A Brief History of the Denial

A denied medical claim looks like a modern artifact — a code, a remittance line, an automated edit — but the machinery behind it is old, and it was assembled deliberately. Nobody set out to build a system in which a correctly delivered service is refused payment; the pieces were added one at a time, each to solve a problem the previous piece had left open. The through-line is a single idea: that a third party, neither the patient nor the treating physician, should decide whether a given service was necessary before the money changes hands. That idea has a birthday, a set of founding statutes, and a documented path from committees of doctors reading charts to software that adjudicates millions of claims a day.

This is an account of that path, reconstructed from the statutes, federal reports, and surveys that record it. It is not an argument that review is wrong or right — utilization review was created in part to check genuine overuse, and it still catches real errors. It is a description of how the denial became an institution.

When President Lyndon Johnson signed the Social Security Amendments of 1965 on July 30 of that year — creating Medicare as Title XVIII and Medicaid as Title XIX — the federal government became, overnight, the largest purchaser of health care in the country.¹It also inherited fee-for-service medicine’s central incentive: providers were paid for each service, and nothing in the payment itself asked whether the service was needed. Congress built a check into the law from the start. To participate in Medicare, a hospital had to operate a utilization review plan — a staff committee, including at least two physicians, that reviewed the medical necessity of admissions and lengths of stay.² Before Medicare, only about a thousand hospitals ran such committees; the condition of participation made them general practice.²

The 1965 committees were run by the treating institution’s own medical staff, and they reviewed care after it was delivered. Within a decade Congress decided that self-review was too weak, and moved the judgment outward — to organizations of physicians who did not work for the hospital.

The pivotal statute is the least remembered. The Social Security Amendments of 1972, Public Law 92-603, signed October 30, 1972, created Professional Standards Review Organizations — regional groups of physicians tasked with reviewing care furnished to Medicare and Medicaid patients.³The mandate, in the words of the Institute of Medicine’s later account of the program, was concrete:

Three phrases in that mandate — medically necessary, professionally recognized standards, and most economical manner — are the vocabulary of nearly every denial written since. The 1972 amendments also strengthened the older utilization-review requirements under Medicare and Medicaid.³What changed was who held the pen. Under the 1965 committees, a hospital’s own staff reviewed its own work; under PSROs, an outside physician organization could find a service unnecessary and Medicare would not pay. The denial had become external.

PSROs were controversial and uneven, and in 1982 the Tax Equity and Fiscal Responsibility Act folded them into a leaner Peer Review Organization program, sharpening the federal review apparatus just as Medicare shifted to paying hospitals a fixed price per diagnosis.³ But the public program was only half the story. The other half was being built in the private market, on a different premise.

A year after the PSRO law, the Health Maintenance Organization Act of 1973, Public Law 93-222, was signed on December 29, 1973.⁴ It provided federal grants and loans to prepaid plans, overrode state laws that had restricted them, and required employers of a certain size that offered health insurance to also offer a federally qualified HMO. In his signing statement, President Nixon described the model in plain terms:

Prepayment changed the economics of review. Under fee-for-service, a payer that denied a claim simply avoided a cost; under capitation, the plan held a fixed sum per member and kept whatever it did not spend on care. Utilization management — prior authorization, concurrent review, case management — became the plan’s core operating discipline rather than a regulatory obligation. The tools that public review had introduced were now run by organizations with a direct financial interest in the answer.

The single fastest change in this history happened in private employer plans in the mid-1980s, and it is documented with unusual precision in the Institute of Medicine’s 1989 report on utilization management. Review organizations had done some preadmission review in the 1970s, the report notes, but “widespread application of this approach to managing health care utilization is a phenomenon of the 1980s.”⁵ The numbers are stark:

Over the same span, the report estimated, the share of employees covered by prior-review programs rose from about 5 percent in 1984 to between half and three-quarters of the workforce.⁵In five years, asking a payer’s permission before a hospital admission went from an exception to a norm. This is the era that produced the everyday experience of the denial: not a rare audit finding, but a routine gate the treating physician had to clear in advance, on the payer’s terms and timeline.

By the end of the decade the two systems — public peer review and private managed care — shared a common grammar of medical necessity and a common set of tools. What they still lacked was scale. Human reviewers could not read every claim.

On January 1, 1996, the Health Care Financing Administration — the agency later renamed the Centers for Medicare & Medicaid Services — implemented the National Correct Coding Initiative for Medicare Part B practitioner claims.⁶ HCFA had contracted in 1994 for the development of national correct-coding rules; the result was a set of automated edits built into claims processing.⁷ The edits compare each submitted claim against tables of code pairs that should not be reported together, flagging them for denial or bundling before a person ever looks at the claim. A contemporaneous account in the Journal of Oncology Practice put the function plainly:

NCCI was aimed at coding errors — unbundling, duplicate reporting, mutually exclusive procedures — not at medical necessity, and it began as a Medicare tool before spreading to hospital outpatient claims in 2000 and, later, to Medicaid.⁶ But its significance for this history is structural rather than clinical: it made the denial a computational event. A claim could now be refused in milliseconds, at national scale, by a rule table rather than a committee. Every automated payer edit since inherits that design.

The four moves — internal committee review in 1965, external physician review in 1972, prepaid managed care in 1973, and automated coding edits in 1996 — are the load-bearing structure of the modern denial. Everything after them is elaboration on the same frame: more code sets, more prior-authorization categories, faster edits, and, most recently, predictive models applied to coverage decisions. That newest chapter — algorithmic review and the public record around it — is the subject of a companion account, When Payers Automate Denials, and is not re-reported here.

What the sixty-year record shows is that none of this arrived by accident, and none of it was designed to refuse correct claims. Each layer was a considered response to a real problem — overuse, cost, coding fraud — and each layer left the treating physician one more step removed from the decision to pay. The denial is not a glitch in American health care. It is one of its oldest deliberate institutions, and it was built, piece by documented piece, out in the open.