Hospital charges vs what Medicare paid, by condition: 30 billing groups in 2024

Three columns, three different measures

The CMS file behind this guide, Medicare Inpatient Hospitals by Geography and Service, reports four numbers for each MS-DRG, the billing group a stay is classified into for payment: a count of discharges, an average covered charge, an average total payment and an average Medicare payment. It does so for the nation and for each state. The 2024 national rows, which are the ones used here, number 767 and add up to 6,905,762 discharges. The smallest national row holds 12 discharges.

CMS's data dictionary defines the three dollar columns as follows, and the definitions are the reason this guide divides one column by another and not a different pair.

Two features of those definitions drive everything below. The first is that the charge column reflects each hospital's own price list. The dictionary says charges "vary from hospital to hospital because of differences in hospital charge structures," which is a statement about price lists and not about care. The federal price transparency rule has a related term for a single item, the gross charge, defined in 45 CFR 180.20 as "the charge for an individual item or service that is reflected on a hospital's chargemaster, absent any discounts." The two are neighboring ideas and not the same measurement: one is a stay-level average CMS builds from claims, the other is a line on a hospital's posted rate sheet.

The second feature is that the total payment column is not Medicare's payment alone. It adds the patient's deductible and copayment amounts and any third-party payments reported for coordination of benefits. The file does not break that remainder into its parts. For that reason the multiple used throughout is the average charge divided by the average total payment, with the Medicare payment shown in its own column beside it. Every multiple is computed from CMS's unrounded averages and shown to two decimals, and every dollar amount is rounded to the whole dollar. A multiple here is a ratio of two averages. It is not the multiple at any one hospital, and the state section further down shows how far single states sit from the national figure.

Scope matters before any number does. The CMS catalog describes the dataset as "hospital discharges for Original Medicare Part A beneficiaries by IPPS hospitals," with a data period of calendar year 2024. The CMS catalog lists the release with a modified date of 2026-04-23 and lists no later data year. It is a file of Original Medicare claims and not a record of what commercial plans pay. The group titles are CMS billing categories, printed here as CMS prints them, and a title says how a stay is classified for payment, not anything about the person who was admitted. On this site a list price is never presented as what a patient pays, so the charge column appears below only to show how far list-price levels sit from payment levels.

One reconciliation for readers of the site's guide to hospital stay cost per day. That guide prints $14,453.04 for MS-DRG 470, a payment built from the FY2027 relative weight and base rates for a hospital at a wage index of 1.0 with no add-ons. The $14,322 here is the average of actual Medicare payments on 2024 discharges, which the dictionary says include teaching, disproportionate share, capital and outlier payments. They are two different measurements from two different years, and the weight arithmetic stays in that guide.

Thirty largest groups, charge against payment

The table lists the 30 national MS-DRG rows with the most discharges in 2024, in descending order. Together they hold 3,043,287 of the 6,905,762 discharges in the file, or 44.1%. The rule for inclusion is volume alone: no group was added or dropped because of how its charges or payments look. Conditions a reader may be looking for appear under their CMS titles, such as septicemia (871 and 872), heart failure and shock (291), simple pneumonia and pleurisy (193 and 194), and major hip and knee joint replacement (470).

The multiple runs from 4.30 for MS-DRG 603, cellulitis without MCC, to 6.63 for MS-DRG 322, percutaneous cardiovascular procedures with intraluminal device without MCC. The median of the 30 is 4.94, with 18 of them below 5.00 and 12 at or above it. In dollars the spread is far wider than in ratios. Average charges run from $36,115 (MS-DRG 309) to $228,948 (MS-DRG 853), a 6.34-fold range, while average total payments run from $7,304 (MS-DRG 309) to $46,478 (MS-DRG 853), a 6.36-fold range. Charges and payments spread by nearly the same factor, and the same two groups sit at the bottom and the top of both ranges.

The single largest row is MS-DRG 871, septicemia or severe sepsis without MV >96 hours with MCC, with 578,073 discharges, 8.4% of every discharge in the file. Its average charge of $90,297 sits against a total payment of $18,228, a multiple of 4.95. The sibling group without MCC, MS-DRG 872, has 107,314 discharges, an average charge of $49,049 and a total payment of $10,091.

MS-DRG 470 ranks 22 by volume. Its average charge exceeds its average total payment by $75,820, and the charge is 6.51 times Medicare's own payment of $14,322. The file reports averages across hospitals and describes no individual bill, so it cannot say that difference was ever billed to anyone. It says only how far the list-price level sat above the payment level on the average 2024 discharge in this group.

How tightly the ratio clusters

Across all 767 national rows, weighting each group by its discharges, the average charge was $93,691 and the average total payment $18,728, a multiple of 5.00. The average Medicare payment was $15,500, which is 82.8% of the discharge-weighted average total payment. Without weighting, the median of the 767 group multiples is 4.97.

The multiples bunch. Of the 767 groups, 682, or 88.9%, fall from 4.00 up to but not including 6.00. Below 3.00 there are 8 groups, and none reaches 8.00. The lowest is 2.31, in MS-DRG 950 (aftercare without CC/MCC, 138 discharges), and the highest is 7.52, in MS-DRG 651 (kidney transplant with hemodialysis without MCC, 917 discharges). Both come from small groups, so the table restricts the extremes to the 507 groups with at least 1,000 discharges.

By title, the lowest six are groups for alcohol and drug rehabilitation, musculoskeletal aftercare, vaginal delivery, psychoses, rehabilitation and depressive neuroses. The highest six are two kidney transplant groups, two percutaneous cardiovascular procedure groups, a transient ischemia group and an inguinal and femoral hernia procedure group. The file gives no reason for either end of the range, and this guide does not supply one.

What the clustering supports is narrow. A ratio near 5 describes how national averages relate across most groups. It does not convert one hospital's charge into a payment, and no such conversion is offered here. The same group's ratio changes by state and by year, both of which are shown below, and a ratio quoted without a state and a year attached is incomplete.

Severity tiers move both columns together

Many MS-DRGs come in sets of two or three that share a base condition or procedure and differ by severity tier. CMS's FY 2027 inpatient payment final rule, recounting the change finalized for FY 2021, describes the structure as a "three-way severity level split" within a base MS-DRG: a subgroup for a major complication or comorbidity (MCC), one for a complication or comorbidity (CC), and a base group with neither. In a title, "with MCC" is the top tier, "with CC" the middle tier, and "without CC/MCC" or "without MCC" the base tier. The tier is a payment classification printed in the title. It says nothing about who has which tier or why.

The table compares the top tier with the base tier in 12 sets, each of which includes at least one group from the 30-row table above. It is not a list of every tiered set in the file.

Moving from the base tier to the top tier raised average charges by a factor between 1.36 and 2.50 across the 12 sets, and average total payments by a factor between 1.37 and 2.62. The two factors are close within each set. For septicemia, MS-DRG 871 against 872, charges are 1.84 times and payments 1.81 times. For heart failure and shock, 291 against 293, the pair is 2.06 and 2.05. In 5 of the 12 sets the charge factor is the larger of the two, and in the other 7 the payment factor is.

The widest differences run toward payment. For intracranial hemorrhage or cerebral infarction (064 over 066) the payment factor is 2.62 and the charge factor 2.36, and for acute myocardial infarction, discharged alive (280 over 282) the pair is 2.05 against 1.75. In both sets the base tier has the higher multiple, 5.89 against 5.30 for the first and 6.02 against 5.13 for the second.

The joint replacement set is worth reading closely. MS-DRG 469, major hip and knee joint replacement or reattachment of lower extremity with MCC or total ankle replacement, averages $149,677 in charges and $29,663 in total payments across 5,530 discharges. MS-DRG 470, without MCC, averages $93,184 and $17,364 across 64,419. The top tier carries 1.61 times the charge and 1.71 times the payment, so the base tier's multiple of 5.37 is higher than the top tier's 5.05.

A statement's total-charge line therefore mixes two things, the tier the stay was classified into and the hospital's own price list, and a single stay cannot separate them. Which tier the claim carries moves both the list-price level and the payment level, which is one reason to ask the billing office for the billing group in writing.

The same group in different states

The file also carries state-level rows, so the national average can be set beside the states it averages. For three high-volume groups, MS-DRG 470, 871 and 193, the table uses the state-level rows with at least 100 discharges. Each of the three has rows for 51 geographies, which the file labels as states and which are the 50 states and the District of Columbia. 49 of them clear the 100-discharge line for MS-DRG 470, 51 for MS-DRG 871 and 51 for MS-DRG 193.

For MS-DRG 470, state average charges run from $28,915 in Maryland to $183,374 in Nevada, a 6.34-fold range, while state average total payments run from $13,216 in Arkansas to $26,000 in Maryland, a 1.97-fold range. The pattern repeats for MS-DRG 871, where the charge range is 6.90-fold and the payment range 1.99-fold, and for MS-DRG 193, at 6.44-fold and 1.79-fold. In these three groups the charge range is between 3.22 and 3.59 times as wide as the payment range.

A concrete pair shows what that means. For MS-DRG 470, the average charge in Nevada was $183,374 and in Massachusetts $43,575, a difference of $139,799. The average total payment in the same two states was $18,747 and $17,875, a difference of $872.

Because the national figure weights every discharge equally, states with many discharges count for more. Florida (3,907 discharges, multiple 9.10) and Texas (4,151 discharges, 7.67) sit well above the national 5.37, and 17 of the 49 state rows are above it. The other 32 are at or below it.

Maryland stands apart in all three groups. Its multiple is 1.11 for MS-DRG 470, 1.10 for 871 and 1.11 for 193, against national figures of 5.37, 4.95 and 4.96. Among the state rows with at least 100 discharges, its average charge is the lowest in all three groups, and its average total payment for MS-DRG 470, $26,000, is the highest. The file states no reason. CMS's page on the Maryland Total Cost of Care Model supplies context: the model built on the Maryland All-Payer Model, launched in 2014, which established global budgets for certain Maryland hospitals, and it aimed "to establish pricing of medical services provided by hospitals, primary care doctors and specialists across all payers." The same page says CMS announced on March 12, 2025 its intention to end the Total Cost of Care Model as of December 31, 2025, and that the model transitioned to the AHEAD model with an implementation period that began in January 2026. The 2024 data predate that change, and this guide does not describe the AHEAD model.

Set Maryland aside and the gap between charge ranges and payment ranges remains. For MS-DRG 470 the other states' average charges still run 5.27-fold, from $34,801 in Rhode Island to $183,374 in Nevada, against a 1.76-fold range in total payments, from $13,216 in Arkansas to $23,235 in New York.

How the ratio changed since 2013

CMS publishes the same file for every data year from 2013 forward. The table puts six groups that have a national row in both 2013 and 2024 beside a line for every national row in each year. Dollar amounts are nominal, with no adjustment for the general price level.

Charges rose faster than payments in every row. For MS-DRG 470 the average charge went from $54,661 to $93,184, a factor of 1.70, while the average total payment went from $14,787 to $17,364, a factor of 1.17, so the multiple moved from 3.70 to 5.37. Across all national rows the average charge per discharge rose by a factor of 1.94 and the average total payment by 1.49. The all-rows multiple was 3.85 in 2013, then 4.22 in 2016, 4.48 in 2019, 4.67 in 2022 and 5.00 in 2024, a rise at every step shown.

Three cautions apply to those comparisons. First, the 2013 and 2024 files do not describe the same population. The number of national rows grew from 741 to 767, discharges in the file fell from 9,882,789 to 6,905,762, and discharges in MS-DRG 470 fell from 452,861 to 64,419. The file does not give a reason for those changes, so the mix of stays inside each group may differ between years, and a group's average can move for that reason as well as for changes in prices. Second, the files were released at different times: the CMS catalog lists the 2013 through 2021 files with a modified date of 2023-05-10, the 2022 file 2024-06-04, the 2023 file 2025-05-22 and the 2024 file 2026-04-23. Third, the dollars are nominal, so some of the dollar growth in both columns is the general price level, which the table does not remove.

The direction in the table is the file's own. The sizes should be read with those cautions, and the multiple is better thought of as a moving quantity than a fixed feature of hospital pricing.

What the Medicare column leaves out

Medicare's payment is the largest piece of the total payment but not all of it. Across all rows the discharge-weighted Medicare payment was 82.8% of the total payment. Among the 30 groups in the table, the gap between total payment and Medicare payment runs from $1,832 in MS-DRG 690 to $6,068 in MS-DRG 853, and Medicare's share of the total runs from 72.9% in MS-DRG 392 to 89.9% in MS-DRG 274. For MS-DRG 470 the gap is $3,042.

The dictionary says that remainder consists of the patient's deductible and copayment amounts and any additional third-party payments for coordination of benefits, and the file does not split it. CMS set the Medicare Part A inpatient hospital deductible at $1,632 for 2024, which covers the beneficiary's share of the first 60 days of Medicare-covered inpatient hospital care in a benefit period. In every one of the 30 groups the gap is larger than that deductible. The comparison shows the order of magnitude of the remainder and no more: a deductible applies once per benefit period, and the file does not say how many of these stays began a benefit period.

Two limits keep this from becoming a patient price. The Medicare payment is the program's payment to the hospital, so it is not the amount a patient owes, and the guide to hospital stay cost per day sets out the Part A schedule a Medicare patient actually follows. And the file does not measure what commercial plans pay for their own members, which is set in the plan's contract with the hospital and can sit above or below Medicare's amount.

When the charge column still matters

A list price is not what anyone owes, but it appears in three places where a patient may meet it, and each has a rule behind it.

Contracts that pay by percentage or formula. A negotiated charge in a hospital's price file is either a dollar amount or a percentage or algorithm. Where it is a percentage or algorithm, 45 CFR 180.50(b)(2)(ii)(C) requires the file to describe that percentage or algorithm and, from January 1, 2026, to encode the 10th percentile, median and 90th percentile allowed amounts in dollars, along with the number of remittances behind them. The guide to using a hospital's published prices shows how to find those fields. Where a contract is written as a percentage of something, the hospital's charge can be an input to the payment and still not a price, and the allowed-amount fields are the figures to look at.

Financial assistance at tax-exempt hospitals. Under 26 CFR 1.501(r)-5, a hospital facility run by a tax-exempt hospital organization must limit what it charges a patient eligible under its financial assistance policy for emergency or other medically necessary care to not more than the amounts generally billed to insured individuals, and for other care covered by the policy to less than the gross charges. The look-back method computes that limit by dividing the sum of insurer-allowed claims over a prior 12-month period by the associated gross charges, and applying the resulting percentage to the gross charges for the care. A second permitted method prices the care as Medicare fee-for-service or Medicaid would. The rule's own second example gets 40 percent for inpatient services from $100 million allowed against $250 million in gross charges. The method uses the same relationship this guide measures, allowed amount over charge, but it is computed from the individual hospital's own claims, not from a national table. The guide to hospital financial assistance and charity care explains how to find the policy and apply.

A statement for someone paying without insurance. A first statement can show total charges of tens of thousands of dollars before any discount is applied. That line is a list price. Hospital price files include a discounted cash price, defined in 45 CFR 180.20 as "the charge that applies to an individual who pays cash (or cash equivalent) for a hospital item or service," and that is the comparison for a self-pay reader. The guide to negotiating a hospital bill covers verifying the bill first, researching published prices for the same codes, and then asking for a reduction, a payment plan or assistance in writing. The Medicare columns on this page can show how list-price levels compare with payment levels, but they are not a quote and no price for self-pay patients is published here.

Putting your statement beside the table

A patient with a hospital statement and an explanation of benefits can compute one number of their own, the charge divided by the allowed amount, and see where it falls relative to the table. The example below uses invented figures for a stay billed under a commercial plan, with a deductible and coinsurance already reflected in the plan's payment.

The computed 5.33 sits near the national 5.37 for MS-DRG 470, and that proximity proves nothing about the bill. The amount the patient owes, $3,500, comes from the allowed amount and the plan's cost-sharing terms, and in this invented example the charge of $120,000 is the starting line from which the contract adjustment is taken. If the ratio had come out at 2 or at 9, it would say something about the hospital's price list relative to its payer and nothing about whether the patient's share was calculated correctly. The check that decides whether a share is right is the one in the guide to why a hospital bill is so high with insurance, which tests each step from allowed amount to patient responsibility.

Three questions to the billing office use the vocabulary of this table. Ask which MS-DRG the stay was assigned, since the tier changes both columns. Ask whether the contract that applies pays a case rate, a per-day rate or a percentage of charges, since the last ties the payment directly to the charges. And ask for the itemized bill, as set out in the guide to checking an itemized bill for errors, because an itemized bill is the document on which a duplicated or mistaken line can be found.

What the averages cannot tell you

The columns are averages of averages across hospitals, years and stays, and several things follow. They are Original Medicare Part A hospital inpatient figures for calendar year 2024 and are not a measure of what other payers pay. Their scope is the catalog's own description of the dataset, which is hospital discharges for Original Medicare Part A beneficiaries by IPPS hospitals. The charge column is a hospital price-list level, the total payment column includes amounts that are not Medicare's, and the multiple divides one by the other. A group's national figure hides the wide spread among its states, as the Nevada and Massachusetts pair shows. A group's figure from 2013 is not the same population as the figure from 2024.

The charge and payment fields belong to the hospital inpatient claim, as the dictionary's count of "discharges billed by all providers for inpatient hospital services" indicates. They are not a source for clinician fees, outpatient or emergency department charges, observation stays or prices for patients without coverage. Whether a stay is billed as inpatient at all is a separate question that decides which pricing system applies, and the guide to observation versus inpatient status covers it. For the hospital's own posted prices for a specific code, the place to look is the hospital's machine-readable file, not this table.

Hospital charge and payment questions

Where each figure comes from

• CMS, Medicare Inpatient Hospitals by Geography and Service, data year 2024, catalog modified date 2026-04-23, API dataset id 0a011f40-7063-4c83-bc05-529f3522b727. National and state rows, 767 national rows. The CSV release (MUP_INP_RY26_P03_V10_DY24_Geo.CSV) matches the API rows. Checked 2026-11-03.
• CMS, Medicare Inpatient Hospitals by Geography and Service data dictionary (PDF posted December 2020), and the Medicare Inpatient Hospital by Provider and Service data dictionary (PDF posted July 2022), which carries identical wording for the three dollar fields. Definitions quoted in full. Checked 2026-11-03.
• CMS data.gov catalog (data.cms.gov/data.json): release dates for the 2013 through 2024 files and the dataset scope statement. The 2013, 2016, 2019 and 2022 files were downloaded from the catalog's CSV links to compute the trend table. Checked 2026-11-03.
• CMS, FY 2027 IPPS/LTCH PPS final rule, FR Doc. 2026-15833 (govinfo.gov text, August 4, 2026), discussion of the three-way severity level split. Checked 2026-11-03. CMS, FY 2027 IPPS Table 5, correction notice version, for the complete MS-DRG 469 title, which the dataset truncates. Checked 2026-11-03.
• CMS, 2024 Medicare Parts A and B premiums and deductibles: inpatient hospital deductible of $1,632 for 2024. Checked 2026-11-03.
• CMS, Maryland Total Cost of Care Model: model description, All-Payer Model background, and end and transition dates. Checked 2026-11-03.
• eCFR, 45 CFR 180.20 (definitions) and 45 CFR 180.50 (required data elements), version in effect October 28, 2026. Checked 2026-11-03.
• eCFR, 26 CFR 1.501(r)-5, limitation on charges, latest amendment dated 2016-07-12 in the eCFR version history. Checked 2026-11-03.

General billing information, not medical, legal or insurance advice. These are Original Medicare averages for 2024, not a quote, a bill or a price for any hospital, and the charge column is a list price that no source here presents as an amount owed. Confirm your price in writing.