Why Two Hospitals Pay Different Prices for the Same Stent

In 2011, an average California patient admitted for a percutaneous coronary intervention — an angioplasty with a drug-eluting stent, one of the most standardized procedures in modern medicine — could be charged between $22,047 and $165,386 depending solely on which of 124 hospitals the ambulance reached. The median charge was $88,350. The patients had no major complications. The stents were comparable. The seven-fold difference in the sticker price was real, documented, and, for the most part, unexplained.

That finding comes from a PLOS ONE study that worked through California hospital billing data and then did something unusual: it checked which explanations actually fit. Private insurers, who negotiate discounts off those charges, paid roughly one-third of the sticker price on average — which only relocated the mystery. Hospital ownership, Medicare patient share, cost of living, and rural location all correlated with charges. But every observable hospital and market factor the researchers could measure explained only 43 percent of the variation between hospitals. The rest had no name.

This is the peculiar structure of hospital purchasing. The stent is a commodity manufactured to precise tolerances by a handful of global firms. The price attached to it is not.

The international control group

How much of that difference is genuinely American? Two London-based researchers, Elodie Grennan and Ashley Swanson, assembled quarterly survey data from hospitals in five countries between 2006 and 2014 and published the results in Health Affairs. The control group was unforgiving.

Drug-eluting stents offer the cleanest comparison. In 2006, the average German hospital paid $1,440 for a drug-eluting stent; the average American hospital paid about $1,000 more. By 2014, the German price had fallen to $340. The American price had fallen too — but remained roughly $1,000 above Germany's, about four times as much. For pacemakers, American hospitals paid up to six times what German hospitals paid. The authors' conclusion was blunt: manufacturers, they wrote, "exploit varying levels of willingness-to-pay and bargaining power between buyers to charge different prices across hospitals and increase profits."

The manufacturer prices the buyer, not the product. A hospital with strong GPO contracts, high procedure volume, and a credible threat to switch vendors pays one number. A hospital with none of those pays another. Neither number is published.

The benchmark that exposed the service side

If devices are priced opaquely, services are priced in fog. The RAND Corporation's Hospital Price Transparency Study — funded by the Robert Wood Johnson Foundation, covering more than 4,000 hospitals and 4,000 ambulatory surgery centers with 2020–2022 claims — gave the fog a measurement. In 2022, employers and private insurers paid hospitals an average of 254 percent of what Medicare would have paid for the same inpatient and outpatient services at the same facilities.

The averages hide the spread. Georgia hospitals averaged 344.73 percent of Medicare; Michigan, 189 percent. RAND's principal investigator, Brian Briscombe, noted the spread with visible restraint: "There's so much variation. There might be valid reasons why one hospital is more expensive than the other. But there's definitely some room for shopping." RAND's health care director, Peter S. Hussey, called the study a tool for employers to become "better-informed purchasers."

The same report measured what hospitals charge for drugs administered on-site: commercial plans paid an average of 278 percent of the average sales price, versus 106 percent for Medicare. The stent on the tray and the drug in the IV bag are subject to the same pricing logic — charge what the particular buyer's ignorance and urgency will bear.

The transparency rule changed the paperwork, not the game

On January 1, 2021, a Centers for Medicare & Medicaid Services rule tried to attack the root cause directly. Every hospital in the country was required to post a machine-readable file containing its negotiated rates for every item and service, plus a consumer-friendly display of 300 shoppable services. No more secret menus.

The rollout was a case study in how a good idea meets an indifferent industry. CMS spent its first year sending warning letters. Its first fines did not land until June 7, 2022, against two Georgia hospitals — $883,180 for Northside Hospital Atlanta, $214,320 for Northside Hospital Cherokee — after both ignored a corrective-action request. As of early 2026, only 28 hospitals had been fined in total, with penalties ranging from $32,301 to $979,000 (the latter against UF Health North in Jacksonville, Florida). A February 2025 executive order demanded stronger enforcement, and the 2026 outpatient payment rule tightened the standard: actual prices instead of estimates, median and percentile breakdowns, and a named CEO attesting the data is accurate.

But a posted number is not the same as a usable benchmark. Payer-posted files run to terabytes of redundant data. Hospital machine-readable files are notoriously inconsistent. And knowing your rival's price still does not tell you what the manufacturer will quote you. Transparency exposed the variation. It did not equalize the leverage.

The honest counterargument

There is a legitimate side to this story, and it deserves a full hearing. Some price variation is not mystery at all: it is cost. Teaching hospitals train the next generation of physicians. Level I trauma centers keep a surgical team standing idle at 3 a.m. in case of a highway pileup. Hospitals in high-wage markets pay high wages. Hospitals that care for the uninsured absorb costs no one pays them for.

More pointedly, hospitals argue the whole frame is rigged. Medicare, the benchmark everything is measured against, underpays by their own accounting — the American Hospital Association put Medicare underpayments at nearly $130 billion in 2023, with Medicare paying roughly 82 cents for every dollar hospitals spent on patient care. If the public program does not cover costs, commercial rates are not so much "inflated" as the only place left to recover them. In this telling, the two hospitals paying different prices for the same stent are both responding rationally: one is subsidizing missions the other does not carry, and the stent line item is just where the subsidy landed.

That argument has force, and any benchmark honest enough to be useful has to absorb it. But even RAND's researchers, generous to the hospital case, could not make the legitimate factors cover the spread. Legitimate variation explains some of the 43 percent that observable factors capture. It cannot explain why, as RAND found, one hospital in a city gets three times the Medicare rate while another hospital across town with the same quality rating gets twice the rate.

The price of a stent is a measurement problem disguised as a negotiation problem. Nobody knows the going rate, so there is no going rate — only thousands of private ones. Until the going rate is visible, two hospitals fifteen miles apart will keep paying two different prices for the same wire mesh, and neither of them will know which one is wrong.