You Captured the Charge. Did You Capture the Revenue?

A revenue operations leader at a physician billing company described what her hospital-based providers did before they moved to charge capture software. They carried “a hospital card in their pocket or the face sheet where they would just hand write, ‘I did a 99232 today,’ and put it back in their pocket with some kind of scribbled diagnosis.”

Most groups have moved past the card in the pocket. The charge gets entered somewhere, and it is easy to treat that as the job done. In the hospital medicine operations I’ve led, an entered charge was only the start. Getting paid correctly on it depended on three more things:

1. Whether the code matched the type of visit: an admission code on the first day of the stay, and a discharge code on the last day when the discharge work was done and documented, rather than a subsequent visit code selected by mistake,
2. Whether it was coded at the level the documentation supported, with the correct diagnoses in the correct order for that code and encounter, and
3. Whether it was ready to go out: whether it cleared payer edits, or needed to be held for a prior authorization number, a provider not yet enrolled with that payer, or another payer-specific requirement.

Each of these is far cheaper to get right before the claim goes out than to fix after it comes back denied or rejected. Working a denial on the back end costs staff time and delays cash, and some of that revenue is never recovered.

That’s why I no longer think “charge capture” is the right name for the process or the software. Capture is one piece of a larger job, which is charge integrity: making sure every charge is right before it hits the payer. That means the correct code type for the visit, a level the documentation supports, diagnoses that support the service in the correct order for that encounter, and payer requirements met, with charges moving quickly because charge lag is delayed cash.

Why the gaps widen as your group adds providers and sites

Physicians keep moving into larger groups. In the AMA’s 2024 Physician Practice Benchmark Survey, 18.3% of physicians worked in practices of 50 or more physicians, up from 12.2% in 2012. Each new provider, facility and system adds another place for a charge to go out wrong. Every new provider also has to be enrolled with each payer before claims for their work will pay, and charges sent before the enrollment date can come back denied.

Hospital medicine groups show the same trend. In SHM’s 2025 State of Hospital Medicine Report, the median adult hospital medicine group was 26.9 FTE physicians. Of 62 groups tracked from 2020 to 2025, 48 grew, and on average the groups added 9.1 FTE physicians, about 40%. Those are FTEs, not headcount, so the number of people entering charges is higher once part-time and per diem providers are counted. I don’t expect this to reverse. Patients are sicker, stays run longer when post-acute beds are hard to find, and hospitalists are taking on a wider scope, so I treat today’s size as the new baseline rather than a peak. Growth also makes it harder to see how each provider is performing, which is why the checks later in this article need to be run regularly.

 

Specialty doesn’t change the underlying process. Code sets differ, but single-specialty and multi-specialty groups ask the same questions: are we coding each encounter correctly, are the charges going out clean, and are we keeping up with Medicare rule changes that commercial payers tend to adopt later?

Right code, right level: what the “safe middle” costs you

Providers who aren’t sure which level their documentation supports often choose a moderate one because it feels safe. In hospital medicine, I believe that more often means underbilling. Patient acuity keeps rising, EHR documentation is thorough, and since 2023 providers can select the level by medical decision making or by time. A busy provider moving to the next patient rarely goes back through the note to check whether it supported a higher level, or met criteria for critical care. Code type matters as much as level. When a subsequent visit code is selected on the day of admission and the documentation supported an initial hospital care code, the difference is money left on the table.

Overbilling carries compliance risk, so the goal is accuracy in both directions. The question is what the safe middle is costing the group. A later post in this series looks at how CPT distribution and work RVU trends by provider show it.

Catch errors before the payer does

Scrubbing applies payer edits before a charge leaves your hands: code pairs, modifiers, diagnosis-to-procedure logic, age and gender edits, medical necessity and patient demographics. Change Healthcare research cited by MGMA found that 86% of denials are potentially avoidable and half start with front-end issues, with registration and eligibility alone accounting for nearly 27% of denials. Reworking a claim costs an average of $25.20. That figure covers only the rework. It doesn’t count the cash that sits unpaid while the claim is corrected, the staff hours pulled away from new charges, or the denials that are never worked and get written off. Too much of that time and money goes to problems that could have been resolved on the front end.

Accuracy includes the diagnoses: the right codes, in the right order, supporting the procedure or visit being billed. Diagnosis-to-procedure mismatches are an error I’ve seen many times. A wound debridement code goes out with acute renal failure or hypertension as the primary diagnosis, and none of the other diagnoses shows a wound at all. The procedure may have been done and documented correctly, but nothing on the claim tells the payer why it was necessary. That is exactly the kind of error a scrub should stop before the claim leaves.

The edits keep changing. ICD-10-CM has always included Excludes1 notes, which flag two diagnoses that shouldn’t be reported together, but payers had not been enforcing them. Now some are, and claims that pair those diagnoses are being denied. Groups with overlapping providers run into a different edit. A group of internists and physiatrists rounding in rehab hospitals described claims kicked back when both providers saw the same patient on the same day and listed the same primary diagnosis. They scrub those claims by hand today.

Some charges are coded correctly and still aren’t ready to go. A service that needs a prior authorization number, or a provider whose enrollment with that payer isn’t in place yet, will be denied no matter how clean the coding is. Those charges should be held until the issue is resolved, not sent and worked later.

When edits run before charges are released, clean charges go straight to billing and staff work only the exceptions. The CEO of a 125-provider infectious disease group described the difference after seeing that workflow:

“Right now, our staff has to touch every single charge that you’re doing today. With this application, if the claim is going to bypass or fly through edits and it’s a clear charge, it automatically goes over to Epic. So the only thing that our staff has to look at are those exceptions.”

How to tell whether it’s happening across your group

Knowing what to check is the easy part. Knowing whether it’s happening for every provider at every site is harder. Physician group owners, and the leaders who run the group’s operations and revenue cycle, should be able to answer these at any point:

1. Do the admission, subsequent visit and discharge codes line up with the patients who were actually admitted and discharged?
2. Which providers’ levels cluster in the middle, and when their notes are reviewed, does the documentation support a different level or different diagnoses?
3. What share of charges go out clean the first time, and how many of the denials that come back trace to an edit, authorization or enrollment issue that could have been caught before release?

These should be reviewed regularly, because as the group grows, the revenue loss grows with it.

What it adds up to, one provider at a time

When the code type, level, diagnoses and payer requirements are right before a charge goes out, the results show up in the numbers leadership already watches: fewer denials and rejections, faster cash, fewer days in accounts receivable, and less staff time spent reworking claims on the back end. HFMA’s benchmarks are a 98% clean claims rate and charges captured within three to five days of service. The next post in this series turns those into a monthly scorecard.

I’d measure the return per provider rather than as one large annual figure. Picture one or two charges per provider per month that go out right the first time: an admission billed as initial hospital care instead of a subsequent visit, a level that matches the documentation, a claim held until the provider’s enrollment date instead of sent and denied. That is a number a group’s leaders can track, and one they can check against their own coding and denial data.

Where Charge Pro fits

medaptus Charge Pro sits between the encounter and your billing system. It takes in census, charge and documentation feeds from the facilities where your providers work and runs every charge through standard coding edits (NCCI, LCD and NCD) in real time, along with custom rules for your payers and workflows, such as shared visits between physicians and advanced practice providers, or holding a charge until a provider’s enrollment with a payer is in place. Clean charges route to billing, and exceptions hold for review. Charge Pro works alongside your billing system or RCM partner rather than replacing it, and building the facility interfaces is part of implementation.

One thing you can check this week

Ask your billing team or RCM partner for the denials and rejections on claims submitted to the primary payer 30 to 60 days ago, sorted by category. Most of those claims should have been adjudicated by now. Rejections are claims the clearinghouse or payer sent back before processing, usually for missing or invalid information. Denials are claims the payer processed and refused to pay. Look at what percentage of those claims fall into each of these categories, all of which can be caught before a claim goes out:

1. Eligibility and registration: patient not covered on the date of service, wrong insurance ID or demographics
2. Prior authorization: missing or invalid authorization number
3. Provider enrollment: rendering provider not enrolled or credentialed with that payer
4. Coding: invalid code or modifier, code-pair (NCCI) edits, bundling
5. Medical necessity: diagnoses that don’t support the service, or LCD or NCD criteria not met
6. Duplicate claims: the same service billed twice, or two providers billing the same patient on the same day
7. Missing or invalid claim information: NPI, place of service, dates or other required fields

The Change Healthcare research cited earlier found 86% of denials potentially avoidable, and your own report will show where your group stands. That’s where to start.

See how Charge Pro supports physician groups across every facility where your providers work. Next in this series: the revenue cycle metrics physician group leaders should review every month.