Lessons to Learn from OIG HCC Reviews

The U.S. Department of Health and Human Services (HHS) Office of Inspector General (OIG) released a report on a Medicare Advantage Compliance Audit of SelectCare of Texas, Inc. The findings of the report—published at the end of November (2023)— were reminiscent of those of the OIG’s review of Geisinger Health Plan in the spring, and Excellus Health Plan in the summer. This audit reinforces the lessons that should be learned.

There were a set of nine conditions appearing in this audit, which were identified by data-mining and discussions with medical professionals. It seems to me that the issue was roughly the same for each category – a diagnosis was claimed and there was no evidence of treatment or continuity. It also seems as though this set of diagnoses was identical to that found in previous audits.

• Acute stroke – there must be a corresponding inpatient or outpatient hospital claim within the service year. Patients do not have chronic strokes, although they may experience sequelae from a previous stroke. The coding can distinguish between a current, acute cerebrovascular (CVA) accident and a previous CVA with residua. As I have said before, if an outpatient provider has the urge to document and code an acute stroke, they should be calling 911 for transport.
• Acute myocardial infarction (MI) – if an acute MI code isn’t found on an inpatient claim within 60 days, then the OIG questioned whether a less-serious condition wouldn’t be more appropriate, like angina pectoris, myocardial injury, or some other ischemic heart disease. Acute MI is picked up for 28 days after the initial diagnosis, but their presumption is that there would have been an index admission.
• Major depressive disorder – if the condition were captured on only a single claim during the service year, but no antidepressant medication was dispensed, the OIG felt the diagnosis might be considered unsupported. I wondered:
  • What if the patient opted for psychotherapy instead of medication?
  • What if it were discovered at the end of the service year? It might still be clinically valid if the documentation supported it.
• Embolism – suspicion arose if there was only one claim with this diagnosis in the service year and no anticoagulant medication had been ordered. Musings:
  • What if the patient has or had an inferior vena cava filter placed? This is usually because there is a contraindication to initiating anticoagulation.
  • This is a diagnosis that has codes available for acute (including subacute), chronic, and history of. Providers need to select the clinically valid code.
• Vascular claudication – these enrollees had the diagnosis found on only one claim, but had not had one of the diagnoses that indicate vascular claudication during the preceding two years – and, furthermore, they were taking a medication usually associated with neurogenic claudication.
• Lung, breast, colon, and prostate cancers – if there was an active cancer diagnosis on only one claim, but there was no surgery, radiation, or chemotherapy administered within six months prior to or subsequent to the diagnosis, the OIG concluded that a “history of” cancer code should have been used instead.
  • Immunotherapy is a commonly utilized treatment for certain cancers now. It does not seem to be on the OIG’s radar.
  • What if a patient declined treatment? The condition is still valid.

Diagnoses that are chronic should be coded as chronic. Having a single documentation of a chronic condition calls into question whether it is really present. Having a chronic condition with a standard-of-care medical treatment should result in that treatment being prescribed unless there are contraindications or extenuating circumstances (which should then be laid out in the record).

One of the cancers was denied because only “lung mass” was documented. This reinforces the position that it is advisable to document pathology diagnoses post-discharge. There was also a lung metastasis, which was submitted as a primary malignancy. The most common issue regarding cancers was that they should have transitioned to “history of,” but had instead been submitted as active, current conditions.

There are multiple prongs to the approach to solving this problem. The first is educating your providers on correct documentation and coding. They need to document conditions accurately and precisely. They need to choose the correct code, supported by the documentation. When diagnoses evolve to “history of,” the documentation should follow suit.

When a diagnosis is made, medically appropriate treatment should be initiated unless there are extenuating circumstances preventing that course of action. That should be explicitly recorded in the chart.

Your organization should consider setting up a technology solution to ensure that any of the high-risk target conditions have evidence of treatment or an explanation as to why treatment is not being undertaken. Perhaps an electronic medical record alert for a programmed set of diagnoses? A reconciliation of diagnoses and administration of medication? A second-level review by clinical documentation integrity (CDI) for high-risk diagnoses?

It should be routine practice that when a chronic diagnosis is made, it is documented repeatedly during a service year whenever it is being addressed. In theory, noting a diagnosis once is sufficient for coding; in practice, auditors question chronic diagnoses if they are only mentioned once and never again.

If your facility has not taken steps to ensure that Hierarchical Condition Category (HCC) diagnoses are properly validated, you are just asking to be the next victim in the OIG headlights.

An ounce of prevention is worth a pound of recoupment.

Programming note: Listen to Dr. Erica Remer as she co-hosts Talk Ten Tuesdays with Chuck Buck today at 10 Eastern.