Billing the Patient for a Denied Claim: When You May and When You May Not

A denial says the payer will not pay. It does not say the patient will. Those are different statements, and the gap between them is where a practice either protects a patient from a bill they do not owe or hands them one — which, when the amount was the practice's to bear, is not merely poor service but a billing-compliance failure. The remittance already carries the start of the answer, in the group code sitting beside every adjusted amount.

The group code assigns responsibility for the amount, and there are four, stable across payers. CO — contractual obligation — puts the amount on the provider; it is the group code on a contractual adjustment and it is written off, not billed to the patient. PR — patient responsibility — puts the amount on the patient and is the code that produces a statement. OA and PI put it somewhere that is usually neither a settled patient balance nor a final write-off yet. How to actually read those codes on the remittance is the subject of reading a denial; what this page is about is what a practice may then do with what it has read.

Before any of this, rule out the case that is not a denial at all. A deductible, a copay, and coinsurance are the patient's share of a covered service the plan processed correctly, and they arrive under PR. Billing them is routine and correct — from billed charge to collected dollar sets out the arithmetic that produces them. A claim can even pay nothing and be entirely a cost-share, when the whole allowed amount fell inside a deductible the patient has not met. That is not a denial, and the patient owes it.

The denial this article is about is a different animal: a covered service the payer refused to pay for — because authorization was missing, the claim was filed late, a coverage or necessity decision went against it, or the claim carried an error. The trap is treating the second like the first: seeing a zero payment, assuming it is the patient's, and sending a statement. A denied covered service is not a cost-share, and whether the patient owes any of it is the question the rest of this page answers.

The largest group of denials cannot be billed to the patient at all — because the practice, not the patient, caused them, and the contract with the payer says the patient is not liable for that. A participating provider agrees to accept the plan's payment as payment in full for covered services and to hold the patient harmless for the difference. That is not a courtesy; it is written into the agreements, and in several programs into federal regulation:

Medicaid — payment in full, by regulation

A state may enroll only providers who accept the agency's payment, plus any authorized cost-sharing, as payment in full (42 CFR §447.15). A Medicaid provider therefore cannot balance-bill the beneficiary the difference on a covered service.

Managed care — the contract must prohibit it

Medicare Advantage and Medicaid managed care go further and require the provider contract itself to bar the charge. An MA plan's contracts must prohibit providers from holding an enrollee liable for fees that are the plan's legal obligation (42 CFR §422.504(g)), and a Medicaid managed-care enrollee may not be held liable for covered services the plan or the State did not pay (42 CFR §438.106).

Original Medicare — the participation agreement

A participating provider accepts the Medicare-approved amount as payment in full and may charge the beneficiary only the applicable deductible and coinsurance — the structure set out in assignment and participation.

Commercial plans — the same clause, in private wording

A commercial participation agreement carries a hold-harmless clause as a matter of course, though the exact wording is the individual contract's. The clause is the reason a denial the practice controls stays with the practice rather than moving to the patient.

Liability moves to the patient in a small number of defined ways, and each has a condition that must have been satisfied before the service — not after the denial arrives.

Genuine cost-share

Deductible, copay, and coinsurance, returned under PR. The patient owes these because the plan covered the service and assigned them their share — not because anything was denied.

A service the patient was warned about, in advance and in writing

This is the only mechanism that shifts a would-be denial to the patient by their own choice. On Original Medicare it is the Advance Beneficiary Notice (Form CMS-R-131): where a service is likely to be denied as not reasonable and necessary, a properly delivered notice lets the provider bill the beneficiary if Medicare denies, because the beneficiary knew of the risk and accepted it — the knowledge standard of §1879 of the Social Security Act. A commercial plan's equivalent is a specific financial-responsibility or waiver form the plan's contract recognizes. Advance is the operative word: a form signed after the denial, or a blanket “I agree to pay whatever insurance doesn't” buried in intake paperwork, generally does not do this work.

A service the plan never covers for anyone

A statutory Medicare exclusion, or a service a commercial plan excludes as a benefit, is the patient's regardless of notice, because there was no coverage to lose. That case has its own page — non-covered service denials — and it is a different situation from a covered service the plan refused.

No contract governs the price at all

A true out-of-network encounter with no participation agreement is one place the hold-harmless clause does not apply. Even there, separate federal and state surprise-billing protections restrict what a patient can be charged in emergency and certain out-of-network situations. Those are a distinct regime, not a general licence to balance-bill, and they are not detailed here.

Billing the patient and writing the balance off are both terminal. Each forecloses recovery from the payer, and each is a mistake if the denial was wrong in the first place. So the disposition comes last, after two checks.

1. 01

   Confirm the denial is actually correct

   A denial can be wrong — the authorization existed, the claim was timely, the service is a benefit. A wrong denial is appealed or corrected; it is not absorbed by the practice and it is certainly not passed to the patient. Deciding it is the practice's to eat, or the patient's to pay, before confirming it was owed by anyone is how recoverable money is lost.
2. 02

   Read the group code and the contract together

   CO with a provider-caused reason is the practice's. PR that is genuine cost-share is the patient's. A provider-caused reason returned under PR is a conflict the contract resolves — and it resolves toward the practice. A mis-split is corrected, not billed.
3. 03

   Match the disposition to the cause

   Provider-caused and correctly denied → write off, and trace the cause upstream so the next hundred claims do not repeat it. Genuinely the patient's, with any required advance notice in hand → a clear, specific patient statement. Wrong → appeal or correct.

Writing off is a legitimate outcome, not a failure — but it should be a decision, and it belongs to the practice when the cause was the practice's. Moving that same amount to the patient does not make it recoverable; it makes it a bill the patient does not owe. What a claim denial is frames the three places a denial ends — corrected, appealed, or written off; this page is about which of those a given denied balance is actually allowed to reach, and why the write-off, not the patient, is where most of them belong.