Medicare Targeted Probe and Educate in Behavioral Health: The 20-to-40 Claim Round and the File That Ends It
Table of Contents
A Targeted Probe and Educate letter rarely arrives as a crisis. It arrives as an Additional Documentation Request for a handful of psychotherapy or psychiatric evaluation claims, often addressed to whoever was listed as the correspondence contact when the program first enrolled in Medicare. By the time the billing manager connects it to a probe, a chunk of the response window is gone. This guide is for owners, clinical directors and compliance officers who want to know what a Targeted Probe and Educate (TPE) review actually tests, where behavioral health programs lose claims, and what file ends the review after the first round instead of the third.
This is operational guidance, not legal advice. Your Medicare Administrative Contractor (MAC) sets the details of its own reviews, so confirm specifics against its published TPE topics and the primary CMS sources linked below.
How Medicare TPE Works: The Short Version
Under Medicare’s Targeted Probe and Educate program, a Medicare Administrative Contractor reviews a sample of 20 to 40 of a provider’s claims per round, for up to three rounds, and offers a one-on-one education session after each round. A behavioral health program that receives a TPE Additional Documentation Request has 45 days to submit the records, and claims without a timely response are denied. A program that still shows a high error rate after the third round is referred to CMS for further action, which can include extrapolation of overpayments or referral to a Recovery Auditor.
CMS describes the program on its Targeted Probe and Educate page, and the contractor rules that govern medical review sit in Chapter 3 of the Medicare Program Integrity Manual. MACs select providers using data analysis, typically because billing patterns stand out from peers or because a provider has a history of errors on a particular service. Reviews can be prepayment (claims are held until records are reviewed) or postpayment. Providers who show improvement can be released after any round, which is the whole point of preparing well for round one.
What the First Round Actually Tests
Each TPE round is narrow. The MAC picks a topic, usually a code family or service type, and pulls claims inside it. In behavioral health, the common targets are time-based psychotherapy codes, psychiatric diagnostic evaluations, evaluation and management visits billed by psychiatric prescribers, and partial hospitalization or intensive outpatient services. MACs publish their active TPE topics on their own websites, and that list is the closest thing you will get to an advance warning.
The reviewer is not judging whether your clinicians are good. They are checking whether each claim, on paper, supports exactly what was billed: the right provider, a valid order or plan where one is required, documented medical necessity, the time that the code requires, and a legible, authenticated signature. Every claim is pass or fail on its own record. A strong program can still post a poor error rate if its records are built for clinical use and not for an outside reader with no context.
Where Behavioral Health Claims Fail in a TPE Round
In the TPE responses we have helped programs assemble, the denials cluster in a small number of places. None of them are about the quality of care. Almost all of them are about what the record can prove.
- Missing or unsupported time: psychotherapy codes are time-based, and a 60-minute psychotherapy code (90837) requires at least 53 minutes of documented face-to-face time, so a note without start and stop times or a total-time statement invites a downcode or denial.
- Unauthenticated signatures: an unsigned note, a stamped signature, or an electronic signature the EHR cannot show an audit trail for is a common failure that has nothing to do with the clinical content.
- Treatment plan gaps: a session billed on a date when the active treatment plan had lapsed, was unsigned, or did not include the service type leaves the note without the plan it depends on.
- Credential mismatch: the rendering clinician on the claim does not match the person who signed the note, or the clinician was not yet enrolled or not eligible to bill that service independently on that date.
- Cloned documentation: identical narrative across sessions reads to a reviewer as a template, not a record of a specific encounter, and undermines medical necessity even when the care was real.
- Incomplete submissions: the program sends the progress note but not the plan, the order, the assessment or the signature log that the note relies on, and the reviewer is not required to ask twice.
The last item is the one we see most often and the one that is fully within your control. Reviewers assess what is in the packet. If the treatment plan that justifies the session sits in a different module of the EHR and nobody printed it, the claim fails as if the plan never existed.
The Address Problem Nobody Checks
An inconvenient operational truth: many TPE responses are late not because the program was slow, but because the ADR went somewhere nobody was watching. Medicare correspondence goes to the addresses on the enrollment record. Programs that moved, changed billing companies, or onboarded a new administrator often have an outdated correspondence contact sitting in PECOS. The letter arrives at the old suite, or in the inbox of a staff member who left last year, and the 45-day clock runs anyway.
The fix takes ten minutes. Pull your enrollment record, confirm every address and contact, and route anything from your MAC to a shared compliance inbox with a named owner and a backup. If you want a broader look at where Medicare enrollment details drift, our team covers this as part of our behavioral health compliance services.
Building the Response File That Ends the Review Early
Treat every TPE ADR as a 20 to 40 chart audit you must finish, check and submit inside the window. A workable sequence:
- Day 1: log the letter, the claim list and the due date. Count 45 days from the date on the letter, set an internal deadline at least ten business days earlier, and assign one owner.
- Days 2 to 10: pull a complete packet for each claim: the note, the active treatment plan, the assessment or order it relies on, signature attestations or an EHR signature log, and the clinician’s credential on that date of service.
- Days 10 to 20: have someone who did not write the notes audit every packet against the code billed. Flag time gaps, signature issues and plan lapses before the reviewer does.
- Days 20 to 30: add a short cover letter per claim that points the reviewer to where each requirement is met. Do not alter or backdate records; late entries must be clearly labeled as such under your documentation policy.
- Before the internal deadline: submit through your MAC’s portal or esMD where available, and save the submission confirmation for every claim.
Then take the education session seriously. It is the one point in the process where the reviewer will tell you, plainly, what failed and why. Write it down, turn it into a documentation change, and train on it before the next round begins.
What to Do Today
You do not need to wait for a letter. Three actions this week will change how a first round goes:
- Open your MAC’s website and read its current TPE and medical review topics. If a behavioral health code family you bill is on the list, you are in the target population.
- Confirm the correspondence address and contact on your Medicare enrollment record, and route MAC mail to a monitored shared inbox.
- Run a mock round: pick 20 recent claims from your highest-volume psychotherapy code and audit them exactly as a reviewer would, with a complete packet per claim. Your error rate on that sample is your best estimate of how round one will go.
If you want help running that mock audit or standing up the response workflow, call us at (888) 458-6619. Programs without in-house compliance capacity often use a fractional compliance officer to own MAC correspondence and audit cycles.
Why TPE Belongs in Your Compliance Program
TPE results are evidence. A program that fails three rounds has a documented history the contractor and CMS can act on, and a program that passes round one has a documented history too. The HHS Office of Inspector General’s General Compliance Program Guidance describes auditing, monitoring and responding to detected problems as core elements of an effective compliance program. A TPE round is exactly that kind of signal. Log it, track your error rate by round, and report the outcome to your compliance committee or board alongside your internal audit results.
The programs that exit early are not the ones with the most elaborate notes. They are the ones whose records can stand alone in front of a stranger, and whose mail reaches the right desk. If you want a second set of eyes on either, reach our behavioral health consultants at (888) 458-6619.




