and it delegated to the Secretary the authority to fill in the working detail. The agency fills it in through the annual payment rules, which set the rates and update the classification systems; through the national coverage determinations, which decide the hardest coverage questions on a public record; through the network of contractors, which decide the rest; and through the sub-regulatory guidance and the appeals ladder that the second half of this article examines. The Federal Register and the Code of Federal Regulations, Title 42, are where this working law is published, and they are the sources a professional consults to know what the program actually does in a given year. The statute is the beginning of the inquiry, not the end of it.

For the reader who works with the program, the practical lesson is to read in the right order. Begin with the statute to understand the framework and the delegation. Move to the annual rules to understand the payment amounts and the agency’s current reasoning. Consult the national and local coverage determinations to understand whether the service is covered and under what conditions. And remember that each of these layers turns over on its own schedule: the rules annually, the coverage determinations as the evidence and the requests require. The program is not a fixed edifice but a moving system, governed by documents that are revised, commented upon, and revised again. The gap between what the statute authorizes and what is actually done is not a failure of the system. It is the system, working as Congress designed it, with the agency making the operative law in public, one rulemaking at a time.

The third machine: the sub-regulatory layer

Beneath the notice and comment rules published in the Federal Register sits the layer where most day-to-day administration of the program actually happens. Rulemaking establishes the large standards: payment methodologies, conditions of participation, benefit definitions. But the daily work of deciding claims, instructing contractors, and answering the operational questions that a payment system of this scale generates every week flows through manuals, transmittals, change requests, and program guidance. A practitioner who reads only the Code of Federal Regulations understands the architecture of the program. A practitioner who reads the manuals understands how the building is operated, who gives the orders, and where a claim actually lives or dies.

The Medicare Benefit Policy Manual states what the benefit covers and under what conditions. Its chapters track the benefit categories Congress created in Title XVIII: inpatient hospital services, skilled nursing facility care, home health, hospice, physician services, durable medical equipment, and the preventive benefits added over the years. When a coverage question turns on whether a service falls within a statutory benefit category, or on the conditions Congress attached to that category, contractors and providers consult this manual first. It does not set payment rates. It maps the boundaries of the benefit, translating broad statutory phrases into the operational criteria that claims reviewers apply to individual cases.

The Medicare Claims Processing Manual is the companion volume for how services are billed, coded, edited, and paid. Organized largely by provider and supplier type, it instructs contractors on claim submission requirements, coding conventions, the automated edits that screen claims before payment, remittance advice, and coordination of benefits. If the Benefit Policy Manual answers whether a service can be covered, the Claims Processing Manual answers how the claim for it must be constructed to survive the contractor’s systems. Much of what providers experience as the working rules of billing lives here rather than in regulation: which modifiers to append, how units are counted, which revenue codes belong on which bill types, and what documentation the claim must reference.

The Medicare Program Integrity Manual governs the defensive side of the operation: medical review, benefit integrity investigations, audits, and the recovery of overpayments. It tells contractors how to select claims for review, what standards reviewers apply when they examine medical necessity or coding, how statistical sampling may support extrapolated overpayment demands, and what process the agency follows when it seeks to recover money already paid. For providers, this manual often matters more than any other, because it controls the audits that can convert years of paid claims into a single demand letter. Its provisions sit at the center of many disputes about whether the agency altered the governing standard after the services were furnished.

CMS revises these manuals through numbered transmittals, each one amending specific sections and carrying an implementation date that tells contractors when the new language takes effect. When a transmittal requires contractors to change their claims processing systems, it travels with a change request, the agency’s formal directive specifying what to build, what to test, and when to activate it. Around these documents sits a wider penumbra of program guidance: MLN Matters articles that explain new policies to the provider community in plain language, frequently asked questions, fact sheets, and joint signature memoranda from agency leadership. None of these documents passes through notice and comment. All of them shape how claims are decided, and contractors treat them as binding instructions under their contracts with the agency.

The agency leaned on this layer heavily for reasons of speed and scale. Full rulemaking, from proposed rule through comment review to final rule, commonly consumes a year or more, and the payment system cannot wait that long for every operational answer. Coding systems update on fixed cycles, new services and technologies enter clinical practice continuously, contractors encounter fact patterns the regulations never contemplated, and Congress sometimes sets implementation deadlines shorter than any rulemaking could satisfy. Against that pressure, a transmittal that revises a manual chapter in a matter of weeks is an irresistible instrument. By 2011 the body of sub-regulatory instruction rivaled the regulations in practical importance: providers built their billing operations around it, contractors denied claims under it, and auditors measured compliance against it.

A well-drafted manual section wears its pedigree openly, citing the statute and the regulation it implements, and the careful reader can trace the chain of authority from the transmittal back through the manual to the Federal Register and ultimately to Title XVIII. That traceability is the layer’s claim to legitimacy: the manual is not inventing policy but recording the agency’s understanding of policy made elsewhere. The difficulty is that not every section carries its pedigree so clearly, and decades of accretion have produced manual language whose connection to any specific regulatory text is attenuated. When the chain of authority is explicit, classification is easy. When it is assumed rather than shown, the document invites the challenge that it is making law rather than explaining it.

The change request deserves attention as the most operational of these instruments, because it is the document that actually moves contractor systems. A change request originates inside CMS when a policy decision, whether made by regulation or by manual revision, requires a systems change: a new edit, a revised payment calculation, an altered reason code. It specifies business requirements, testing expectations, and an implementation date, and contractors build to it. Providers rarely read change requests directly, but they feel them immediately when claims begin denying under a new edit on the implementation date. The change request is thus the point where policy becomes software, and where an abstract dispute about a manual’s meaning becomes a concrete denial on a remittance advice.

The hierarchy among these sources is strict even when it is invisible. A manual section cannot contradict the regulation it implements, a transmittal cannot contradict the manual, and program guidance cannot contradict any of them. When conflicts arise, the higher authority controls, and contractors are instructed to apply the regulation over the manual where the two point in different directions. This ordering is what makes the classification dispute intelligible: the question is never whether the manual outranks the regulation, but whether the manual faithfully carries out the regulation or quietly rewrites it. Practitioners who find a genuine conflict between a manual instruction and its regulatory source have found something more valuable than a classification argument; they have found a direct inconsistency that the higher authority resolves in their favor.

MLN Matters articles illustrate the layer’s softer edge. Written by the agency’s provider education staff, they translate regulatory and manual changes into plain language summaries with examples, and they carry prominent disclaimers that they are not a substitute for the underlying authority. Providers rely on them anyway, because they are readable in a way that transmittals are not. The disclaimers matter legally: when a summary and a manual section point in different directions, the manual controls, and the article’s educational character does not convert it into a binding standard. But the articles reveal something important about the layer as a whole, which is that much of it exists to teach rather than to command, and the interpretive-versus-substantive question is sharpest where teaching shades into commanding.

Below even the manuals sits the softest layer of all: the informal guidance that never receives a transmittal number. Joint signature memoranda from agency leadership announce policy directions to contractors, regional office emails answer questions, medical directors discuss review approaches on contractor calls, and help-desk responses resolve individual billing puzzles. None of this is published in any formal sense, yet it shapes contractor behavior in the aggregate, because contractors calibrate their reviews to what they understand the agency to expect. This layer rarely appears in litigation, precisely because it is hard to cite and harder to prove, but it explains why two contractors can apply the same manual differently: the informal signals each receives are not identical.

What guidance can CMS issue without notice and comment?

CMS may issue manuals, transmittals, change requests, and other program guidance that interpret existing law without notice and comment. But under 42 U.S.C. 1395hh, any rule, requirement, or policy statement that establishes or changes a substantive legal standard governing benefits, payment, or eligibility must be promulgated by regulation, except for national coverage determinations.

The answer draws the boundary where the statute draws it, and the practical meaning of that boundary deserves emphasis. Interpretation, in this context, is the agency’s explanation of what existing law requires: a manual section that restates a regulatory standard in operational detail, a transmittal that applies a payment regulation to a newly introduced procedure code, a set of questions and answers that clarifies how contractors should handle a recurring fact pattern. These documents add no obligation that the underlying law did not already impose; they reduce uncertainty about obligations that already exist. That is why they may issue without public comment. The regulated community is entitled to know the agency’s reading of the law, and notice and comment would add little where the law itself has already been through the process.

The subtlety is that documents without legal force can still determine outcomes. A contractor that must follow manual instructions under its contract will deny the claim that the manual says to deny, and the provider experiences that denial as law regardless of the document’s formal status. This practical binding effect is what makes the classification question so contested: the agency points to the document’s formal character as interpretation, while the affected party points to its practical character as command. Both descriptions can be accurate at once, which is why courts look past labels to function, asking whether the document genuinely explains existing law or effectively creates new law under the guise of explanation.

Congress confronted this tension directly. In the Omnibus Budget Reconciliation Act of 1986, it added a Medicare-specific rulemaking provision, codified at 42 U.S.C. 1395hh, that reaches further than the general requirements of administrative law and speaks explicitly to the sub-regulatory layer. The provision reflects a congressional judgment formed during the program’s first two decades: that the agency’s power over benefits, payment, and eligibility was too consequential to be exercised entirely through manuals, and that the public, including the providers and beneficiaries whose money and care were at stake, deserved a formal voice in the standards that governed them.

Subsection (a)(2) provides that no rule, requirement, or other statement of policy (other than a national coverage determination) that establishes or changes a substantive legal standard governing the scope of benefits, the payment for services, or the eligibility of individuals, entities, or organizations to furnish or receive services or benefits shall take effect unless promulgated by the Secretary by regulation. The provision fences three domains where the agency’s power is greatest and the temptation to govern by manual is strongest: what the benefit includes, how services are paid, and who may furnish or receive them. Within those domains, a substantive legal standard must travel through regulation. The parenthetical exception for national coverage determinations reflects a separate statutory scheme for coverage policy, not a license to smuggle payment or eligibility standards into coverage documents.

Subsection (b) then sets the procedure: Federal Register notice and a public comment period of not less than 60 days before a final regulation issues. The 60-day floor is deliberate, signaling how seriously Congress took public participation in a program that touches nearly every provider and a large share of the population. The subsection recognizes two situations in which the full sequence may bend: where a statute affirmatively permits the agency to proceed in interim final form, and where a statute sets a deadline too short for the ordinary sequence. These are narrow allowances for congressional direction, not general grants of speed. When Congress has not spoken, the notice and the 60 days are the price of a binding substantive standard.

Subsection (e) addresses the dimension of time. It bars retroactive application of a substantive change in regulations, manual instructions, interpretative rules, statements of policy, or guidelines of general applicability. The list matters because it names the sub-regulatory layer explicitly: manual instructions appear alongside regulations as documents whose substantive changes cannot reach backward. A provider that furnished services in reliance on the manual language then in effect cannot be judged later under a rewritten manual section, at least not where the change is substantive. The provision protects reliance interests and forces the agency to announce new standards prospectively, which is the minimum condition of fair notice in a program where billing decisions are made daily against written instructions.

Taken together, the three subsections form a coherent theory of legitimate administration. Substantive standards in the three sensitive domains must be made by regulation, under subsection (a)(2). Regulations must be made with notice and at least 60 days of public comment, under subsection (b). And substantive changes, wherever they appear, must apply prospectively, under subsection (e). The statute thus draws a circle around the sub-regulatory layer: manuals and guidance may explain, interpret, and implement, but they may not create new substantive obligations in the domains Congress fenced off, and they may not rewrite history. Everything that follows in the long-running dispute over manuals is an argument about where that circle’s boundary lies.

The design of 1395hh reflects a particular theory of the program’s politics. Congress understood that providers and beneficiaries, unlike the agency, cannot issue their own instructions; their only protection against being governed by unpublished policy is a guaranteed opportunity to comment before the policy becomes binding. The 60-day floor, longer than the periods general administrative law typically produces in practice, gives national associations time to analyze proposals, survey members, and submit data the agency may not have. The Federal Register requirement ensures the proposal is findable. Together they convert rulemaking from a bureaucratic formality into the program’s central forum for contesting payment policy, which is why the fight over what must go through that forum has never ended.

The layer’s size by 2011 reflected decades of accumulation rather than any single design. Each generation of administrators added instructions to address new benefits, new payment systems, and new integrity initiatives, while old instructions were revised piecemeal rather than retired wholesale. Transmittals were numbered in the thousands, manual chapters ran to dozens, and the full corpus constituted the program’s institutional memory: the record of how countless individual policy questions had actually been answered. This accretion is both the layer’s strength and its vulnerability. It means contractors rarely face a question no one has addressed, but it also means the corpus contains provisions whose original justification has been forgotten, whose regulatory anchor has eroded, and whose current application no one at the agency could defend as mere interpretation with a straight face.

Providers learned to monitor the layer with the seriousness it deserved. Trade associations maintain transmittal-tracking services, consultants publish summaries within days of issuance, and large provider organizations assign staff to review each transmittal for changes affecting their billing. The agency facilitated this surveillance through electronic mailing lists and a public transmittal archive, recognizing that instructions cannot fairly bind parties who cannot find them. This ecosystem of monitoring is itself evidence of the layer’s practical authority: no industry builds an entire information infrastructure around documents that do not matter.

Publication is the often-overlooked companion to procedure. Regulations appear in the Federal Register, the government’s official daily record, where they are indexed, searchable, and citable by date and page. Manuals and transmittals appear on the agency’s website, organized by manual and number, findable to those who know the system and obscure to those who do not. The difference matters for reliance: a provider can fairly be held to a standard published in the Federal Register after 60 days of public comment in a way that strains fairness when the standard arrived via an obscure transmittal. Section 1395hh’s insistence on Federal Register notice is, among other things, an insistence on visibility, a requirement that the standards governing benefits, payment, and eligibility be made where the governed can find them.

Drawing the line: interpretive guidance or substantive rule

Every argument about the sub-regulatory layer eventually reduces to a single classification question: is the document interpretive or substantive. The statute makes the distinction decisive. An interpretive document explains or applies a standard already established in law, and it may issue through a transmittal without public comment. A substantive document establishes or changes a legal standard in one of the three fenced domains, and under 1395hh(a)(2) it must go through notice and comment rulemaking or it cannot take effect. The classification determines the procedure required, the document’s legal force, and whether a denial resting on it will survive challenge. It is the most consequential recurring dispute in program administration, and it has no mechanical answer.

How do you tell an interpretive instruction from a substantive rule?

An instruction is interpretive when it explains or applies a standard found in statute or regulation, adding no new obligation a reader of the law could not anticipate. It is substantive when it creates a duty, condition, or eligibility test that changes what providers or beneficiaries must do to receive payment, which is the line 42 U.S.C. 1395hh(a)(2) polices.

The answer states the working test that practitioners apply, and each of its elements carries weight. “Explains or applies” captures the legitimate office of guidance: a regulation governing payment for medically necessary services will inevitably require instructions about what documentation demonstrates necessity, and a manual that supplies those instructions is interpreting, not legislating. “A reader of the law could not anticipate” captures the reliance interest: if the obligation would surprise someone who studied the statute and the regulation with care, the document has likely crossed from explanation into creation. The test is functional rather than formal. It asks what the document does, not what the agency calls it.

General administrative law supplies the background against which 1395hh operates. Under the Administrative Procedure Act, legislative rules, which create rights and obligations, require notice and comment, while interpretive rules, which explain existing law, do not. Decades of case law have wrestled with the boundary, producing multi-factor tests that weigh a document’s language, its practical effect, and the agency’s own characterization. Section 1395hh overlays this framework with a program-specific command that is in some respects sharper: in the three fenced domains, the question is not merely whether the document interprets, but whether it establishes or changes a substantive legal standard, a formulation Congress chose because it had watched the agency do exactly that through manuals. The APA tests remain relevant, but the statute’s text is the starting point, and practitioners brief the statutory language first.

The agency, for its part, sometimes blurred the line voluntarily by soliciting comment on significant guidance even where the statute did not require it. Draft manual revisions on consequential topics were occasionally posted for public input, and the agency maintained channels for providers to question guidance and receive clarification. These practices served the agency’s own interests: comment improved the quality of instructions, surfaced operational problems before implementation, and built a record of reasonableness that helped defend the guidance later. But voluntary comment is not the statutory right that 1395hh(b) guarantees, and it can be withdrawn as easily as it is offered. Practitioners welcomed the practice while recognizing its limits: a comment opportunity the agency grants as a courtesy is not the same as a comment right Congress commands.

The agency’s strongest case begins with the arithmetic of administration. The program processes an immense volume of claims across thousands of provider types, and the regulations cannot possibly address every coding question, documentation scenario, or coverage fact pattern that contractors encounter. Congress legislates at altitude: it defines benefit categories, directs the Secretary to establish payment methodologies, and sets eligibility criteria in broad strokes. Someone must translate those strokes into the claim-level instructions that make payment possible, and that translation cannot wait for a rulemaking cycle each time a new question arises. Manuals are the translation mechanism, and to demand notice and comment for every clarification would be to demand that administration stall while the Federal Register catches up with clinical practice.

The agency’s case also draws on the long pedigree of interpretive documents in administrative law. Agencies have always issued manuals, rulings, and guidance to tell regulated parties how they read their own regulations, and courts have traditionally permitted such documents without notice and comment precisely because they claim no independent legal force. A transmittal that applies an existing payment regulation to a newly introduced procedure code, the agency argues, adds nothing to the law; it states what the regulation has always meant as applied to new facts. On this view, challengers confuse the burden of learning new instructions with the creation of new obligations, and 1395hh(a)(2) was aimed at genuine substantive policymaking, not at the everyday work of interpretation without which the regulations would be inert.

The challengers’ strongest case begins with the provider’s experience. Manual provisions function as binding law whatever the agency calls them: contractors deny claims citing manual sections, auditors extrapolate overpayments from manual standards, and no provider can safely disregard a transmittal on the theory that it is merely interpretive. When a manual section introduces a documentation requirement that appears nowhere in the regulation, and claims are denied for failing it, the distinction between interpretation and legislation collapses in practice. The challengers argue that 1395hh(a)(2) was written for exactly this situation: Congress watched the agency govern the program’s most sensitive domains through manuals and insisted that substantive standards in those domains go through public rulemaking, with the 60-day comment period as the public’s seat at the table.

The challengers’ second point is structural. If the agency’s label controlled the classification, subsection (a)(2) would be a dead letter, because no agency would ever label its own document substantive and thereby trigger the rulemaking obligation it seeks to avoid. The statute must therefore invite independent judicial assessment of what a document does, not deference to what the agency calls it. The comment right that subsection (b) guarantees is meaningless if the operative decisions migrate to transmittals that no one outside the agency sees until they take effect. And subsection (e)’s bar on retroactive manual changes proves, the challengers note, that Congress understood manuals to be potent instruments of policy rather than mere restatements of settled law; a provision guarding against retroactive interpretive documents would have been unnecessary.

The allocation of argumentative burden in these disputes follows the classification itself. The agency typically begins by characterizing the document as interpretive and pointing to the regulatory text it implements, which shifts the practical burden to the challenger to demonstrate the gap between the document and its anchor. The challenger meets that burden, when it can, by placing the two texts side by side and showing that the document adds an obligation the regulation does not contain. Courts then decide how much weight to give the agency’s characterization, and the honest answer is that the weight varies: some opinions treat the label as nearly dispositive, others as nearly irrelevant, and most as one factor among several. The variability is another reason the dispute never settles, since each side can find support for its preferred approach.

Courts confronted with these competing positions reached varying conclusions, and the variation itself is instructive. Some manual provisions were upheld as genuine interpretation, closely tethered to regulatory language that a careful reader could have anticipated. Others were held to impose new duties, conditions, or eligibility tests that the underlying regulation did not contain, and therefore to require notice and comment before they could take effect. Outcomes turned on the specific document: how far its language traveled from its regulatory anchor, whether it introduced obligations a regulated party could not have foreseen, and whether the agency applied it as binding in practice. No single formula emerged, because the question is irreducibly fact-intensive. A single manual could contain both kinds of provisions, interpretive in one section and substantive in the next, which is why practitioners analyze documents section by section rather than manual by manual.

A hypothetical illustrates the test in motion. Suppose a regulation provides that the program pays for durable medical equipment that is medically necessary, and a new manual section requires that claims for a particular device include a physician’s written order containing five specified elements, with claims lacking any element denied. The agency would defend the section as interpretive: the regulation demands medical necessity, the order documents it, and the elements merely specify what a complete order contains. The challenger would attack it as substantive: the regulation says nothing about five elements, a provider reading the regulation would not anticipate the requirement, and claims are denied for failing an obligation the manual invented. The outcome would turn on whether the five elements genuinely elaborate the regulatory standard or effectively create a new condition of payment, which is precisely the inquiry 1395hh(a)(2) demands.

The verdict for the practitioner is direct: learn to classify documents on this axis, because the classification predicts most of the litigation the layer generates. Before arguing about what a manual section means, ask what it is. Trace its authority from the transmittal through the manual section to the regulation and the statute, and test each link. Ask whether a provider who read only those underlying sources would have anticipated the requirement. Ask whether contractors apply the document as binding. A document that fails these inquiries is vulnerable under 1395hh(a)(2) regardless of the heading it carries, and a document that passes them will likely survive regardless of how aggressively it is challenged. The lawyers who win manual disputes are usually the ones who classified the document correctly before they briefed it, because every subsequent argument, from procedure to remedy, flows from that first determination.

Many classification disputes never reach a courtroom because they are resolved inside the system the dispute concerns. Contractors sometimes reverse denials at redetermination or reconsideration when the appellant demonstrates that the manual section applied has no regulatory anchor, preferring to pay the claim rather than defend the document. The agency sometimes revises the manual section itself, either narrowing it to fit the regulation or, where the policy is genuinely needed, proposing it as a regulation through notice and comment. These quiet resolutions leave no published opinion, which means the visible case law understates the classification question’s practical importance: for every document a court holds substantive, several more were softened or withdrawn before a judge ever saw them.

Certain kinds of provisions drew challenges more often than others, and the pattern confirms the test. Documentation requirements that conditioned payment on records the regulation never mentioned, medical review criteria that narrowed coverage beyond the regulatory standard, and enrollment or eligibility conditions that added qualifications the statute did not state were the recurring flashpoints. In each pattern, the challengers’ argument had the same shape: the regulation set a standard, the manual added a requirement, and claims were denied for failing the addition. Where the agency could show that the manual merely specified how to demonstrate compliance with a standard the regulation already imposed, it generally prevailed. Where it could not, the document looked substantive, and the agency’s characterization could not save it.

The retroactivity bar in subsection (e) generated its own line of disputes, concentrated where the Program Integrity Manual meets overpayment recovery. Auditors measuring years of past claims against newly revised manual sections created the precise scenario Congress had in mind: providers judged under standards announced after the services were furnished. The agency’s response, at its strongest, distinguished clarification from change, arguing that a transmittal restating what the regulation always required could be applied to earlier periods without offending the bar. Challengers answered that the distinction was manipulable, and that any manual revision that altered audit outcomes was substantive in the only sense that mattered to the provider writing the refund check. Here again the classification question controlled: clarification looks backward freely, while substantive change must face forward.

For the practitioner, the dispute has a practical shape that repeats across cases. The classification argument is raised within the appeals ladder itself, typically at reconsideration or before the administrative law judge, where the appellant contends that the denial rests on a document that should have gone through rulemaking. Building that argument requires the record work described earlier: the manual section, the transmittal that introduced it, the regulation it purports to interpret, and evidence of how contractors apply it. The argument’s strength depends entirely on the gap between the document and its regulatory anchor, which is why the most effective challenges attach the two texts side by side and let the distance speak. Counsel who master this comparison win more manual disputes than counsel who master the case law, because the documents themselves are the evidence.

The dispute recurs because its structural causes are permanent. The agency will always need to issue guidance faster than rulemaking allows, regulated parties will always experience guidance as binding, and the 1986 compromise will always invite argument about where interpretation ends and legislation begins. Each generation of manuals revives the question in new factual clothing: new benefits, new payment methodologies, new program integrity initiatives, each requiring instructions that someone will call substantive. The practitioner should therefore treat the interpretive-versus-substantive line not as a controversy awaiting resolution but as a permanent feature of the terrain, like a river that must be forded on every journey through the program’s law. General administrative law supplies the background vocabulary of interpretive and legislative rules, but 1395hh supplies the sharper, program-specific edge: in the three fenced domains, the question is not only whether the document interprets, but whether Congress permitted it to do anything more.

The fourth machine: the appeals ladder

Section 1869 of the Social Security Act, codified at 42 U.S.C. 1395ff and added by the Medicare, Medicaid, and SCHIP Benefits Improvement and Protection Act of 2000, built the modern appeals structure. Before that statute, the program’s appeals were fragmented: different tracks, different rules, and different decision-makers depending on whether the claim arose under Part A or Part B, a division that confused beneficiaries and providers alike. BIPA 2000 unified the process into a single five-stage ladder with defined filing deadlines and decision timeframes at every rung, and it paired the appeals reform with contractor reform, recognizing that the entities deciding claims and the process for challenging those decisions had to be modernized together. Readers who encounter claims under the private plan options created by the 2003 legislation should note that those plans operate under separate appeal rules, described in the series account of the 2003 act and its drug benefit, the series account of the 2003 act and its drug benefit; the ladder described here governs original fee-for-service claims.

The pre-2000 fragmentation that BIPA 2000 replaced had grown up alongside the program’s two-part structure. Part A claims moved through intermediary reconsiderations under one set of procedures, while Part B claims moved through carrier hearings and administrative law judge review under another, and beneficiaries whose care spanned both parts could face parallel proceedings under different rules. Providers complained of inconsistency, beneficiaries complained of confusion, and oversight reviews documented a system whose complexity itself denied meaningful review to many appellants. The 2000 statute’s answer was structural rather than cosmetic: a single ladder, identical for both parts, with the same deadlines and the same decision-makers regardless of where the claim originated.

The contractor reform culminated in section 911 of the Medicare Prescription Drug, Improvement, and Modernization Act of 2003, which replaced the old fiscal intermediaries and carriers with Medicare administrative contractors. Under the prior structure, fiscal intermediaries processed Part A claims and carriers processed Part B claims, often applying different local policies to the same service. The 2003 statute directed the agency to award consolidated contracts covering both parts within defined jurisdictions, so that a single contractor would administer the full range of claims in its territory. For appeals, the change mattered structurally: redetermination would be performed by the contractor that made the initial determination, a unified entity accountable for the whole claim rather than a fragment of it, and appellants would face one contractor instead of two.

The ladder serves two populations whose interests sometimes align and sometimes diverge. Beneficiaries appeal when the program denies coverage for care they received or need, and the dispute is about access to the benefit itself. Providers and suppliers appeal when the program denies payment for care they furnished, and the dispute is about compensation for work performed. Both travel the same five stages under section 1869, though the economics differ sharply: a beneficiary’s appeal may concern a single course of treatment, while a provider’s appeal may concern thousands of claims denied under the same edit. The design accommodates both, but the volume that strains the system comes overwhelmingly from the provider side, where automated denials generate appeals by the batch.

What are the five stages of a Medicare claim appeal?

Redetermination by the contractor, reconsideration by an independent contractor, an ALJ hearing, review by the Medicare Appeals Council, and judicial review in district court form the five stages. Filing deadlines run 120, 180, 60, 60, and 60 days; decisions at the first four stages are due within 60, 60, 90, and 90 days; amount thresholds restrict the last two.

The first stage, redetermination, is a paper review performed by the same contractor that made the initial determination. The appellant must file within 120 days of receiving the initial determination, and the contractor must decide within 60 days. No hearing occurs; a reviewer uninvolved in the original decision examines the claim, the submitted documentation, and any additional evidence the appellant supplies. This stage corrects the correctable: clerical mistakes, coding errors, missing pages in the record, and misapplications of the contractor’s own edits. Because it is fast, inexpensive, and decided by the entity with the claim file already in hand, redetermination resolves an enormous share of disputes, and experienced appellants treat it as the stage where careful documentation does the most work.

The second stage, reconsideration, moves the file to a qualified independent contractor, an entity with no role in the initial determination or the redetermination. The filing deadline is 180 days from the redetermination notice, and the decision is due within 60 days. Independence is the point: the QIC reviews the record fresh, and where medical necessity is at issue, the review is conducted by clinicians with relevant expertise. Appellants may submit additional evidence, and the reconsideration decision must explain its reasoning in greater detail than the terse notices of earlier stages. The QIC stage is the last purely paper review and the last decision made within the contractor system; what survives it proceeds to adjudicators outside that system entirely.

The qualified independent contractor model was itself an innovation worth understanding. Rather than assigning reconsiderations to another unit of the same contractor, the statute required the agency to engage separate entities whose sole function is second-level review, organized in jurisdictions distinct from the claims-processing contractors. The design serves two purposes at once: it removes the appearance, and the reality, of an institution reviewing its own work, and it concentrates reconsideration expertise in organizations that do nothing else. The clinicians these contractors employ for medical necessity reviews are the first medically trained reviewers many appellants encounter, which is why reconsideration sometimes reverses denials that turned on clinical judgment rather than on coding.

The third stage is the hearing before an administrative law judge in the Office of Medicare Hearings and Appeals, and it changes the character of the proceeding entirely. The request must be filed within 60 days of the reconsideration notice, the decision is generally due within 90 days, and the stage is available only when the claim meets a minimum amount in controversy set by regulation. For the first time, the appellant may testify, present witnesses, and be represented by counsel in a live proceeding, with the judge developing the record through questioning. The ALJ decides independently of the contractors, applying the statute, the regulations, and any guidance that qualifies as authoritative. Because credibility, clinical judgment, and expert testimony can enter the record, cases that turned on paper at the contractor stages can turn on testimony here, and the outcomes sometimes diverge sharply from what the paper record predicted.

Representation rights expand as the ladder ascends. At every stage, appellants may appoint a representative, whether an attorney, a billing specialist, a family member, or an advocacy organization, through a formal appointment that the adjudicator must honor. Beneficiaries navigating the system without counsel rely heavily on these representatives, and on the written notices the statute requires at each stage explaining what was decided and what comes next. Providers, by contrast, typically arrive with professional appeals staff or outside counsel from the first filing, and the quality gap between represented and unrepresented appellants is one of the system’s persistent, if rarely discussed, features. The ladder is formally neutral between them; in practice, expertise compounds at every rung.

The fourth stage is review by the Medicare Appeals Council, a component of the Departmental Appeals Board. The request must be filed within 60 days of the ALJ decision, and the Council’s decision is generally due within 90 days. Review is on the record created below; there is no new hearing, though the Council may consider additional evidence in limited circumstances. The Council may affirm, reverse, or modify the ALJ’s decision, or remand the case for further proceedings, and its decision becomes the final administrative determination of the Secretary. The Council may also take review of ALJ decisions on its own motion, a reminder that the stage serves the agency’s interest in uniformity as well as the appellant’s interest in correction. This is the last forum in which the dispute remains inside the executive branch.

The fifth and final stage is judicial review in federal district court. The action must be filed within 60 days of the Council’s decision, it requires exhaustion of the four administrative stages, and it is available only when the claim satisfies a higher minimum amount in controversy than the ALJ stage demands. The court decides questions of law without deference to the tribunals below, while reviewing their factual findings with the restraint courts traditionally afford administrative records. Few claims travel this far: the amount thresholds, the cost of litigation, and the attrition of the earlier stages winnow the docket to disputes where the dollars or the legal principle justify a federal case.

The amount-in-controversy thresholds contain a practical accommodation for the economics of small claims: appellants may aggregate multiple claims to reach the required amount, provided the claims present common issues. A provider with hundreds of identical denials under a single edit need not find one claim large enough to qualify; the batch itself, sharing a common question of law or fact, can cross the threshold together. Aggregation is what makes the ALJ stage accessible for systemic disputes, and it explains why some of the most consequential hearings concern legal principles worth far more in the aggregate than any single claim in the batch.

The word “generally” in the decision targets carries a procedural safety valve worth understanding. When an adjudicator at any stage fails to decide within its timeframe, the appellant need not wait indefinitely; the structure permits escalation to the next stage, preserving forward motion when the clock breaks down. Escalation is the statute’s answer to the timeliness problem: instead of leaving appellants stranded behind a missed deadline, it lets them move up the ladder and keep the dispute alive. The mechanism is used more often in periods of heavy backlog, and its existence shapes strategy, because an appellant weighing escalation must decide whether the existing record is strong enough to carry upward without the decision that never came.

The notices issued at each stage do more than communicate outcomes; they structure the appeal itself. An initial determination must state the reason for denial in terms the appellant can understand and must explain the right to redetermination, including the 120-day deadline. Each subsequent decision must similarly explain the next available stage and its filing window. These notice requirements are the ladder’s internal signage, and their adequacy is itself appealable: a notice that fails to explain the basis for denial deprives the appellant of the ability to respond, which is why careful practitioners attack defective notices as well as defective decisions.

Most disputes end at the contractor stages for reasons that are economic rather than legal. The great majority of denied claims involve modest dollar amounts for which the cost of pursuing an ALJ hearing would exceed any recovery, and many denials reflect fixable defects: a missing modifier, an unsigned order, a diagnosis code that failed an edit. Redetermination exists precisely to catch these, and the 120-day filing window with a 60-day decision clock makes it the fastest correction mechanism in the system. Providers with high claim volumes build their appeals operations around these early stages, appealing systematically at redetermination and selectively at reconsideration, reserving the ALJ stage for claims where the amount justifies the investment or where the legal issue recurs across many claims and a favorable decision would carry weight across their operations.

The ladder also carries a second kind of dispute that behaves differently from ordinary claim denials: challenges to overpayment determinations. When program integrity audits extrapolate a sample of reviewed claims into a demand covering years of payments, the provider appeals the overpayment determination through the same five stages, but the economics invert. The amounts are large enough to justify climbing the full ladder, the legal issues often concern the sampling methodology or the manual standard applied rather than any single claim’s facts, and the provider’s survival may depend on the outcome. These are the appeals that most often reach the ALJ and the Council, and they are where the classification arguments about manuals do their heaviest work, because the overpayment usually rests on a manual provision the provider contends is substantive.

When does a denied claim reach federal court?

A denied claim reaches federal district court only after exhausting all four administrative stages, filing within 60 days of the Medicare Appeals Council decision, and meeting a higher amount-in-controversy threshold. Exhaustion gives the agency the first chance to correct errors and build a complete record, and courts then decide questions of law fresh while reviewing factual findings with restraint.

Exhaustion is the doctrine that holds the ladder together, and its justifications are institutional. The agency that administers the payment system should have the first opportunity to correct its own errors, develop a complete factual record, and apply its expertise to the clinical and coding questions that courts are poorly equipped to resolve in the first instance. Exhaustion also protects the courts: by the time a claim arrives at the district court, four prior decision-makers have narrowed the issues, assembled the evidence, and explained their reasoning, so the judge confronts a defined legal question rather than a raw billing dispute. And it protects the trust funds, because a system that routed every denied claim directly to federal court without administrative filtering would collapse under its own weight.

The standard of review reflects the division of labor the ladder creates. Questions of law, including the interpretation of the statute and the validity of the regulations and guidance applied below, are decided by the court on their merits, without deference to the ALJ or the Council. Factual findings, by contrast, are reviewed with restraint, sustained where the record provides adequate support. The amount-in-controversy thresholds perform a parallel gatekeeping function: the lower threshold reserves the ALJ hearing, with its live testimony and attendant expense, for disputes of sufficient size, while the higher threshold reserves the federal courts for the disputes that matter most. Together, exhaustion, the thresholds, and the standard of review express a single judgment about institutional competence: contractors and clinicians establish the facts, the agency’s adjudicators develop the record, and courts declare the law.

The ladder’s deadlines created a persistent structural tension between due process and volume. Each stage carries a decision clock, 60 days at the contractor stages and generally 90 days at the ALJ and Council stages, but the number of appeals grew as program spending grew and as contractors intensified medical review. Adjudicators faced dockets the clocks were not designed to accommodate, and federal oversight offices documented the resulting strain: decisions issued after their target timeframes, backlogs accumulating at the hearing stage, and appellants waiting well beyond the periods the statute contemplated. The reports described an institutional mismatch rather than individual failure, a ladder engineered for a smaller caseload carrying a larger one. For appellants, the practical consequence was delay layered atop denial, and for the system, a standing question about whether the promise of timely review could be kept at scale.

Step back and the ladder reveals its logic in what changes at each rung. Redetermination offers a second look by the same institution, cheap and fast, catching error. Reconsideration offers a first look by an independent institution, adding clinical expertise the contractor may lack. The ALJ hearing offers the first live proceeding, adding testimony and credibility to a record that was paper until then. The Council offers the first review for legal uniformity across the system, adding the agency’s last word. Judicial review offers the first determination by an institution outside the executive branch, adding independent judgment on the law. Each rung adds something the previous ones could not supply, and each rung’s deadline and threshold reflects a judgment about when that addition is worth its cost. One further link between the machines deserves notice: the ALJ decides according to statute, regulation, and agency rulings, giving deference to manual guidance without being bound by it the way contractors are, which makes the hearing stage the first forum where a classification challenge to a manual can receive a genuinely independent assessment.

The complication and the verdict

The complication is a belief that survives every explanation of the system: that Congress sets what the program pays. The belief is understandable. Congress created the program, Congress writes the statutes that govern it, and Congress occasionally intervenes to set a specific rate or mandate a particular update. But the belief mistakes the frame for the picture. Congress sets frameworks and occasionally fixes numbers; the operative payment numbers, the ones that determine what a hospital receives for a discharge or a physician receives for a procedure, emerge from the rulemaking process and the sub-regulatory layer that implements it. To follow the money is to follow the agency, not the statute books.

Follow a single payment from authorizing statute to deposited funds and the four machines reveal themselves in sequence. Congress enacts the framework: the benefit categories, the payment methodologies, the appeals rights in section 1869, the rulemaking duties in section 1395hh. The agency translates the framework into payment rules through notice and comment, setting the rates and methodologies that contractors will apply. Contractors implement those rules through manuals, transmittals, and change requests, converting regulatory language into claims edits and payment systems. The initial determination applies all of it to a single claim. And when the claim is denied, the appeals ladder tests whether each machine did its work lawfully: whether the rule was properly promulgated, whether the manual faithfully interpreted it, and whether the facts fit the standard.

The chain is visible from the beneficiary’s side as well, though it looks different. The beneficiary never sees the rulemaking or the change request; she sees the denial notice, which by law must explain the reason for the determination and the appeal rights that accompany it. Those rights are the fourth machine made personal: 120 days to seek redetermination, then the climb through reconsideration, hearing, Council, and court. The notice is the hinge between the machines, the document that converts an administrative determination into an appealable event. A practitioner advising beneficiaries learns to read these notices the way providers read remittance advices, because the stated reason for denial dictates which machine’s work is being challenged and which rung of the ladder must be climbed first.

The chain also dictates a reading order for anyone confronting a denial. Start with the notice, which identifies the reason and therefore the document under challenge. Move to the manual section or edit cited, and ask whether it faithfully implements its regulatory source. Move to the regulation, and ask whether it faithfully implements the statute. At each step upstream, the question repeats: did this document have the authority to do what it did. Practitioners who read downstream, from statute to denial, understand the system’s design; practitioners who read upstream, from denial to statute, win the individual case. The discipline is to do both, and the ladder rewards those who do.

None of this denies Congress its moments of direct rate-setting. When Congress freezes an update, mandates a specific percentage, or writes a payment amount into statute, the agency’s discretion narrows to implementation and the chain shortens. But these interventions are the exception that confirms the pattern: they make news precisely because they depart from the ordinary course, in which Congress legislates the methodology and the agency produces the number. The practitioner who waits for Congress to set payment waits for the exception; the practitioner who learns the agency’s machines is ready for the rule.

That leaves the question of where a practitioner should begin. The deciding factor is procedural primacy: the machine through which every dispute must pass, and whose deadlines extinguish rights most unforgivingly, deserves the first hours of study. By that measure, the answer is the fourth machine, the appeals ladder. Every dispute generated by the other three machines must travel the ladder to be resolved: a challenge to a payment rule, a fight over a manual’s classification, and a denial under a contractor’s edit all enter through redetermination and climb or die by the same deadlines. Those deadlines are jurisdictional in effect, and a brilliant argument that a manual section is substantive under 1395hh(a)(2) is worthless if the 120-day redetermination window has closed.

Moreover, the classification fight that defines the third machine is waged on the ladder’s rungs. It is at reconsideration, before the ALJ, and before the Council that the interpretive-versus-substantive argument is raised, briefed, and preserved for judicial review. The ladder is thus not merely a procedure for resolving disputes but the platform on which every other machine’s disputes become litigable. Master the ladder and the arguments generated by the other machines have somewhere to go; neglect it and even the strongest substantive position never reaches a decision-maker. Procedural primacy does not mean procedure matters more than substance. It means substance without procedure is inaudible.

The sub-regulatory layer is the close second, and in daily practice the two are inseparable. The ladder supplies the procedure; the classification question supplies the substance of most payment disputes. A practitioner who knows the deadlines but cannot tell an interpretive instruction from a substantive rule will preserve claims only to lose them on the merits, just as one who knows the classification test but misses a filing date will lose claims without reaching the merits. Learn the ladder first, the manuals second, and then read the payment rules with the question that governs everything: which document actually decided this claim, and did it have the authority to do so.

Learning the ladder first has an operational meaning that goes beyond memorizing deadlines. It means calendaring every denial notice on receipt, because the 120-day redetermination clock runs from receipt and excuses are narrow. It means reading each notice for the stated reason, because the reason identifies the manual section, the edit, or the rule under challenge. It means building the evidentiary record early, because the contractor stages are the cheapest place to win and the record built there constrains everything above. These habits are unglamorous and decisive, which is why they come first: the practitioner who internalizes them rarely loses on procedure, and the practitioner who does not will eventually lose on procedure regardless of substantive merit.

The stakes of these machines are not abstract. For beneficiaries, the appeals ladder is the difference between a denial that stands and care that is covered; for providers, the sub-regulatory layer is the difference between billing operations that survive audits and overpayment demands that threaten viability. The four machines together determine how the program’s promises translate into paid claims, and the practitioner who masters them holds the map of that translation. Procedural primacy says to learn the ladder first. Everything else follows from knowing where every dispute must go.

The rule-to-denial pipeline

One pipeline connects the statute on the shelf to the denial in the mailbox, and every stage has a decisionmaker, a governing standard, and a deadline. Read the table across a row before you read down a column, because the pipeline is a sequence: each stage applies what the stage above produced, and the deadline in one cell is the clock that starts the next.

Stage	Decisionmaker	Governing standard	Deadline
Statute	Congress	Authorizes benefit categories, payment methodologies, coverage exclusions, and appeal rights in Title XVIII	Public law; amendment requires legislation
Regulation	Secretary of HHS, acting through CMS	Federal Register notice and at least 60 days of public comment under 42 U.S.C. 1395hh(b); substantive standards in benefits, payment, and eligibility by regulation under 42 U.S.C. 1395hh(a)(2)	Proposed rule, then final rule; payment rules finalize on a fixed annual calendar
Sub-regulatory guidance	CMS headquarters and components	Interprets existing statute and regulation in manuals, transmittals, change requests, and memoranda; may not create new substantive legal standards	Implementation dates set by transmittal; substantive changes apply prospectively under 42 U.S.C. 1395hh(e)
Initial determination	Medicare administrative contractor	Applies statute, regulation, national and local coverage determinations, and manual guidance to the submitted claim	Contractor processes claim; appealable within 120 days of receipt
Redetermination	Same contractor, new reviewer	Paper review of the claim file and any additional documentation; terse notice requirement	File within 120 days; decide within 60 days
Reconsideration	Qualified independent contractor	Fresh review by an entity separate from the contractor, with clinical expertise where medical necessity is at issue; reasoned decision	File within 180 days of the redetermination notice; decide within 60 days
ALJ hearing	Administrative law judge, Office of Medicare Hearings and Appeals	Live proceeding with testimony under oath; applies statute, regulation, and authoritative guidance; amount in controversy threshold	Request within 60 days of the reconsideration notice; decide generally within 90 days
Medicare Appeals Council	Medicare Appeals Council, Departmental Appeals Board	Record review of the ALJ decision; may affirm, reverse, modify, or remand; decision becomes the final determination of the Secretary	Request within 60 days of the ALJ decision; decide generally within 90 days
Federal court	United States district court	De novo review of legal questions; restraint on facts supported by substantial evidence; the Supreme Court’s 1975 and 2000 channeling decisions require completion of the administrative ladder, subject to narrow exceptions	File within 60 days of the Council decision

For readers keeping study notes across the series, the VaultBook study notebook offers a place to record how each payment rule builds on its statute, and the ReportMedic civics study guide provides the broader government context for the agency and its role.

Frequently Asked Questions

Q: What does CMS do under the Medicare statute?

The Centers for Medicare and Medicaid Services administers Title XVIII of the Social Security Act under authority delegated by the Secretary of Health and Human Services. In practice that means the agency writes the regulations that implement the statute, publishes the annual payment rules that set hospital, physician, and post-acute rates, issues national coverage determinations on whether particular items and services meet the reasonable and necessary standard, and oversees the Medicare administrative contractors that process claims and issue local coverage determinations. It also maintains the manuals and program guidance that instruct contractors on daily operations, and it defends the resulting payment and coverage decisions through the five-stage administrative appeals ladder. The statute sets the framework and the boundaries, including the section 1801 bar on federal control of medical practice; CMS fills in the working detail that determines what providers are paid and which services are covered.

Q: How does CMS write Medicare rules?

CMS writes Medicare rules through notice and comment rulemaking published in the Federal Register. The agency publishes a proposed rule that states the policy, discloses the data and methodology behind it, and invites public comment, then publishes a final rule that responds to the comments and amends the Code of Federal Regulations at 42 C.F.R. For Medicare, section 1871(b) of the Social Security Act, 42 U.S.C. 1395hh(b), adds a program-specific requirement: before issuing a final regulation, the Secretary must provide Federal Register notice and a comment period of not less than 60 days. The annual payment rules follow this sequence every year, with proposed rules in the spring or summer and final rules in time for the October 1 fiscal year or January 1 calendar year payment periods. Manuals and guidance, by contrast, issue without notice and comment and may only interpret existing law.

Q: What is the annual Medicare payment rule cycle?

It is the yearly sequence in which CMS proposes and finalizes the rates and formulas that determine what providers are paid. The inpatient prospective payment rule is proposed in the spring and finalized by August 1 for the federal fiscal year beginning October 1. The outpatient prospective payment rule follows a calendar year cycle, proposed in the summer and finalized by November 1 for the payment year beginning January 1. The physician fee schedule is proposed in the summer and finalized by November 1 for the calendar year beginning January 1. Post-acute rules for skilled nursing facilities, home health agencies, inpatient rehabilitation facilities, and long term care hospitals follow the same proposed-then-final pattern. Each rule updates base rates, classification weights such as diagnosis-related group weights and relative value units, geographic adjustments, and policy details like the wage index and outlier thresholds, with distributional tables showing the estimated effects on provider categories and a comment-and-response preamble recording the agency’s reasoning.

Q: What is a Medicare national coverage determination?

A national coverage determination is the Secretary’s decision whether a particular item or service is covered nationally under Title XVIII, made under section 1862(l) of the Social Security Act, 42 U.S.C. 1395y(l). The process begins with a formal request, which anyone may submit, and the agency must publish a proposed decision within six months, or nine months when it commissions an external technology assessment or consults its advisory committee. Thirty days of public comment follow, then a final determination that may grant coverage, deny it, or grant it conditionally, for example for specified diagnoses, settings, or providers, or only when furnished in an approved clinical study under coverage with evidence development. The determination rests on published clinical evidence and technology assessments, and the agency’s coverage memorandum explains which studies it found persuasive and why, making the memorandum a durable piece of the program’s reasoning.

Q: Does CMS have to use notice and comment for Medicare rules?

Yes, and the Medicare statute imposes a stricter, more explicit requirement than the general default. Under 42 U.S.C. 1395hh(a)(2), no rule, requirement, or other statement of policy, other than a national coverage determination, that establishes or changes a substantive legal standard governing the scope of benefits, payment for services, or eligibility to furnish or receive services may take effect unless promulgated by the Secretary by regulation. Under subsection (b), a final regulation requires Federal Register notice and a public comment period of not less than 60 days, with narrow exceptions where a statute permits interim final form or sets a short deadline. Interpretive rules and guidance that merely explain existing law may issue without comment, but the line between interpretation and a substantive standard is the program’s most litigated question, and documents that cross it without rulemaking cannot lawfully take effect in the three fenced domains.

Q: How do you appeal a Medicare coverage denial?

Start with the denial notice, which by law must state the reason for the denial and explain the appeal rights. The reason identifies which document is being challenged, whether a manual section, a local coverage determination, a coding edit, or a payment rule, and that identification shapes everything above. File for redetermination with the Medicare administrative contractor within 120 days of receiving the initial determination, submitting the missing documentation or the argument that the standard was misapplied, since this stage corrects clerical and coding errors fastest. If denied, request reconsideration by a qualified independent contractor within 180 days, adding clinical evidence where medical necessity is at issue. Beyond that lie the administrative law judge hearing, the Medicare Appeals Council, and federal court, each with 60-day filing windows. Build the record early, because the evidence assembled at the contractor stages constrains every rung above.

Q: What was HCFA before it became the Medicare agency CMS?

The Health Care Financing Administration was the federal agency that administered Medicare and Medicaid from its creation in March 1977 until its renaming in 2001. Established under the Department of Health, Education, and Welfare, it consolidated federal health financing responsibilities that had previously been divided between the Social Security Administration, which handled Medicare enrollment and premium processing, and the Social and Rehabilitation Service, which handled Medicaid. HCFA oversaw the 1983 creation of hospital prospective payment, the 1989 physician fee schedule, and the contractor system of fiscal intermediaries and carriers. In 2001 the agency was renamed the Centers for Medicare and Medicaid Services, a change announced that June and later codified in the 2003 Medicare legislation, to reflect that its mission extended beyond financing to the quality and administration of the programs. The regulations it issued as HCFA remain in force under the new name.

Q: What did Azar v. Allina decide about Medicare rulemaking?

This article does not describe that decision, because it falls outside the article’s scope. Azar v. Allina Health Services, 587 U.S., was decided in 2019, eight years after this article’s 2011 date wall, and this series does not treat later developments as part of the law it explains. The underlying question the case concerned, however, can be answered from the statute as it stood at the wall. Under 42 U.S.C. 1395hh(a)(2), added in the 1986 budget reconciliation act, no rule, requirement, or other statement of policy, other than a national coverage determination, that establishes or changes a substantive legal standard governing the scope of benefits, payment for services, or eligibility to furnish or receive services may take effect unless promulgated by the Secretary by regulation. Subsection (b) requires Federal Register notice and at least 60 days of public comment before the final regulation. Whether a given manual or guidance document crosses that line was already the program’s central rulemaking dispute by 2011.

Q: How do you find a current Medicare payment rule in the Federal Register?

Use the two publications together, because they answer different questions. The Federal Register carries every proposed and final rule with its preamble, organized by date of issuance, and it is where you find what the agency did in a particular rulemaking and why: the proposed rates, the distributional tables, the comment-and-response discussion, and the effective dates. The Code of Federal Regulations at 42 C.F.R. carries the codified regulatory text organized by subject, showing what the regulation says as amended to date. To trace a payment rate, find the final rule for the relevant payment year in the Federal Register, read the preamble for the methodology and the agency’s responses to comments, then check the current 42 C.F.R. text for the operative language. The preamble is not regulation in the strict sense, but contractors and auditors apply the rule as the preamble describes it, so professionals read both.

Q: What is a Medicare local coverage determination and who issues it?

A local coverage determination is a coverage policy issued by a Medicare administrative contractor for its own jurisdiction, applying the reasonable and necessary standard of section 1862(a)(1)(A) where no national coverage determination exists. The contractor’s medical directors develop the policy, typically publishing a draft, holding a comment period, and consulting a contractor advisory committee of physicians from the jurisdiction before issuing the final determination with its covered indications, documentation requirements, coding guidance, and limitations. Because contractors weigh evidence independently, coverage for the same service can differ by region. A local determination cannot contradict a national one; it operates only in the space the national determinations leave open. Interested parties may ask the contractor to reconsider a local determination, and the agency reviews new and revised determinations for consistency with national policy without fully standardizing their substance.

Q: What is the reasonable and necessary standard for Medicare coverage?

It is the statutory gate for coverage, stated in section 1862(a)(1)(A) of the Social Security Act, 42 U.S.C. 1395y(a)(1)(A): no payment may be made for items or services that are not reasonable and necessary for the diagnosis or treatment of illness or injury or to improve the functioning of a malformed body member. Congress left the phrase undefined, and the agency has applied it primarily through evidence of clinical effectiveness, asking whether the item or service improves health outcomes for beneficiaries rather than applying a formal cost effectiveness threshold. National coverage determinations apply the standard on a public record with published memoranda reviewing the evidence; local coverage determinations apply it through contractor medical directors; and individual claims reviewers apply it case by case. The standard’s generality is its strength and its controversy, since it lets the agency evaluate technologies that did not exist when the statute was written while leaving wide discretion in how much evidence is enough.

Q: What was the sustainable growth rate system for physician payment?

The sustainable growth rate was the statutory formula, created by the Balanced Budget Act of 1997, that set the annual update to the physician fee schedule conversion factor. It compared actual cumulative physician spending against a target derived from economic growth and other factors: when spending exceeded the target, the formula called for update reductions, and when spending fell short, it called for increases. By 2010 the formula was generating calls for steep payment cuts that Congress found unacceptable, and Congress intervened year after year with short-term legislation overriding the scheduled reductions. The episode showed the limits of formula-driven payment policy, because a statutory formula can produce results the political system will not sustain. Each annual physician fee schedule rule recorded the gap, publishing the update the formula would produce alongside the patched number Congress permitted, and the repeated patches became a yearly feature of the rulemaking calendar.

Q: What is a Medicare administrative contractor and what does it do?

A Medicare administrative contractor is the private entity that administers original Medicare claims within a defined geographic jurisdiction. Section 911 of the Medicare Prescription Drug, Improvement, and Modernization Act of 2003 replaced the old fiscal intermediaries, which processed Part A claims, and carriers, which processed Part B claims, with consolidated contractors handling both parts in their territories. The contractor processes claims and makes initial determinations, issues local coverage determinations where no national policy exists, conducts medical review and audits, performs redeterminations as the first stage of appeal, and implements agency policy through claims processing systems built to the agency’s change requests. Because the contractor is the face of the program for providers and beneficiaries, most coverage and payment disputes begin and end with its decisions, and its local policies are the working law for the services no national determination addresses.

Q: What happens at the redetermination stage of a Medicare appeal?

Redetermination is the first appeal stage: a paper review by the Medicare administrative contractor that made the initial determination, conducted by a reviewer uninvolved in the original decision. The appellant must file within 120 days of receiving the initial determination, and the contractor must decide within 60 days. No hearing occurs. The reviewer examines the claim, the submitted documentation, and any additional evidence the appellant supplies, and issues a Medicare redetermination notice explaining the result and the next appeal right. This stage corrects the correctable: clerical mistakes, coding errors, missing pages, and misapplications of the contractor’s own edits. Because it is fast and inexpensive and the claim file is already in hand, redetermination resolves an enormous share of disputes, and experienced appellants treat careful documentation at this stage as the highest-leverage work in the entire ladder.

Q: What is a qualified independent contractor in Medicare appeals?

A qualified independent contractor is the separate entity that decides the second stage of a Medicare appeal, reconsideration. Created under the appeals reform of the 2000 benefits improvement act, the QIC has no role in the initial determination or the redetermination, and its jurisdictions are organized separately from the claims-processing contractors. The appellant files within 180 days of the redetermination notice, and the QIC must decide within 60 days. Independence is the design’s point: the file gets a fresh review, and where medical necessity is at issue, clinicians with relevant expertise conduct the review. Appellants may submit additional evidence, and the reconsideration decision must explain its reasoning more fully than the terse notices of earlier stages. Reconsideration is the last purely paper review and the last decision made inside the contractor system; what survives it proceeds to adjudicators outside that system entirely.

Q: How does an administrative law judge hearing work in a Medicare appeal?

The hearing is the third appeal stage, conducted by an administrative law judge in the Office of Medicare Hearings and Appeals, and it transforms the proceeding from paper review into a live adjudication. The appellant requests the hearing within 60 days of the reconsideration notice, and the decision is generally due within 90 days. For the first time the appellant may testify, present witnesses, and be represented by counsel while the judge develops the record through questioning, with testimony taken under oath and the proceeding recorded. The stage is available only when the claim meets a minimum amount in controversy set by regulation, and multiple claims with common issues may be aggregated to reach it. The judge decides independently of the contractors, applying the statute, the regulations, and authoritative guidance, and gives manual guidance deference without being bound by it the way contractors are.

Q: What does the Medicare Appeals Council review?

The Medicare Appeals Council, a component of the Departmental Appeals Board, reviews administrative law judge decisions on the record created below. The appellant requests review within 60 days of the ALJ decision, and the Council’s decision is generally due within 90 days. There is no new hearing, though the Council may consider additional evidence in limited circumstances. It may affirm, reverse, or modify the judge’s decision or remand the case for further proceedings, and its decision becomes the final administrative determination of the Secretary, the last word of the executive branch before judicial review. The Council may also take review of ALJ decisions on its own motion, which serves the agency’s interest in legal uniformity across adjudicators as well as the appellant’s interest in correction. Like the judge below, the Council may not disregard a national coverage determination; it may decide only whether the determination applies to the claim and was applied correctly.

Q: What is the difference between a Medicare manual and a Medicare regulation?

A regulation is made through notice and comment rulemaking under 42 U.S.C. 1395hh, published in the Federal Register, and codified at 42 C.F.R.; it carries the force of law and binds the agency, contractors, adjudicators, and the public. A manual is sub-regulatory guidance issued through transmittals without public comment; it interprets existing law for the contractors the agency contracts with, and contractors must follow it under their contracts. The hierarchy is strict: a manual cannot contradict its regulation, and where the two conflict the regulation controls. The recurring dispute is classification. When a manual section merely explains a regulatory standard, it is interpretive and lawful. When it creates a duty, condition, or eligibility test the regulation does not contain, challengers argue it is a substantive legal standard that should have gone through rulemaking. Practitioners test every manual section against its regulatory anchor before relying on it.

Q: What is the Office of Medicare Hearings and Appeals?

The Office of Medicare Hearings and Appeals is the HHS component that employs the administrative law judges who decide the third stage of Medicare claim appeals. Created to separate the hearing function from the claims-processing contractors, it provides the first forum in the ladder where the appellant may testify, present witnesses, and be represented by counsel in a live proceeding. Its judges decide independently of the contractors and of CMS policy offices, applying the statute, the regulations, and authoritative guidance, and they are the first adjudicators who can give a classification challenge to a manual a genuinely independent assessment, since contractors must follow manuals under their contracts while judges only owe them deference. The office’s decisions are subject to review by the Medicare Appeals Council, and its hearing stage is the gateway to the upper rungs of the ladder for any dispute large enough to meet the amount-in-controversy threshold.

Q: Can CMS apply a new Medicare manual provision retroactively?

Not when the provision is a substantive change. Section 1395hh(e) bars retroactive application of a substantive change in regulations, manual instructions, interpretative rules, statements of policy, or guidelines of general applicability. The provision names the sub-regulatory layer explicitly: manual instructions appear alongside regulations as documents whose substantive changes must apply prospectively. A provider that furnished services in reliance on the manual language then in effect cannot be judged under a rewritten manual section for those past services, at least where the change is substantive. The agency’s defense, at its strongest, is that a transmittal merely clarified what the regulation always required, and clarification looks backward freely. Challengers answer that any revision that changes audit outcomes is substantive in the only sense that matters. The dispute concentrates where the Program Integrity Manual meets overpayment recovery, with auditors measuring years of past claims against newly revised standards.