The Money You Already Earned: What Payer Underpayments Are Quietly Costing Your Health System

You Did the Work. The Payment Didn’t Follow.

Here is the part of the revenue conversation nobody likes to have out loud: your health system already earned a meaningful share of the money it is missing. The care was delivered. The beds were staffed. The physicians documented, the nurses covered, the supplies were used. And then, somewhere between the claim and the deposit, the payment came back smaller than the contract said, later than it should have, or not at all.

I am not talking about charity care, or about services nobody expected to be paid for. I am talking about earned revenue — money owed under contracts your organization already signed, for care your teams already gave — that leaks out through denials, underpayments, and claims that were never fought in the first place.

Most leaders I talk to do not have a revenue problem they can see. They have a revenue problem they have stopped noticing, because it arrives in small pieces: a denial here, a re-level there, a short-pay that looks close enough to let go. Added up, those pieces are not small at all.

The Numbers Are Not Small

Start with how common denials have become. Kodiak Solutions’ benchmarking of 2,300 hospitals and 350,000 physicians found the average initial denial rate rose to 11.6% in 2025, up from 11.4% in 2024, while the median final denial rate — the claims still denied after appeals — rose from 2.5% to 2.7%. Clinical denials, the ones tied to prior authorization and medical necessity, accounted for nearly all of that increase, and providers got slightly worse at overturning them, with the overturn rate dipping from 42.7% to 42.1%.

Two and a half percent sounds manageable until you put dollars on it. That same Kodiak analysis put total 2025 revenue leakage at $48.4 billion across its hospitals — final denials and uncollected bills combined — up 25% in a single year. Small rate movements, applied to hospital-scale volume, produce very large numbers. That is the first thing to understand about underpayments: the percentages look boring. The totals are not.

The cost of chasing the money is its own line item. The American Hospital Association’s “Costs of Caring” report found hospitals spent $43 billion in 2025 simply trying to collect payments insurers already owed for care already delivered — denials, prior authorizations, repeated documentation requests, and changing billing and coverage rules. The AHA reports the average hospital now staffs about 64 administrative and billing employees dedicated to those functions, roughly 6.5% of total hospital employment. That is not a billing department. That is a second workforce whose entire job is getting paid for the first workforce’s work.

And this is not a hospital-only story. In a March 2024 MGMA poll of 235 medical group leaders, 60% said their claim denial rates had increased over the prior year, while only 11% said denials had gone down. When those leaders named the reasons for their most recent denials, the list was familiar: insufficient documentation, patient eligibility or incorrect ID numbers, untimely filings, incorrect modifier usage, and payers claiming a lack of prior authorization — despite the practice having it.

Where the Money Hides

In my experience, earned revenue hides in four places. If you only look at one of them, you will conclude the problem is smaller than it is.

1. Denials that should never have been denied

Some denials are legitimate. Many are not. A claim denied for “insufficient documentation” may simply need the record that already exists, presented the way the payer requires. A claim denied for lack of authorization may have an authorization on file that was never matched to the claim. MGMA’s leaders reported exactly that pattern — payers claiming no prior authorization despite the practice having secured one.

These are winnable fights, but only if someone actually fights them, with the clinical record to back them up. A complete, physician-reviewed record changes these conversations, because the appeal stops being an argument about what happened and becomes evidence of what happened. Every element traceable to its original source document and provider is a much harder thing for a payer to wave away.

2. Paid, but paid less than the contract says

This is the quietest leak of all, because it never shows up as a denial. The claim pays. It just pays short.

Contract terms drift. Fee schedules update. A payer’s system applies an edit your contract does not authorize, or re-levels a visit to a lower code than the documentation supports. Nobody flags it, because the deposit arrived and the account looks closed. The only way to catch it is to compare, line by line, what the contract promised against what actually landed — and to do it payer by payer, service line by service line, because the pattern is never uniform. One payer underpays one service line. Another short-pays a different one. Averages hide both.

3. Claims that were never appealed at all

Some revenue is not denied. It is abandoned. A denial lands in a queue, the queue is longer than the staff, the timely-filing window passes, and the claim is written off — not because it was unpayable, but because nobody got to it in time.

The consumer side of the market shows how rarely denials get challenged at all: KFF’s analysis of 2023 HealthCare.gov data found consumers appealed only about 1% of denied in-network claims. Provider appeals are a different process, and hospitals do appeal far more than that — but the principle holds on the provider side too. Every denial that ages out unworked is a decision, whether anyone made it deliberately or not. And the Kodiak data says the odds of winning are not bad: providers still overturned more than four in ten of the denials they challenged. The money left on the table is, in large part, the money nobody picked up.

4. The cost of the chase itself

Even the denials you win cost you. Staff time, record retrieval, peer-to-peer reviews, rework, delayed cash — the AHA’s $43 billion figure is not lost revenue, it is the price of collecting revenue that should have arrived the first time. A health system that wins every appeal and still spends a fortune winning them has a revenue cycle problem, not a victory. The goal is not to get better at appealing. The goal is to get paid correctly the first time, so there is less to appeal.

Why It Keeps Happening

It keeps happening because the system that produces the denial is faster than the system that fixes it. Payer rules change constantly. Documentation requirements tighten. Prior authorization expands into new service lines. And every change lands on teams that are already carrying full caseloads.

It also keeps happening because the people closest to the problem rarely see the whole pattern. The biller sees one denial. The coder sees one code. The physician sees one peer-to-peer request. Almost nobody sees that the same payer has denied the same service line forty times this quarter for the same reason — and that forty-first denial was preventable the day the pattern was spotted.

Patterns are where the money is. Single claims are where the time goes.

That is why the first step in any serious recovery review is not a new tool or a new vendor. It is a pattern review: denials by payer, by service line, by reason code, over time.

Your RCM Team Is Not Underperforming

Here is something worth saying plainly, because it rarely gets said: if your denial numbers look bad, your first instinct should not be to blame your revenue cycle team.

Their job, as it is actually structured, is not to close every claim. It is to generate as much revenue as possible with the staff and hours they have. Those are two very different missions, and they point in different directions.

A team chasing complete closure on every single claim would spend its week fighting low-dollar denials that cost more to work than they return, while high-value claims sit in the queue. A team doing its job well does the opposite: it triages. It works the claims most likely to pay, in the order most likely to move the total, and it lets the ones that cost more to chase than they are worth go. That is not underperformance. That is the job, done correctly, under a workload no team can fully clear.

So when a claim ages out unappealed, the honest question is rarely “why didn’t someone work it?” The honest question is “what were they working instead — and was that the right call?” If your team is bringing in the most revenue the staffing allows, they are succeeding at the mission they were given, even if the queue is never empty. The queue will never be empty. Judge the team by the revenue they recover, not by whether every claim closes — because no team, at any size, closes them all.

And that is the larger point: you are losing this money because the payment system is built to leak in small, deniable amounts, and because nobody is paid to notice the pattern until it is already a write-off.

The money is not theoretical. It is sitting in your denial queues, in your short-paid remits, and in the claims that aged out while everyone was busy. It was earned the day the care was delivered. The only question is whether someone goes back for it.

The conversation costs nothing. Missing it might.