Payer Prior Authorization Approval Rates: 2026 Data
The CY2025 prior authorization approval, appeal and turnaround data that UnitedHealthcare, Humana, Aetna, Cigna, Anthem, Centene and 3 states posted.
On this page: Direct answer
Direct answer
Prior authorization approval rates by insurer: what operators need to know
The CY2025 prior authorization approval, appeal and turnaround data that UnitedHealthcare, Humana, Aetna, Cigna, Anthem, Centene and 3 states posted. We found CY2025 postings from UnitedHealthcare, Humana, Aetna, Cigna, Anthem (California Medi-Cal only), Centene's Health Net, and the New Jersey, Utah, and Indiana Medicaid agencies. We could not locate Molina's.
Under the CMS Interoperability and Prior Authorization Final Rule (CMS-0057-F), Medicare Advantage organizations, state Medicaid and CHIP fee-for-service programs, Medicaid and CHIP managed care plans, and QHP issuers on the federally facilitated exchanges had to post calendar-year 2025 prior authorization metrics on their public websites by March 31, 2026. We read what large payers and three state Medicaid agencies actually posted and compiled the figures into one table, with a CSV you can download.
The data is useful, but it is not a behavioral-health dataset and it is not a league table. The rule asks for figures aggregated across all items and services, each payer chose its own format, and the entity a payer reports on (a Medicare contract, a Medicaid plan, a state exchange issuer) rarely matches the exact product your patients carry. Read the methodology and caveats before you quote a number.
Key takeaways
The short version
- We found CY2025 postings from UnitedHealthcare, Humana, Aetna, Cigna, Anthem (California Medi-Cal only), Centene's Health Net, and the New Jersey, Utah, and Indiana Medicaid agencies. We could not locate Molina's.
- For the largest entity each payer reported, standard approval rates ranged from 73% (Cigna's exchange plans) to 99.27% (New Jersey Medicaid fee-for-service).
- Where Medicare Advantage entities reported appeal outcomes, 56.9% to 89.70% of appealed standard denials were approved. The Medicaid entities we compiled ranged from 7.60% to 46.1%.
- Payers counted appeals, time, and partial approvals differently, so treat any cross-payer ranking as a question to ask, not a finding.
- Indiana was the only posting that broke results out by service type. Its IOP psychiatric (56.6%) and detox (56.7%) lines were fully approved far less often than all services combined (87.84%).
- From 2026 on, Medicare Advantage, Medicaid, and CHIP standard decisions are due within 7 calendar days and expedited decisions within 72 hours. The CY2026 metrics due March 31, 2027 will be the first measured against those deadlines.
Take the template with you
Free to copy · no email required
Every figure in this article with entity, line of business, appeal denominator, time units, source URL, and retrieval date, plus Indiana's behavioral-health service lines, our two roll-ups, and the payers we could not find.
1. What CMS-0057-F requires payers to post
Each impacted payer must post, on its public website, a list of the items and services that require prior authorization plus nine metrics for the prior calendar year: the percentage of standard requests approved, denied, and approved after appeal; the percentage of requests approved after the review timeframe was extended; the percentage of expedited requests approved and denied; and the average and median time from submission to decision for standard and expedited requests. Every metric is aggregated for all items and services, and drugs are excluded.
CMS's optional reporting template recommends publishing counts alongside percentages. It defines approved-after-appeal as a share of appealed requests and asks payers to report medians under one day in hours rather than rounding them to 0 days. It also says metrics buried in password-protected portals do not meet the publicly accessible standard.
| Payer type | Reporting level | Metrics rule | Decision deadline beginning 2026 |
|---|---|---|---|
| Medicare Advantage organizations | Each MA contract | 42 CFR 422.122(c) | 7 calendar days standard (42 CFR 422.568); 72 hours expedited (42 CFR 422.572) |
| State Medicaid fee-for-service | State | 42 CFR 440.230(e)(3) | 7 calendar days standard; 72 hours expedited (42 CFR 440.230(e)(1)) |
| Medicaid managed care plans | Each plan | 42 CFR 438.210(f) | 7 calendar days standard; 72 hours expedited, for rating periods starting on or after January 1, 2026 (42 CFR 438.210(d)) |
| State CHIP fee-for-service | State | 42 CFR 457.732(c) | 7 calendar days standard; 72 hours expedited, per CMS's template |
| CHIP managed care entities | Each plan | CHIP managed care rules in 42 CFR part 457 | 7 calendar days standard; 72 hours expedited, per CMS's template |
| QHP issuers on federally facilitated exchanges | Issuer | 45 CFR 156.223(c) | Not changed by CMS-0057-F; CMS's template lists 72 hours urgent and 15 days standard |
2. What payers posted for calendar year 2025
Each row shows one reporting entity exactly as the payer published it. Where a payer reported many contracts, plans, or states and also published request counts, we show the entity with the largest published volume. Aetna and Cigna did not publish counts, so we show a named contract or the single entity reported. Timing appears as mean / median, in the units the payer used. All figures were retrieved on September 21, 2026.
| Payer and entity shown | Line of business | Standard approved | Standard denied | Approved after appeal | Expedited approved / denied | Standard time (mean / median) | Expedited time (mean / median) |
|---|---|---|---|---|---|---|---|
| UnitedHealthcare: contract H2001 (largest of 63; 1,854,050 standard requests) | Medicare Advantage | 76.6% initial | 23.4% initial not approved | 56.9% of appeals | 88.5% / 11.5% | 1 / 1.8 days* | 0 / 0.6 days* |
| UnitedHealthcare: New Jersey FamilyCare (largest of 45 plan entries; 279,419) | Medicaid managed care | 84.5% initial | 15.5% initial not approved | 46.1% of appeals | 91.4% / 8.6% | 0 / 2.2 days* | 0 / 0.6 days (as published)* |
| UnitedHealthcare: Texas individual and family plans (largest state; 66,385) | ACA Marketplace (FFE) | 76.4% initial | 23.6% initial not approved | 38.6% of appeals | 87.7% / 12.3% | 0 / 0.9 days* | 0 / 0.5 days* |
| Humana: contract H5216 (largest of 32; 4,238,164) | Medicare Advantage | 93.13% | 6.87% | 65.92% (5,982 of 9,075 appeals) | 89.80% / 10.20% | 1 day / 0 days | 5 hours / 0 hours |
| Humana: Florida Medicaid (largest of 9 states; 746,974) | Medicaid managed care | 97.44% | 2.56% | 7.60% (190 of 2,499 appeals) | 59.14% / 40.86% | 1 day / 1 day | 13 hours / 3 hours |
| Aetna (CVS Health): contract H5521 (one of 42; counts not published) | Medicare Advantage | 92.14% | 7.86% | 89.70% (denominator not stated) | 89.13% / 10.87% | 1.44 days / 0 days | 0.32 days / 0 days (expedited) |
| Cigna Healthcare: IFP plans on federally facilitated exchanges | ACA Marketplace (FFE) | 73% | 27% | 16% overturned | 78% / 22% | 3.95 days / 0 (no unit) | 1.53 days / 0 (no unit shown) |
| Anthem Blue Cross (Elevance Health): California Medi-Cal, all plans | Medicaid managed care | 97.4% | 2.6% | 34.1% (denominator not stated) | 91.9% / 8.1% | 3 / 1 (days implied) | 2 / 1 (days implied, expedited) |
| Health Net (Centene): Medi-Cal, reported as CA-SHP (3,096,168) | Medicaid managed care | 97.55% | 2.45% | 39.72% (404 of 1,017 appeals) | 95.68% / 4.32% | 2 days / 0 days | 1 day / 0 days |
| Wellcare By Health Net (Centene): contract H0562 (184,541) | Medicare Advantage | 98.06% | 1.94% | 77.17% (71 of 92 appeals) | 97.13% / 2.87% | 3 days / 1 day | 1 day / 1 day |
| New Jersey DMAHS: NJ FamilyCare fee-for-service (889,627) | Medicaid and CHIP FFS | 99.27% | 0.27% | 0 appeals reported | No expedited requests reported | 1 day / 1 day | NA (as published) |
| Utah Medicaid: fee-for-service (30,072 standard) | Medicaid FFS | About 97% | About 3% | Not reported | About 98% / about 2% (94 requests) | 4.81 / 4 days | 4.48 / 4 days |
| Indiana FSSA via Acentra Health: all service types (204,568 standard service lines) | Medicaid and CHIP FFS | 87.84% (plus 7.05% partial) | 5.10% | 27.78% (162 of 583 fair hearings) | 84.82% / 3.03% (plus 12.14% partial) | 2.11 / 2 days | 19.19 / 9 hours |
3. The one behavioral-health breakout: Indiana fee-for-service
Indiana's fee-for-service contractor, Acentra Health, published its 2025 results by service type, counted by service line, with partial approvals reported separately. It is the only posting in our set that shows behavioral-health categories. The figures below are standard requests, rounded to one decimal from the workbook's fractions.
Fully approved rates for IOP and detox lines were far below the all-service rate. Much of the gap is partial approvals (typically fewer units or days than requested) rather than outright denials. For mental health services, 131 of 444 standard cases that reached an administrative law judge fair hearing were overturned (29.5%). Indiana's managed care plans report on their own websites, and fee-for-service rules do not transfer to them.
| Indiana FFS service type | Standard service lines | Fully approved | Partially approved | Denied | Mean / median days |
|---|---|---|---|---|---|
| Mental health services | 54,117 | 84.0% | 10.8% | 5.2% | 2.33 / 3 |
| Inpatient psychiatric (plus 1,757 expedited) | 6,412 | 83.1% | 13.7% | 3.2% | 1.18 / 1 |
| IOP, chemical dependency | 1,219 | 67.2% | 18.5% | 14.3% | 2.59 / 3 |
| IOP, psychiatric | 572 | 56.6% | 36.2% | 7.2% | 2.81 / 3 |
| Detox admission (plus 65 expedited) | 517 | 56.7% | 26.1% | 17.2% | 1.37 / 1 |
| MRO (Medicaid Rehabilitation Option) | 494 | 73.5% | 18.4% | 8.1% | 2.07 / 3 |
| PRTF (psychiatric residential treatment facility) | 414 | 98.1% | 1.7% | 0.2% | 1.99 / 2 |
| All service types | 204,568 | 87.84% | 7.05% | 5.10% | 2.11 / 2 |
4. Patterns across the postings
- Contract-level spread is wide. Standard approval ran from 80.14% (H9431) to 100.00% (H0562) across Aetna's 42 Medicare Advantage contracts. It ran from 74.8% (H1537) to 99.3% (H7833) across the 52 UnitedHealthcare contracts with at least 1,000 standard requests, and from 87.37% (H0783) to 96.64% (H2029) across Humana's 32 contracts. Your patients' contract matters more than the brand.
- Headline figures can differ from the detail. Summing the counts in UnitedHealthcare's 63 Medicare Advantage contract entries gives 7,942,400 standard requests with 86.5% initially approved (our calculation). UnitedHealthcare's landing page reports 95.4%, a figure that counts approvals after appeal and, per its footnote, excludes requests handled by delegates, including behavioral health delegates.
- Humana's 32 Medicare Advantage contracts sum to 9,530,027 standard requests with 93.1% approved. Of 17,690 appealed standard denials, 64.7% were approved after appeal (our calculation from published counts).
- Expedited does not always mean easier. Humana's Florida, Oklahoma, and South Carolina Medicaid plans denied 40.86%, 55.36%, and 59.44% of expedited requests, compared with 2.56%, 8.53%, and 10.86% of standard requests.
- Few denials are appealed. Humana's H5216 posting records 9,075 standard appeals against 291,089 standard denials (about 3%). Health Net's Medi-Cal posting records 1,017 against 75,814 (about 1.3%). These counts come from the same posting, not a tracked cohort, but the gap is large.
- Appeals fare better in Medicare Advantage than in Medicaid. Approved-after-appeal ran 56.9% to 89.70% for the Medicare Advantage entities above and 7.60% to 46.1% for the Medicaid managed care entities.
- Medians are fast. Most entities reported same-day or next-day medians, which puts the operational bottleneck on submission completeness and follow-up rather than on waiting for a decision. Utah fee-for-service was the exception at a 4-day median.
5. Methodology
We verified the reporting requirement against the current eCFR text of 42 CFR 422.122, 438.210, 440.230, and 457.732 and 45 CFR 156.223, together with CMS's reporting template. We then looked for each payer's CY2025 posting through its public website navigation and web search, and read the payer's own page, PDF, or workbook. Numbers in the tables were read from those documents on September 21, 2026. Anything labeled as our calculation was derived only from counts the payer published.
We did not adjust, weight, or normalize any payer figure. Percentages appear at the precision the payer used, except the Indiana service-type rows, which we rounded to one decimal. Third-party aggregators and vendor blogs were not used as data sources.
| Payer | What we found | Status |
|---|---|---|
| UnitedHealthcare | Medicare Advantage (63 contract entries), Medicaid and CHIP (45 plan entries), and ACA Marketplace (by state) PDFs with counts | Compiled |
| Humana | 32 Medicare Advantage contract PDFs and 9 state Medicaid PDFs with counts | Compiled |
| Aetna (CVS Health) | One Medicare Advantage PDF covering 42 contracts, percentages only; Aetna Better Health state Medicaid PDFs also exist | Medicare Advantage compiled; Medicaid not compiled |
| Cigna Healthcare | One-line disclosure for exchange (IFP FFE) plans | Compiled |
| Elevance Health (Anthem) | California Medi-Cal scorecard; no Medicare Advantage or Marketplace posting located | Medi-Cal compiled; others not found |
| Centene | Health Net Medi-Cal and Wellcare By Health Net H0562 summaries; CalViva and Imperial Valley summaries on the same page | Two entities compiled |
| Molina Healthcare | No CY2025 metrics posting located on the provider pages we checked or through web search | Not found |
| Oscar Health | Statistics PDFs for 12 exchange issuers | Located; not compiled |
| State Medicaid agencies | New Jersey, Utah, and Indiana fee-for-service reports | Compiled |
6. Caveats before you cite a number
- Aggregate, not behavioral health. The rule requires figures across all items and services. Except for Indiana's service types, nothing here isolates IOP, PHP, residential, or SUD care.
- Entities are not products. A Medicare contract, a Medicaid plan, and an exchange issuer can each span many benefit designs, and a payer's largest entity may not be the one your patients carry.
- Lines of business are not comparable. Medicare Advantage, Medicaid, and exchange plans cover different populations, benefits, and requirement lists.
- Denominators differ. UnitedHealthcare, Humana, and Health Net report approved-after-appeal as a share of appeals. Indiana counts only administrative law judge fair hearings. Aetna and Anthem do not state a denominator, and Cigna reports percent overturned.
- Approval is defined differently. UnitedHealthcare reports initial decisions, Indiana splits out partial approvals and counts service lines, and New Jersey's approvals and denials leave 4,107 of 889,627 standard requests unaccounted for.
- Units and rounding differ. Several postings show 0 days where CMS's template asks for hours, Cigna shows medians without a unit, and Anthem's units are implied by a footnote.
- Utah excluded 13,070 delayed requests (mean 39.44 days, median 21 days) from its standard and expedited figures and reported them separately.
- Extensions are mostly unreported. UnitedHealthcare says it does not extend reviews, and several payers show N/A or 0, so the extension metric cannot be compared.
- Payers can revise their postings. Keep the retrieval date with any figure you cite.
7. How behavioral-health operators can use the data
- 01
Build a payer scorecard
For each payer and contract you bill, record the published approval, appeal, and timing figures next to your own approval rate, partial-approval rate, and submission-to-decision time. Take large gaps to the provider representative or joint operating committee.
- 02
Cite the payer's own appeal numbers
In escalations and appeal cover letters, cite the entity's published approved-after-appeal rate with the URL and retrieval date. Present it as context for why the denial deserves review, not as proof the denial was wrong.
- 03
Frame parity questions with data
Published aggregates cannot show a mental health parity disparity on their own. If your IOP, PHP, or residential requests are approved at rates well below a plan's published aggregate, use that gap to request the plan's comparative analysis for prior authorization as a nonquantitative treatment limitation.
- 04
Staff to the real bottleneck
When a payer's median decision time is same-day or next-day, delays usually come from incomplete submissions, missing clinicals, and slow follow-up. Staff utilization review for complete first submissions and for concurrent reviews, not for waiting.
- 05
Hold 2026 decisions to the new deadlines
Record the submission and decision timestamps for every Medicare Advantage, Medicaid, and CHIP request. Escalate standard requests that pass 7 calendar days, or expedited requests that pass 72 hours, without a documented extension notice.
- 06
Re-pull every April
CY2026 metrics are due by March 31, 2027. Refresh the CSV so you can see whether the new timeframes changed each payer's speed and denial mix.
Common questions
Answers before you build.
Which payers have to publish prior authorization metrics?+
Medicare Advantage organizations (by contract), state Medicaid and CHIP fee-for-service programs (by state), Medicaid managed care plans and CHIP managed care entities (by plan), and QHP issuers on the federally facilitated exchanges (by issuer), under CMS-0057-F. Employer plans and state-exchange-only plans are not covered by this federal requirement.
When were the first metrics due, and what period do they cover?+
The first postings were due by March 31, 2026 and cover calendar year 2025. Payers must post the prior year's metrics each year by March 31, so CY2026 data is due by March 31, 2027.
Are these approval rates specific to behavioral health?+
No. The rule requires figures aggregated across all medical items and services, excluding drugs. Among the postings we compiled, only Indiana's fee-for-service workbook broke results out by service type, including mental health services, IOP, detox, inpatient psychiatric, MRO, and PRTF.
Can I compare one payer's approval rate with another's?+
Only with caution. Payers report different entities, lines of business, and appeal denominators, and some count partial approvals or service lines differently. Compare the same payer and entity over time, and treat cross-payer gaps as questions rather than conclusions.
What decision deadlines apply in 2026?+
For Medicare Advantage, Medicaid (fee-for-service and managed care), and CHIP, standard prior authorization decisions are due within 7 calendar days and expedited decisions within 72 hours, with limited extensions allowed. CMS-0057-F did not change decision timeframes for QHP issuers on the exchanges.
Can I cite these numbers in an appeal?+
Yes, as context. Quote the payer's own published figure, name the entity and line of business, and include the source URL and retrieval date. Your appeal still has to rest on the clinical record and the plan's criteria.
Practical closeout
Use this operator checklist.
- We found CY2025 postings from UnitedHealthcare, Humana, Aetna, Cigna, Anthem (California Medi-Cal only), Centene's Health Net, and the New Jersey, Utah, and Indiana Medicaid agencies. We could not locate Molina's.
- For the largest entity each payer reported, standard approval rates ranged from 73% (Cigna's exchange plans) to 99.27% (New Jersey Medicaid fee-for-service).
- Where Medicare Advantage entities reported appeal outcomes, 56.9% to 89.70% of appealed standard denials were approved. The Medicaid entities we compiled ranged from 7.60% to 46.1%.
- Payers counted appeals, time, and partial approvals differently, so treat any cross-payer ranking as a question to ask, not a finding.
- Indiana was the only posting that broke results out by service type. Its IOP psychiatric (56.6%) and detox (56.7%) lines were fully approved far less often than all services combined (87.84%).
- From 2026 on, Medicare Advantage, Medicaid, and CHIP standard decisions are due within 7 calendar days and expedited decisions within 72 hours. The CY2026 metrics due March 31, 2027 will be the first measured against those deadlines.
Continue through the cluster
Verified customer case studies are added only with customer permission and supporting evidence; none is implied by these operational examples.
Sources & methodology
Trace the operational claims.
Marsa Health Editorial reviewed the primary and research sources below on September 21, 2026. We translate them into workflow controls, distinguish proposals from final rules, and flag where plan, program, state, contract, or clinical requirements vary.
- 01CMS Interoperability and Prior Authorization Final Rule (CMS-0057-F) Centers for Medicare & Medicaid ServicesOfficial CMS hub for the final rule, its compliance dates, and the prior authorization metrics reporting resources.Accessed or rechecked September 21, 2026
- 02Prior Authorization Metrics Reporting: Overview & Template Centers for Medicare & Medicaid ServicesOptional CMS template defining expected numerators and denominators, reporting levels by payer type, and the instruction to report sub-one-day medians in hours.Accessed or rechecked September 21, 2026
- 0342 CFR 422.122: Prior authorization requirements (Medicare Advantage) Electronic Code of Federal RegulationsParagraph (c) requires MA organizations to post the prior-year metrics at the contract level by March 31, beginning in 2026.Accessed or rechecked September 21, 2026
- 0442 CFR 422.568: Standard organization determination timeframes Electronic Code of Federal RegulationsSets the 7-calendar-day standard decision deadline for items and services subject to prior authorization beginning January 1, 2026; 42 CFR 422.572 keeps the 72-hour expedited deadline.Accessed or rechecked September 21, 2026
- 0542 CFR 438.210: Coverage and authorization of services (Medicaid managed care) Electronic Code of Federal RegulationsParagraph (f) requires plan-level public metrics; paragraph (d) sets 7-calendar-day standard and 72-hour expedited limits for rating periods starting on or after January 1, 2026.Accessed or rechecked September 21, 2026
- 0642 CFR 440.230: Sufficiency of amount, duration, and scope (Medicaid fee-for-service) Electronic Code of Federal RegulationsParagraph (e) sets 2026 decision timeframes for state Medicaid agencies and requires state-level public metrics by March 31.Accessed or rechecked September 21, 2026
- 0742 CFR 457.732: Prior authorization requirements (CHIP) Electronic Code of Federal RegulationsParagraph (c) requires state CHIP agencies to post the same prior authorization metrics at the state level.Accessed or rechecked September 21, 2026
- 0845 CFR 156.223: Prior authorization requirements (QHP issuers on FFEs) Electronic Code of Federal RegulationsParagraph (c) requires issuer-level public metrics from QHP issuers on the federally facilitated exchanges.Accessed or rechecked September 21, 2026
Organizational author. Editorial review covers source accuracy, search intent, workflow boundaries, and human-oversight requirements. This material is educational and does not provide clinical, legal, coding, or coverage advice.
No named clinical or legal expert reviewer is attributed to this version. Marsa Health does not invent reviewer credentials.
Read our editorial methodRevision history
What changed and when
September 21, 2026
Initial publication, source review, and operational editing.