A 65% E/M Error Rate and a Memo That Says 'No Refund Required'
An internal audit of 40 established-patient visits billed by five orthopedic surgeons found that 26 (65%) were billed 99214 or 99215 while the documentation supported 99213; 7 of 12 claims with modifier 25 lacked a separately identifiable service; and one surgeon's notes were near-identical across three dates. The practice is hospital-owned, bills professional services under the hospital's tax ID, and Medicare is 44% of visits. The compliance director's draft memo says the errors reflect 'documentation weakness, not coding error,' recommends one 30-minute refresher, and concludes no refund is due because the services were medically necessary. Last year's external audit found a 22% error rate in the same specialty and no corrective action was documented.
1. Critique the draft memo. Address the medical necessity argument, the documentation-versus-coding distinction, and the 60-day overpayment rule.
Read the worked answer
'Medically necessary' is the wrong standard. The claim asserts a specific code and, for modifier 25, a specific fact - that a significant, separately identifiable evaluation and management service occurred. If the documentation does not support the level billed, the claim is inaccurate even when the visit was appropriate and the patient needed care. Medical necessity is a coverage question; code support is an accuracy question, and false claims exposure attaches to the latter.
The documentation/coding distinction is a mitigation argument, not a defense. It matters for intent, for the tone of any disclosure, and for how the corrective action is designed. It does not change the refund obligation. It also becomes weaker each year the pattern persists: a 22% error rate in the same specialty last year with no documented corrective action converts a documentation problem into evidence of reckless disregard, which is enough for knowledge under the False Claims Act.
The 60-day rule applies. An overpayment must be reported and returned within 60 days of identification. Identification occurs when the organization has credible information of a potential overpayment - these audit results qualify. The organization may take a reasonable, good-faith investigation period (generally capped at six months, with the 60-day clock following, and an aggregate not to exceed eight months) to quantify the exposure, and that investigation should be scoped, dated, and documented. Concluding 'no refund' without quantification is the single most dangerous sentence in the memo.
Also flag: the near-identical notes may indicate cloning or templating, which is a documentation-integrity issue requiring separate review, and the modifier 25 findings carry their own refund exposure that the memo does not address at all.
2. Design a corrective action plan with measurable thresholds, owners, and a timeline. What data would you pull first?
Read the worked answer
Data first (within 14 days): two years of E/M level distribution by provider against the same-specialty peer benchmark; modifier 25 frequency by provider; documentation from the 40 audited visits plus a re-audit of 30 randomly selected visits per surgeon; coder-to-provider feedback records; the prior external audit workpapers and any management response; and the EHR template and dot-phrase inventory for the five surgeons.
Corrective action, phased:
- Days 0–30: written report to the compliance committee and audit committee; suspend the internal audit 'no findings' conclusion; interim 100% pre-bill review of the five surgeons' E/M and modifier 25 claims by a credentialed coder; retire or rebuild the templates showing cloning.
- Days 30–90: role-specific education - surgeons on the documentation elements of medical decision making and on what 'separately identifiable' means, coders on query practice and on not upcoding to match habit. Education must be case-based using the group's own de-identified records, with post-testing and attendance tracking; a single 30-minute session does not remediate a 65% error rate.
- Days 90–180: prospective monitoring transitions to monthly random samples of 10 records per provider; add a physician champion and a monthly scorecard by provider.
Thresholds for exit: error rate below 5% for two consecutive quarters at the provider level and below 3% in aggregate, with no individual provider above 10% in any month. Report the trend to the compliance committee quarterly.
Owners: compliance director owns the plan, the coding manager owns the audits, the chief medical officer owns physician accountability, and the audit committee receives a written status update each quarter until exit criteria are met.
3. When would you recommend self-disclosure to the OIG rather than a routine refund, and what single fact would most change your recommendation?
Read the worked answer
Start with the default. The baseline obligation is to quantify the overpayment and refund it through the Medicare Administrative Contractor with a written explanation, then remediate. Self-disclosure is not the default for a documentation-driven error rate; it is a risk-management decision.
Factors favoring OIG self-disclosure: a large extrapolated dollar exposure; a pattern across many providers or several years; evidence of intentional conduct, such as cloning or editing notes after the fact; a prior audit that identified the same problem without correction; a credible internal report or a whistleblower who may file; or conduct that also implicates the Anti-Kickback Statute or Stark, which the Self-Disclosure Protocol is designed to handle. The protocol offers a path to a negotiated resolution and is generally viewed more favorably than being found out, but it opens a multi-year relationship and requires a robust, verifiable corrective action plan.
Factors favoring refund plus remediation: a confined group of providers, modest dollars, an honest documentation failure, prompt self-identification, and strong corrective action already underway.
The single fact that would most change the recommendation is the prior year's 22% external audit finding in the same specialty with no documented corrective action. That fact converts the current findings from an isolated documentation failure into a repeat pattern with notice, which is the classic profile for a reckless-disregard theory under the False Claims Act. If that fact is confirmed, I would refer the matter to counsel with a recommendation to consider self-disclosure, and I would not let the memo's 'no refund' conclusion stand in the interim.