The Extraction Index
Who profits from America's $3.24 trillion healthcare overspend. Ranked by documented dollars, updated with every issue.
The Extraction Index
Who profits from the $3.24 trillion gap, and how the documented dollars have accumulated over the past decade.
$519B documented annual extraction across 15 issues, against a $3.24T annual gap. Every figure traces to a published issue and its source.
- Industry-wide scale, not attributable to any single company. MedPAC measures coding intensity above the statutory adjustment, a different and broader measure than the per-company Kronick lines; the company lines do not sum into this one. See scorecard/backfill/medpac_industry_line.py.
- OptumRx processes ~22% of US prescriptions; named in the FTC's 2024 insulin suit; litigation active (Issue #4: The Middlemen)
- Optum owns Change Healthcare, which touches roughly a third of all US claims (Issue #5: The Paper Chase)
- 13.5% volume-weighted prior-auth denial rate across 61 MA contracts; 57.9% of appealed denials overturned (Issue #8: The Denial Machine)
- Largest MA insurer; $13.9B in 2021 differential-coding revenue per Kronick 2025 (Issue #11: The MA Overpayment)
- Optum pays its own providers more: payer-to-provider rates 17% above competitors (61% in concentrated markets) (Issue #12: The Consolidation Tax)
- ~90,000 employed physicians; profits on both sides of the specialist-pay structure (Issue #14: The Specialist Tax)
- Optum ASC acquisitions associated with 11% price increases to competing insurers (Health Affairs 2026) (Issue #15: The Facility Fee Scam)
- Lowest appeal-initiation rate of the Big 5 (2.7%); 64.7% of appealed denials overturned (Issue #8: The Denial Machine)
- Second-largest MA insurer; coding-intensity payment beneficiary (Issue #11: The MA Overpayment)
- $7.3B in specialty-drug markups at PBM-owned pharmacies vs independents for identical drugs, 2017-2022 (FTC Interim Report #1)
- CVS Caremark ~34%, Express Scripts ~24%, OptumRx ~22% of US prescription claims
- FTC insulin case: Express Scripts settled February 2026; litigation against Caremark and OptumRx remains active
- Documented in Issue #4: The Middlemen
- Received an estimated 6% of the $33B in 2021 Medicare Advantage differential-coding payments (Kronick 2025)
- Same parent as the CVS Caremark entry, which extracts separately through PBM spread pricing
- Documented in Issue #11: The MA Overpayment
- Caremark processes ~34% of US prescriptions; designs the formularies and owns the pharmacies that fill them (Issue #4: The Middlemen)
- Insurer + PBM + pharmacy under one roof; denial economics in MA (Issue #8: The Denial Machine)
- Express Scripts processes ~24% of US prescriptions; kept high-list-price insulins preferred because they carried the largest rebates (Issue #4: The Middlemen)
- Prior-authorization denial economics in Medicare Advantage (Issue #8: The Denial Machine)
How to read this: each line is cumulative extraction at the company's documented annual rate, climbing from the year a source first detected it to today; the number on the right is the total accumulated since then. Solid segments are the source's documented window. The Big 3 PBM line follows the FTC's reported acceleration (a 42% annual growth rate, 2017-2021, to a $7.3B total), with the per-year split reconstructed to those published figures. The Medicare Advantage coding lines (UnitedHealth, Humana, Aetna) anchor to Kronick 2025's measured 2021 insurer-level estimate. Dashed segments carry the documented annual rate forward through years the mechanism kept operating but no source has independently re-measured. The axis is logarithmic. The faint gray MA industry line is the all-insurer Medicare Advantage coding-intensity total (MedPAC, real per-year data) for scale; it is a broader, different measure than the company lines, which do not sum into it. The chart shows only mechanisms our published issues have sourced (2016-2026); it grows as issues publish and as we backfill earlier years. Full methods and data →
The same companies, ranked by total documented dollars. Each figure traces to a published issue and its source.
Differential-coding revenue in Medicare Advantage: payments received for coding the same patients as sicker than they would appear in traditional Medicare. 42% of the $33B industry-wide total in 2021.
Medicare Advantage differential-coding revenue: 19% of the $33B 2021 industry total, the second-largest share after UnitedHealth. Payment for coding the same patients as sicker than they would appear in traditional Medicare.
Specialty-drug markups at PBM-owned pharmacies relative to independent pharmacies for identical drugs: $7.3B over 2017-2022. The FTC did not publish a per-company split, so neither do we.
Medicare Advantage differential-coding revenue: 6% of the $33B 2021 industry total. This is CVS Health's insurance arm; CVS also appears separately for its Caremark PBM spread pricing.
Spread pricing billed to a single state Medicaid program above what pharmacies were paid: $224.8M in one year from Ohio alone (Kentucky's companion audit found another $123.5M). Caremark was Ohio Medicaid's dominant PBM.
Projected patient out-of-pocket savings from the February 2026 FTC settlement restructuring Express Scripts' PBM economics: $7B over ten years, a measure of what one PBM's practices were costing patients.
Settlements, consent decrees, and open government actions cited in our published issues. Settlements are one-time amounts, not annual figures, and most involve no admission of wrongdoing.
False Claims Act settlement, the record for Medicare Advantage risk adjustment. DOJ alleged ~500,000 post-visit addenda diagnoses producing ~$1B in improper payments, 2009-2018. Kaiser denied wrongdoing.
First FTC settlement of the Big 3 PBM insulin case; restructures PBM economics with projected $7B in patient savings over ten years. FTC litigation against CVS Caremark and OptumRx remains active.
Publicly acknowledged DOJ investigation into its Medicare Advantage risk-adjustment practices. Open as of publication.
False Claims Act settlement over alleged inflated risk scores in a regional New York MA plan; includes a separate settlement by a named former executive.
False Claims Act settlement over alleged inflated Medicare Advantage risk scores through improper diagnosis submissions.
False Claims Act settlement over diagnosis codes submitted for risk adjustment that were not supported by the medical record. Sutter denied liability.
Pennsylvania AG petition alleged ~$40M/yr in property-tax avoidance unjustified by charitable mission; settled with a 10-year consent decree on insurance access and no quantitative community-benefit floor.
False Claims Act settlement over alleged one-way chart reviews and improper diagnosis-code guidance to increase Medicare Advantage risk scores.
Companies profiled in our "Who Profits" sidebars as beneficiaries of a documented mechanism, ranked by number of appearances. Tap a company to see each issue. The dollar shown with each issue is the size of the documented problem, not that company's share of it.
UnitedHealth Group / OptumUNH7 issues
Issue #4: The Middlemen (a $30B/yr problem) · OptumRx processes ~22% of US prescriptions; named in the FTC's 2024 insulin suit; litigation active
Issue #5: The Paper Chase (a $200B/yr problem) · Optum owns Change Healthcare, which touches roughly a third of all US claims
Issue #8: The Denial Machine (a $32B/yr problem) · 13.5% volume-weighted prior-auth denial rate across 61 MA contracts; 57.9% of appealed denials overturned
Issue #11: The MA Overpayment (a $28B/yr problem) · Largest MA insurer; $13.9B in 2021 differential-coding revenue per Kronick 2025
Issue #12: The Consolidation Tax (a $13B/yr problem) · Optum pays its own providers more: payer-to-provider rates 17% above competitors (61% in concentrated markets)
Issue #14: The Specialist Tax (a $28B/yr problem) · ~90,000 employed physicians; profits on both sides of the specialist-pay structure
Issue #15: The Facility Fee Scam (a $2.55B/yr problem) · Optum ASC acquisitions associated with 11% price increases to competing insurers (Health Affairs 2026)
HCA HealthcareHCA5 issues
Issue #3: The 254% Problem (a $73B/yr problem) · Largest for-profit hospital system; primary beneficiary of commercial rates at 254% of Medicare
Issue #10: The Procedure Mill (a $7.6B/yr problem) · 190 hospitals and 121 surgery centers generate margin from procedure volume, including low-value procedures
Issue #12: The Consolidation Tax (a $13B/yr problem) · Cross-market mergers associated with 5-17% commercial price uplift
Issue #14: The Specialist Tax (a $28B/yr problem) · ~50,000 affiliated physicians concentrated in high-RVU procedural specialties
Issue #15: The Facility Fee Scam (a $2.55B/yr problem) · Employed-physician conversions shift clinic billing to hospital facility rates immediately
Tenet Healthcare / USPITHC4 issues
Issue #10: The Procedure Mill (a $7.6B/yr problem) · USPI runs the largest US outpatient surgery network; low-value procedures are high-margin in that setting
Issue #12: The Consolidation Tax (a $13B/yr problem) · Consolidation-driven commercial pricing
Issue #14: The Specialist Tax (a $28B/yr problem) · ASC compensation structures tied to high-RVU procedure volume
Issue #15: The Facility Fee Scam (a $2.55B/yr problem) · USPI portfolio bills facility rates for services in MedPAC's site-neutral candidate set
The Cigna Group / Express ScriptsCI2 issues
Issue #4: The Middlemen (a $30B/yr problem) · Express Scripts processes ~24% of US prescriptions; kept high-list-price insulins preferred because they carried the largest rebates
Issue #8: The Denial Machine (a $32B/yr problem) · Prior-authorization denial economics in Medicare Advantage
CVS Health / Caremark / AetnaCVS2 issues
Issue #4: The Middlemen (a $30B/yr problem) · Caremark processes ~34% of US prescriptions; designs the formularies and owns the pharmacies that fill them
Issue #8: The Denial Machine (a $32B/yr problem) · Insurer + PBM + pharmacy under one roof; denial economics in MA
HumanaHUM2 issues
Issue #8: The Denial Machine (a $32B/yr problem) · Lowest appeal-initiation rate of the Big 5 (2.7%); 64.7% of appealed denials overturned
Issue #11: The MA Overpayment (a $28B/yr problem) · Second-largest MA insurer; coding-intensity payment beneficiary
CommonSpirit Healthprivate nonprofit2 issues
Issue #12: The Consolidation Tax (a $13B/yr problem) · One of the largest products of hospital merger waves; 137 hospitals in 24 states
Issue #13: The Nonprofit Lie (a $5.38B/yr problem) · Property, sales, and state tax exemptions persist across 137 facilities regardless of operating losses
WaystarWAY1 issue
Issue #5: The Paper Chase (a $200B/yr problem) · 4.5B+ transactions/yr; demand grows as prior-auth volume and denial rates grow
R1 RCMprivate (since 2024)1 issue
Issue #5: The Paper Chase (a $200B/yr problem) · An $8.9B valuation is a bet that billing complexity persists
Novo NordiskNVO1 issue
Issue #7: The GLP-1 Gold Rush (a $40B/yr problem) · Semaglutide at $1,000-1,350/month in the US vs $190-300 internationally; 80-90% US gross margin
Eli LillyLLY1 issue
Issue #7: The GLP-1 Gold Rush (a $40B/yr problem) · Tirzepatide at $1,000-1,400/month in the US vs $400-600 internationally; patents to 2036-2038
Elevance HealthELV1 issue
Issue #8: The Denial Machine (a $32B/yr problem) · Prior-authorization denial economics in Medicare Advantage
Premier Inc.private (since Nov 2025)1 issue
Issue #6: The Supply Closet (a $28B/yr problem) · GPO fees are a percentage of contracted volume: revenue is higher when contracted prices are higher
Cardinal HealthCAH1 issue
Issue #6: The Supply Closet (a $28B/yr problem) · Thin-margin, high-volume distribution; no incentive for hospitals to consume less
Owens & MinorOMI1 issue
Issue #6: The Supply Closet (a $28B/yr problem) · Hospital supply distribution; divesting the segment under structural margin pressure
DaVitaDVA1 issue
Issue #14: The Specialist Tax (a $28B/yr problem) · Captures specialist fee-for-service economics at the corporate level through dialysis bundles
American Medical Association501(c)(6)1 issue
Issue #14: The Specialist Tax (a $28B/yr problem) · Convenes the RUC and collects a royalty every time a CPT code is billed to any US payer
Advocate Healthprivate nonprofit1 issue
Issue #12: The Consolidation Tax (a $13B/yr problem) · Product of one of the largest cross-market hospital mergers
RadNetRDNT1 issue
Issue #10: The Procedure Mill (a $7.6B/yr problem) · 418 imaging centers; imaging is the largest category on the low-value-care list
Marsh McLennan / MercerMMC1 issue
Issue #9: The Employer Trap (a $6.6B/yr problem) · Dual compensation: plan-sponsor consulting fees plus carrier-paid indirect compensation; disclosed, not prohibited
Willis Towers WatsonWTW1 issue
Issue #9: The Employer Trap (a $6.6B/yr problem) · Benefits-consulting compensation rises when premiums rise; structural carrier-commission incentive
AonAON1 issue
Issue #9: The Employer Trap (a $6.6B/yr problem) · Benefits architect, exchange operator, and carrier-facing broker in the same client relationship
Kaiser Foundation Hospitalsprivate nonprofit1 issue
Issue #13: The Nonprofit Lie (a $5.38B/yr problem) · One Form 990 covers all 43 hospitals; per-facility community benefit is invisible at that filing level
UPMCprivate nonprofit1 issue
Issue #13: The Nonprofit Lie (a $5.38B/yr problem) · PA AG alleged ~$40M/yr in unjustified property-tax avoidance; settled by consent decree with no community-benefit floor
Ascension Healthprivate nonprofit1 issue
Issue #13: The Nonprofit Lie (a $5.38B/yr problem) · Consolidated Form 990 across 19 states; state regulators end up doing the audit the IRS does not
The rules that decide what appears on this page.
The ranked tier includes companies with a specific dollar figure documented in a published issue or in the peer-reviewed or government source that issue is built on; each entry shows its source.
Watch List companies are profiled beneficiaries of documented mechanisms; no specific annual figure has been attributed to them yet.
Pools are never split across companies without a source that does the splitting.
Settlements are listed in the Enforcement Record, not annualized into the ranked tier.
Allegations are labeled as allegations. Companies may dispute any figure via the contact address; corrections are published.
The chart plots cumulative extraction at each company's documented annual rate, climbing from the year a source first detected it to the present. The annual rate is measured; the accumulation across later years assumes that documented rate persisted while the mechanism kept operating. Solid line segments are years a source measured directly; dashed segments are the rate carried forward. Each company card shows both the annual rate and the accumulated total, so the measured figure is always visible alongside the projection.
Think a number is wrong, or have data that would sharpen it? Write us: contact@ahcdata.fund. All analysis code is open source on GitHub.
Last updated 2026-06-10, through Issue #15. Updated within a day of each new issue.