Total Revenue Integrity for AdvancedMD Practices

What it takes for every visit you deliver to become cash you can count, and how to find where yours stops.

The Short Version

At one multi-provider behavioral health practice, we traced every completed visit from the appointment to the deposit and asked what share had reached paid in full. The answer was 17%. The other 83% sat somewhere in between: documented but unsigned, signed but unbilled, billed but denied, paid but never fully posted.

Revenue was fine. The practice was growing. Every department’s report looked normal, because every report covered one stage, and every stage looked survivable on its own. No report covered the chain.

That chain is what revenue integrity actually means. Work done becomes cash collected, every time, on a clock you can see. When the chain holds, the bank account finally matches the P&L. When it leaks, the owner funds the gap personally and calls it seasonality.

This guide walks three places the chain breaks, with real numbers from practices running AdvancedMD. Then it shows what the chain looks like when it’s watched every day and the same failure can’t happen twice. Found it. Fixed it. Stays fixed. By the end you’ll know which question to ask of your own system first, and each section gives you a way to check yours this afternoon.

What Revenue Integrity Actually Means

The industry uses “revenue integrity” to mean compliance. Audit readiness. Charge capture policy binders. Useful work, and none of it answers the question an owner actually has: did the work we did this month become money?

Here is the operating definition. Every unit of care your practice delivers moves through a chain. The visit happens. The note gets written and signed. The charge gets created. The claim goes out clean. The payer pays what the contract says. The payment posts and the balance closes. Integrity means the chain completes for every visit, and you can prove it without pulling records by hand.

Now look at how your practice is organized. Providers own documentation. The billing team owns claims. Your RCM service, if you use one, owns denials and follow-up. The front desk owns scheduling and intake. Each of them has reports, and each report covers their stage. Ask who owns the whole chain, visit to deposit, and the answer in almost every practice is nobody. Not because anyone dodged the job. The role was never created, so the measurement was never built.

Your AdvancedMD system records every event in that chain. Every appointment, every note, every charge, every claim, every payment is in there right now. The recording was never the gap. The connected view, one visit followed all the way through, is something you build on top of what the system already captured. Most practices never have. That’s why the three failures below stay invisible for years inside data that recorded every one of them.

Failure Point One: The Work You Can’t Bill Yet

The finding. 628 chargeslips sat unsigned at one practice. Roughly $94,000 of care that had been delivered and documented, and could not be billed, because a signature is the gate between work done and revenue that exists. The care was real. The revenue wasn’t, yet.

Why nobody saw it. Unsigned work doesn’t look like a problem in any single week. A provider finishes a long day, leaves four notes for the morning, and the morning brings a new schedule. Four becomes forty over a quarter. Nothing alerts, because an unsigned note isn’t a failure event. It’s an absence, and absences don’t show up on reports built to count what happened.

The total also wasn’t spread evenly. It concentrated in specific providers, which means the practice didn’t have a documentation culture problem. It had a few individual backlogs hiding inside a practice-wide average that looked tolerable. Nobody in this story did anything wrong. The providers were seeing patients, which is what they’re for. The backlog grew because no one could see it growing, and you can’t manage a number nobody surfaces.

The fix. Daily visibility, by provider, with age. Every morning a list rebuilds from the system: who has unsigned work, how much, and how old the oldest item is. The conversation changes from “please stay current on notes,” which fails everywhere it’s tried, to a specific count with a specific age next to a specific name. The prevention is structural. The list regenerates itself every day, so a backlog can’t quietly become 628 again. It gets caught at six.

Check yours today. Pull your unsigned chargeslip list, sort oldest first, and write down two numbers: the count, and the age of the oldest item. That’s your baseline, and it took five minutes.

The question. How many chargeslips are unsigned in your system this morning, and how old is the oldest one?

Failure Point Two: The Claim That Was Built to Die

The finding. At a second behavioral health group, one payer and procedure combination was denying at 81.8%. Four claims out of five with that pairing were dead on arrival, and the practice kept submitting them, because each denial came back alone and each one got worked alone.

Why nobody saw it. Denials are handled one at a time. A claim comes back, someone corrects and resubmits, the queue moves. That workflow is built for volume, and it does its job. What it can’t do is step back and ask whether the denials share a shape. Aggregated by payer and procedure, this practice’s denial noise collapsed into a short list of repeat offenders, and the worst pairing failed 81.8% of the time. One claim at a time, that pattern is invisible. It’s also permanent that way, because correcting claim four hundred does nothing to stop claim four hundred one from going out tomorrow with the same fatal pairing.

Worth saying plainly: this sits upstream of your billing team and upstream of any RCM service. Claims follow-up is the bottom half of the revenue cycle, and the people doing it are doing it well. By the time a denial lands in a work queue, the cause is already weeks old. The pattern lives in the top half, where the claim gets assembled, and finding it there was never anyone’s assigned work.

The fix. The combinations get measured, then gated. Every payer and procedure pairing carries a running denial rate built from the practice’s own claim history. Anything past the threshold gets flagged before submission and routed to the right path: a different authorization step, a coding review, or a contract conversation with the payer, depending on the cause. The prevention mechanism is a watchlist that updates as payers change behavior, because they do, and a pairing that was fine last year can quietly become this year’s 81.8%.

Check yours today. Export ninety days of denials. Group by payer plus procedure code and sort by count. The top three rows are your candidates. Divide denials by submissions for each, and you’re holding your own version of the 81.8%.

The question. Which payer and procedure combination going out this week has a denial rate you would never accept if it were printed at the top of the claim?

Failure Point Three: The 83% Problem

The finding. Back to the first practice. We followed every completed visit through every stage between the appointment and the deposit, then asked the only question the bank account cares about: what share of this work has reached paid in full? Paid in full means closed, every dollar posted, balance at zero. The answer was 17%. Eighty-three percent of delivered care was parked at a stage in the middle: unsigned, uncharged, unbilled, denied and waiting, partially paid, or paid and never fully posted.

Why nobody saw it. Because every stage owner was right. Documentation looked a little behind, which is normal. The claim queue was moving, normal. Denials were being worked, normal. AR was aging about like last quarter, normal. Every single report was locally true, and the practice still had five of every six visits stuck short of done, because “done” was never a stage anyone measured. Stage reports answer “is my part moving.” Nothing answered “did the whole thing finish.” Those are different questions, and only the second one pays your bills.

This is also the finding that explains the feeling that probably brought you to this page. Revenue up, cash tight, and no report anywhere that admits anything is wrong. The reports aren’t lying. Each one is telling a small truth, and the large truth lives between them.

The fix. One funnel, visit to paid in full, every stage in a row, with volume and age at each stage, built from the records your system already holds. The first time an owner sees it, the conversation stops being about whether there’s a problem and becomes about which stage to drain first, because the money at each stage finally has a number on it. The prevention mechanism: the funnel refreshes on schedule, and any visit stuck past the expected age of its stage surfaces by itself, with a dollar value, on somebody’s list. Stuck stops being a state a visit can silently stay in.

Check yours today. Take ten completed visits from ninety days ago and trace each one to its current state. Count how many reached paid in full. Ten visits, one afternoon, and you’ll have your own version of the 17%.

The question. Of the visits your practice completed ninety days ago, what percentage has reached paid in full, and can anyone in the building produce that number today?

Why It Stays Broken

The three failures share one anatomy. Work stalled at a stage. The stall was an absence rather than an event. And absences don’t appear on reports built to count events. A visit that happened creates a record. A signature that didn’t happen creates nothing, and nothing is very hard to see.

There’s a second layer. Everyone inside a practice inherits the reports, the categories, and the definitions of normal that came with the building. A little documentation lag is normal. A steady denial hum is normal. AR aging like last quarter is normal. None of these were decisions anyone made. They’re defaults nobody was ever positioned to question, because questioning them was nobody’s job and every stage owner’s numbers said their stage was fine.

So the leak persists for years in plain sight, recorded completely by the system and surfaced by nothing, while the owner covers the gap and wonders which report to distrust. If that’s been your experience, the problem was never your effort, and it wasn’t your team. The chain had no owner, and nothing was watching it.

What Integrity Looks Like When It Holds

The repair is the same shape every time. Make the absence visible. Decide in advance who acts on it. Let the system watch for drift so the fix outlives everyone’s attention span.

Held integrity looks like this in a running practice. Unsigned work surfaces every morning by provider and age, and the oldest item has a name next to it. Toxic payer pairings get caught before submission instead of after denial. The visit-to-paid funnel refreshes on its own, and anything aging past its stage shows up with a dollar amount attached. None of it depends on a heroic office manager remembering to check. The checking is the machine’s job, and the acting is a pre-decided human step with a named owner.

Practices run this way stop being surprising. The standard is measurable: cash conversion in single digits, counted from the day the work is done to the day the money clears. Clean claims above 95% and holding. A month’s revenue you can predict before the month starts, because nothing between the visit and the deposit is invisible anymore. Those numbers are operating results from practices running on AdvancedMD, produced by watching the same chain this guide just walked.

And the three certainties an owner should demand from anyone who touches this, including us. Found it, with the records to prove every line. Fixed it, with the dollar figure that moved. Stays fixed, with the mechanism that catches the drift, because a finding without prevention is a report, and you have enough reports.

Where to Start

You don’t need a project to find out whether your chain holds. You need one look at the right cut of your own data.

Book a 30-minute look. Bring nothing. We’ll pull the thread on one of the three questions above against your own system, live, and you’ll leave knowing whether your chain holds or where it stops. If you’d rather start smaller, the Practice Cash Scorecard takes three minutes and tells you where you stand against the standard.

PracticePath is not affiliated with, endorsed by, or sponsored by AdvancedMD. AdvancedMD® is a registered trademark of Global Payments. All references to AdvancedMD are for informational purposes and to identify the software environment our services support.