Skip to content
Back to CMS Watch
OIGSeptember 21, 2026·7 min read

OIG: Methodist Hospital (San Antonio) billed at least $12.4M in estimated Medicare overpayments

OIG report A-09-23-03001 (issued September 14, 2026; posted September 17, 2026) found that Methodist Hospital in San Antonio, Texas, did not fully comply with Medicare billing requirements for 27 of 100 sampled inpatient, IRF, and outpatient claims, producing $256,926 in sample net overpayments. OIG estimated at least $12.4 million ($12,499,714) in net overpayments for selected high-risk claim types in 2020 and 2021. This is an OIG performance-audit estimate, not an adjudicated recovery or fraud finding. The Hospital did not concur with the refund and post-period audit recommendations.

OIGhospital billingIRFMedicare Part ADRG
HCC Buddy

By the HCC Buddy Coding Team

Published September 21, 2026

Hospital coding binder and claim checklist on a desk, depicting the OIG Methodist Hospital Medicare billing compliance audit
OIG sampled 100 Methodist Hospital claims across IRF, inpatient, and outpatient high-risk areas for calendar years 2020 and 2021.Illustration: HCC Buddy

Key Takeaways

  • HHS-OIG report A-09-23-03001 (issued September 14, 2026; posted September 17, 2026) estimated that Methodist Hospital (San Antonio, Texas) received at least $12.4 million ($12,499,714) in net Medicare overpayments for selected high-risk inpatient, IRF, and outpatient claim types during January 1, 2020, through December 31, 2021.
  • In the sample, 27 of 100 claims did not fully comply with Medicare billing requirements, producing $256,926 in sample net overpayments; 73 claims complied. Four of the 27 erred claims involved requirements that were not conditions of payment.
  • IRF claims drove most of the sample dollars: 16 of 20 IRF claims were incorrectly billed ($247,457). The payment driver on 12 of 20 was medical necessity: patients did not reasonably require supervision by a rehabilitation physician (42 CFR § 412.622(a)(3)(iv)). Intensity and participation rules in § 412.622(a)(3)(ii) through (iii) were waived during the COVID-19 PHE audit window.
  • Inpatient errors (8 of 60; $8,850 net) included unsupported diagnosis/procedure codes that changed DRG assignment, unsupported outlier units/charges, and one incorrect discharge-status code. Outpatient errors (3 of 20; $619) were unsupported HCPCS and medical-necessity failures.
  • This is an OIG performance-audit statistical estimate at the lower limit of a two-sided 90% confidence interval. It is not an adjudicated recovery, not a fraud finding, and not an FCA settlement. CMS action officials (through the MAC) determine whether an overpayment exists. The Hospital did not concur with the refund and post-period internal-audit recommendations.

On September 14, 2026, HHS Office of Inspector General issued report A-09-23-03001, a Medicare hospital provider compliance audit of Methodist Hospital in San Antonio, Texas. The report was posted September 17, 2026. The full PDF is A-09-23-03001.

OIG reviewed a stratified random sample of 100 inpatient, inpatient rehabilitation facility (IRF), and outpatient claims with Medicare payments totaling $1,426,020. The Hospital complied for 73 claims. For the remaining 27 claims, it did not fully comply with Medicare billing requirements, producing $256,926 in sample net overpayments. On that basis, OIG estimated that the Hospital received at least $12.4 million ($12,499,714) in net overpayments for selected high-risk claim types during the audit period (January 1, 2020, through December 31, 2021).

Read that dollar figure carefully. OIG estimates overpayments at the lower limit of a two-sided 90-percent confidence interval. Footnote 17 states the exact estimate as $12,499,714 and explains that lower limits calculated this way are designed to be less than the actual overpayment total 95 percent of the time. This is a performance-audit projection from a sample, not a court judgment, not a CMS final determination, and not a fraud allegation. Footnote 22 says OIG audit recommendations do not represent final determinations; CMS, acting through a Medicare Administrative Contractor or other contractor, decides whether an overpayment exists and how to recoup.

What OIG sampled

Methodist Hospital is described in the report as a 1,976-bed short-term acute-care hospital in San Antonio, Texas. During the audit period OIG says Medicare paid the Hospital about $621 million across 40,110 inpatient and 97,557 outpatient claims. The sampling frame for this audit was narrower: 6,625 claims totaling $62,048,673 in selected OIG-designated high-risk areas.

OIG selected 100 claims for review:

  • 20 IRF claims
  • 60 inpatient claims (CERT error-prone DRGs, high-severity DRGs, severe malnutrition, mechanical ventilation, and related risk areas)
  • 20 outpatient claims with bypass modifiers

An independent medical review contractor validated the sampled claims against the medical records. This audit is part of OIG's ongoing hospital compliance series targeting facilities with a high volume of claims in previously identified high-risk billing patterns. It is not a Medicare Advantage risk-adjustment diagnosis-code audit (those companion MA reports for HumanaChoice and UnitedHealthcare of Wisconsin are already on /news).

Sample results by claim type

Claim typeSampledIncorrectly billedSample net overpayments
IRF2016$247,457
Inpatient608$8,850
Outpatient203$619
Total10027$256,926

Appendix D (Table 5) breaks the same sample by risk area. IRF alone accounts for nearly all of the sample overpayment dollars. Four of the 27 claims in error involved requirements that were not conditions of payment (OIG still counted them as compliance errors but did not treat the amounts as overpayments for those specific findings).

Appendix C (Table 4) shows the extrapolated frame estimate: point estimate $18,766,379, lower limit $12,499,714, upper limit $25,033,044. OIG recommends recovery at the lower limit.

IRF claims: rehab-physician supervision drove the dollars

For 16 of 20 selected IRF claims, the independent medical review contractor determined that the Hospital incorrectly billed Medicare Part A for stays that did not meet Medicare criteria for inpatient rehabilitation. Overpayments on those claims totaled $247,457.

The load-bearing pattern for the coding and utilization desk:

  • Medical necessity not met (12 of 20): OIG's payment finding is that, at admission, patients did not reasonably require supervision by a rehabilitation physician under 42 CFR § 412.622(a)(3)(iv). OIG's example: an enrollee admitted after a fall caused pelvic fractures managed non-operatively, with no significant change from baseline; the reviewer determined a non-rehabilitation physician could have overseen recovery and monitoring, so an IRF stay was not required. The audit period fell during the COVID-19 Public Health Emergency, when CMS waived the intensive-therapy participation rules in § 412.622(a)(3)(ii) through (iii). Readers should not invent an intensity or interdisciplinary-team payment test for these 12 claims.
  • IRF-PAI discharge assessment timing (10 of 20): Medicare requires the IRF Patient Assessment Instrument (IRF-PAI) discharge assessment to be completed by the 5th calendar day after discharge. OIG counted late assessments as compliance errors but states the amounts remained allowable because IRF-PAI timing is not a condition of payment.
  • Other Matters (individualized plans of care): Documentation for all 20 IRF claims, per the reviewer, did not support that the plan of care was individualized. OIG places this under Other Matters. It is not the payment finding and is not the $247,457 driver. Interdisciplinary-team language appears in the plan-of-care regulation background (42 CFR §§ 412.622(a)(4)(ii)(A) and (B)), not as the 12-claim payment test.

If your desk touches IRF CMG assignment or admission screening for this audit window, the takeaway is documentation supporting a reasonable expectation that the patient required rehab-physician supervision under § 412.622(a)(3)(iv), not hospital-level rehab intensity as the payment test.

Inpatient claims: unsupported codes, outliers, and discharge status

For 8 of 60 selected inpatient claims, the reviewer found incorrect Part A billing, producing $8,850 in net overpayments (the inpatient total includes both overpayments and underpayments).

Desk-level patterns OIG called out:

  • Diagnosis and procedure codes not supported (4 of 60; $5,622 net): Claims data must support the DRG. OIG's example: a claim carried an unsupported principal diagnosis of atrial fibrillation when the actual reason for admission was cancer-related leg pain, which would have mapped to a lower-paying DRG.
  • Outlier payments not supported (3 of 60; $822 net): Units of service and charges used to support outlier payments were incorrect.
  • Discharge status code not supported (1 of 60; $2,406): The Hospital coded discharge to home when the status should have been to home-health services from a home health agency (post-acute transfer rules under 42 CFR § 412.4(c)).

Appendix D also shows the inpatient risk-area split: high-severity-level DRG codes (6 incorrectly billed; $8,225), CERT error-prone DRGs (1; $157 underpayment), severe malnutrition (1; $782), with mechanical ventilation and home-health-resume strata showing zero incorrectly billed claims in sample.

OIG separately notes (Other Matters / footnote 21) that 19 of 60 inpatient claims did not meet InterQual Level of Care screening criteria. Failure to meet InterQual is not the same as failing Medicare inpatient criteria; OIG presents it as quality-assurance information for the Hospital, not as the payment finding.

Outpatient claims: HCPCS support and medical necessity

For 3 of 20 selected outpatient claims, the reviewer found incorrect Part B billing totaling $619:

  • HCPCS not supported (2 claims; $318): Medical record documentation did not sufficiently support the HCPCS code billed.
  • Medical necessity not supported (1 claim; $301): OIG's example describes a CT abdomen plus IV antibiotic administration claim that lacked clinical indications (such as abdominal complaints or signs of infection) to justify the services.

Bypass-modifier outpatient claims were the sampled outpatient risk area (Appendix D). The dollar scale is small next to IRF, but the pattern is the same desk rule: the code and the medical necessity both have to be in the note.

OIG recommended that the Hospital:

1. Refund the estimated $12,499,714 in net overpayments (excluding amounts presumed unrecoverable under the § 1870 waiver),

2. Consider internal audits for claims after the audit period based on the identified risks, and

3. Provide additional training on Medicare billing requirements.

The Hospital did not concur with the first and second recommendations and said it is willing to conduct education to address the third. The report's Hospital Comments section also challenges extrapolation, reopening authority, and PHE-era IRF-PAI expectations. OIG responds that its sampling methodology was statistically valid, that CMS kept most requirements in place during the PHE, and that providers retain five levels of appeal. OIG also states it revised the final report by removing two Two-Midnight Rule inpatient claims from the findings before publishing.

Again: an OIG recommendation is not a CMS recoupment letter. Novitas (the MAC named in the report) and CMS action officials decide whether and how to recover.

Why facility coders should care this week

This is the same OIG hospital high-risk claim series your HIM and IRF teams already watch, not a new MA plan diagnosis series. The Monday-morning checks are concrete:

  • IRF admissions: can the record defend a reasonable expectation of rehab-physician supervision under § 412.622(a)(3)(iv)? For PHE-era stays, do not treat waived intensity or participation rules as the payment test; keep individualized plan-of-care gaps in the Other Matters bucket.
  • Inpatient DRG: does the principal diagnosis and any CC/MCC or procedure code match what the attending documented as the reason for admission and what was actually performed?
  • Outliers: do units and charges supporting outlier payment match the record?
  • Discharge status: does the patient status code match the actual post-acute plan (home vs home health vs transfer)?
  • Outpatient: does every HCPCS on the claim have documentation and medical necessity in the note?

Keep the companion MA diagnosis-code audits in your peripheral vision via /news/oig-humana-choice-diagnosis-codes-130-9-million-2026 and /news/oig-unitedhealthcare-wisconsin-diagnosis-codes-46-9-million-2026, but do not blur those risk-adjustment patterns into this hospital Part A/B billing audit.

What coders should do now

  1. 1Pull a sample of recent IRF admissions and confirm the record supports a reasonable expectation that the patient required supervision by a rehabilitation physician (42 CFR § 412.622(a)(3)(iv)) before CMG assignment ships. For PHE-era stays, do not treat waived intensity or participation rules as the payment test; keep individualized plan-of-care documentation in the Other Matters bucket, not as the overpayment driver.
  2. 2For inpatient DRG queries, confirm the principal diagnosis matches the reason for admission in the attending note (OIG's atrial fibrillation vs cancer-related leg pain example is the pattern to avoid).
  3. 3Re-check discharge status codes on inpatient claims with home-health plans that start within three days; a 'home' status can incorrectly avoid post-acute transfer payment adjustments.
  4. 4For outpatient claims using bypass modifiers, confirm each HCPCS has supporting documentation and a clear clinical indication in the record before drop.
  5. 5If your facility sits in an OIG high-risk hospital series, map OIG's risk areas (IRF, CERT-prone DRGs, high-severity DRGs, malnutrition, outliers, bypass modifiers) to your internal audit queue for dates after December 31, 2021.

Frequently Asked Questions

Is the OIG Methodist Hospital $12.4 million figure a final CMS recoupment?

No. Report A-09-23-03001 is an OIG performance-audit estimate at the lower limit of a two-sided 90% confidence interval ($12,499,714). Footnote 22 states OIG recommendations are not final Medicare determinations. CMS, acting through a MAC or other contractor, decides whether an overpayment exists and how to recoup, and the provider can appeal.

What drove most of the Methodist Hospital sample overpayments?

IRF claims. OIG found 16 of 20 sampled IRF claims incorrectly billed, with $247,457 in sample overpayments. For 12 of 20, the payment finding was that patients did not reasonably require rehab-physician supervision under 42 CFR § 412.622(a)(3)(iv). Intensity and participation rules in § 412.622(a)(3)(ii) through (iii) were waived during the COVID-19 PHE. Inpatient ($8,850 net across 8 of 60) and outpatient ($619 across 3 of 20) errors were much smaller in the sample.

Did Methodist Hospital agree with OIG's recommendations?

No on the money and the post-period audits. The Hospital did not concur with OIG's recommendations to refund the estimated $12,499,714 and to consider internal audits after the audit period. It said it is willing to conduct education to address the training recommendation. Disagreement is not the same as a court ruling or an admission.

Is this the same kind of OIG audit as the HumanaChoice and UHC Wisconsin diagnosis-code reports?

No. A-09-23-03001 is a fee-for-service hospital compliance audit of inpatient, IRF, and outpatient billing. The HumanaChoice (A-05-24-00010) and UnitedHealthcare of Wisconsin (A-07-24-01214) reports are Medicare Advantage high-risk diagnosis-code / risk-adjustment audits. Different program, different documentation test.

Which dates does the Methodist Hospital OIG audit cover?

The audit period is January 1, 2020, through December 31, 2021. The report was issued September 14, 2026, and posted September 17, 2026. OIG also recommends the Hospital consider reviewing similar risks on claims after that audit period.

Related topics:OIGhospital billingIRFMedicare Part ADRG
HCC Buddy

HCC Buddy Coding Team

Editorial

Every HCC Buddy news article is checked against the current CMS-HCC model and the active FY ICD-10-CM tabular release before it publishes.

Get CMS Updates in Your Inbox

RADV news, model changes, and coding guidance — within days of CMS publishing, not quarters.