Reference
The strongest support is the burden-gap shape, not a single universal health-insurance number.
KFF’s ACA marketplace analysis supports one part: in HealthCare.gov plans, consumers appealed fewer than 1% of denied in-network claims in 2024.
KFF’s Medicare Advantage prior-authorization analysis supports a different part: 11.5% of denied prior-authorization requests were appealed in 2024, and more than eight in 10 appealed denials were partially or fully overturned.
The policy sources support the "rails are not workflow" point. CMS, HIPAA, ONC, and HealthCare.gov pages describe rights, APIs, deadlines, records, and appeal steps. They do not create the default burden-reducing workflow the essay is asking for.
More on the references
The issue's Reference layer is strongest where it names public statistics and weakest where it projects a workflow standard. The KFF sources can support low appeal rates, high overturns in one procedural category, low awareness of appeal rights, and the need to keep categories separate. They cannot tell us how often denials are wrong, how many unappealed denials are contestable, or what a one-button workflow would do at scale.
The CMS, HHS, ONC, and HealthCare.gov sources support the existence of procedural rails, access rights, and appeal rules. They also constrain the draft by showing why rights and APIs are not the same as appeal execution.
What the references did
Changed
- The data setup now says the sources give pieces of the shape from different parts of the system, not one single metric.
- ONC information blocking was added to the source package because the prose mentioned that rail.
Narrowed
- Product examples are category signals only.
- CMS/HIPAA/ONC are rails and rights, not a complete workflow.
- Medicare Advantage overturns are a warning light with a documentation caveat.
Bounded
- The issue does not measure the true number of contestable unappealed denials.
- Near-term feasibility remains medium-low.
- The final upstream-pressure claim is a plausible mechanism, not a measured effect.
Source ledger
Open any entry to see the outside constraint and what it could have changed.
Marketplace consumers appealed fewer than 1% of denied in-network claims.
Public sourceKFF ACA marketplace denials and appeals analysis.
How it could have changed the claimIf appeal rates were much higher, the burden-gap frame would weaken.
What actually happenedConfirmed with limits.
Remaining limitPost-service claims in HealthCare.gov QHPs, not all insurance or prior authorization.
Medicare Advantage prior-authorization appeals were often overturned.
Public sourceKFF Medicare Advantage prior authorization analysis.
How it could have changed the claimIf overturns were low, the warning-light paragraph would need major narrowing.
What actually happenedConfirmed with limits.
Remaining limitKFF notes missing documentation may explain some reversals.
Most people with denied claims did not know whether they had appeal rights and did not formally appeal.
Public sourceKFF consumer survey on denied claims.
How it could have changed the claimIf awareness and appeal filing were high, the patient-burden mechanism would weaken.
What actually happenedConfirmed with limits.
Remaining limitThe survey cannot prove how often denied claims were wrong.
CMS prior-authorization interoperability rules are rails, not the whole workflow.
Public sourceCMS CMS-0057-F fact sheet.
How it could have changed the claimIf rules already required default appeal execution, the feasibility section would change.
What actually happenedConfirmed with limits.
Remaining limitAPI compliance is phased; relevant provisions exclude drug prior authorization; APIs are not workflow execution.
HIPAA access rights help but do not erase friction.
Public sourceHHS HIPAA Privacy Rule Summary.
How it could have changed the claimIf access were instant and strategy-ready, the outside-integrated-setting burden would be overstated.
What actually happenedConfirmed with limits.
Remaining limitAccess to records is not explanation, strategy, appeal assembly, submission, or follow-up.
Patients have a legal right to their own records — but that is only the first step of an appeal.
Federal "information-blocking" rules mean a patient can get their own electronic health records. That matters because you usually need those records to contest a denial — but obtaining them is one early step, not the appeal itself.
Public sourceONC information-blocking page.
How it could have changed the claimIf no rail existed, the policy paragraph would lose support; if it created workflow, the issue would need stronger feasibility.
What actually happenedConfirmed with limits.
Remaining limitAccess to electronic health information is not the full denial-to-appeal workflow.
Internal and external appeals still require process navigation.
Public sourceHealthCare.gov internal appeals; HealthCare.gov external review.
How it could have changed the claimIf the pages described automatic appeal execution, the burden section would need narrowing.
What actually happenedConfirmed with limits.
Remaining limitHealthCare.gov pages do not cover every plan-specific procedure.
Current tools and advocates show an emerging category, not default burden transfer.
Public sourcesFight Health Insurance; Counterforce Health; Claimable; Patient Advocate Foundation; Solace.
How it could have changed the claimIf tools were embedded at scale by default, the state-of-the-art section would need upgrading.
What actually happenedConfirmed-limited.
Remaining limitVendor/service pages are self-descriptions; no independent success or scale claim is imported.
Remaining risks
About this Reference record
A source counts here only if it could have changed, narrowed, killed, or bounded a claim. This is not a bibliography.
Remaining risks, plainly: the issue does not measure the true number of contestable unappealed denials; product pages are self-descriptions; the final pressure-upstream claim is a plausible mechanism, not measured; and near-term feasibility remains medium-low because data, authorization, records, and submission paths are split across systems.