Medicare Advantage Compliance Audit of Diagnosis Codes That Humana Submitted to CMS

04-19-2021 | A-07-16-01165 | Complete Report | Report in Brief

Why OIG Did This Audit

Under the Medicare Advantage (MA) program, the Centers for Medicare & Medicaid Services (CMS) makes monthly payments to MA organizations according to a system of risk adjustment that depends on the health status of each enrollee. Accordingly, MA organizations are paid more for providing benefits to enrollees with diagnoses associated with more intensive use of health care resources than to healthier enrollees who would be expected to require fewer health care resources.

To determine the health status of enrollees, CMS relies on MA organizations to collect diagnosis codes from their providers and submit these codes to CMS. CMS then maps certain diagnosis codes, on the basis of similar clinical characteristics and severity and cost implications, into Hierarchical Condition Categories (HCCs). CMS makes higher payments for enrollees who receive diagnoses that map to HCCs.

For this audit, we reviewed one of the contracts that Humana, Inc., has with CMS with respect to the diagnosis codes that Humana submitted to CMS. Our objective was to determine whether Humana submitted diagnosis codes to CMS for use in the risk adjustment program in accordance with Federal requirements.

How OIG Did This Audit

We selected a sample of 200 enrollees with at least 1 diagnosis code that mapped to an HCC for 2015. Humana provided medical records as support for 1,525 HCCs associated with the 200 enrollees. We used an independent medical review contractor to determine whether the diagnosis codes complied with Federal requirements.

What OIG Found

Humana did not submit some diagnosis codes to CMS for use in the risk adjustment program in accordance with Federal requirements. First, although most of the diagnosis codes that Humana submitted were supported in the medical records and therefore validated 1,322 of the 1,525 sampled enrollees' HCCs, the remaining 203 HCCs were not validated and resulted in overpayments. These 203 unvalidated HCCs included 20 HCCs for which we identified 22 other, replacement HCCs for more and less severe manifestations of the diseases. Second, there were an additional 15 HCCs for which the medical records supported diagnosis codes that Humana should have submitted to CMS but did not.

Thus, the risk scores for the 200 sampled enrollees should not have been based on the 1,525 HCCs. Rather, the risk scores should have been based on 1,359 HCCs (1,322 validated HCCs + 22 other HCCs + 15 additional HCCs). As a result, we estimated that Humana received at least $197.7 million in net overpayments for 2015. These errors occurred because Humana's policies and procedures to prevent, detect, and correct noncompliance with CMS's program requirements, as mandated by Federal regulations, were not always effective.

What OIG Recommends and Humana's Comments

We recommend that Humana refund to the Federal Government the $197.7 million of net overpayments and enhance its policies and procedures to prevent, detect, and correct noncompliance with Federal requirements for diagnosis codes that are used to calculate risk-adjusted payments.

Humana disagreed with our findings and with both of our recommendations. Humana provided additional medical record documentation which, Humana said, substantiated specific HCCs. Humana also questioned our audit and statistical sampling methodologies and said that our report reflected misunderstandings of legal and regulatory requirements underlying the MA program. After reviewing Humana's comments and the additional information that it provided, we revised the number of unvalidated HCCs for this final report. We followed a reasonable audit methodology, properly executed our sampling methodology, and correctly applied applicable Federal requirements underlying the MA program. We revised the amount in our first recommendation from $263.1 million (in our draft report) to $197.7 million but made no change to our second recommendation.

Complete Report available here. 

 

Humana banks on big data to boost patient health — and profits

From fewer heart attacks to stronger bones and longer functioning brains, insurer Humana hopes that a data analysis partnership with a California biotech company will help improve people’s health, lower the cost of care and increase profits.

Humana’s collaboration with Amgen will focus on the enormous amounts of health claims data that Humana collects on its roughly 9 million customers. While details about the research are still being finalized, the insurer hopes to analyze the data to detect health problems earlier and intervene before they become more serious — and costly to the patient, the hospital and the insurer.

“We believe that you lower costs by improving patient outcomes,” said Laura Happe, the company’s chief pharmacy officer.

The companies also plan to combine real-world evidence with data from wearable tech and apps or even Bluetooth-enabled drug delivery devices to target serious conditions such as cardiovascular disease, osteoporosis, neurologic disorders, inflammatory diseases and cancer.

While Humana and Amgen have worked together before, the new partnership marks the first research collaboration.

“Overall, there is a trend towards using data and learning from data to optimize care,” Happe said.
Businesses and governments hope the ability to analyze big sets of health care data will allow them to improve care and reduce waste, fraud and costs — but the sheer amount of data that is available and being collected also presents significant challenges.

In a recent report by Stanford Medicine, Dr. Lloyd Minor, dean of the Stanford School of Medicine, said the health care industry’s increasing connectivity and complexity “poses both an opportunity and a challenge.”

“By leveraging big data, we can create a vision of health care that is more preventive, predictive and precise,” he said.

California-based data analytics company MapR recently said that data analysis can help health care providers with early detection of serious health conditions, such as congestive heart failure.
CHF “accounts for the most health care spending. The earlier it is diagnosed, the better it can be treated, avoiding expensive complications, but early manifestations can be easily missed by physicians,” the company said. “A machine learning example from Georgia Tech demonstrated that machine learning algorithms could look at many more factors in patients’ charts than doctors, and by adding additional features, there was a substantial increase in the ability of the model to distinguish people who have CHF from people who don’t.”

Happe said that when CHF is undetected or uncontrolled, patients can end up with fluid retention, which can become life-threatening.

Humana and Amgen will sift through data to figure out how to detect the condition earlier and prevent bad outcomes. Healthier patients will mean fewer doctor and hospital visits, which will mean lower costs to health care providers and payers, including the patient and Humana.

Amgen, based in Thousand Oaks, Calif., conducts research and develops therapies for illnesses, primarily in six areas: cardiovascular disease, oncology, bone health, neuroscience, nephrology and inflammation. Its products include Enbrel, which treats arthritis; Neulasta, which stimulates white blood cell growth and is used often for chemotherapy patients; and Sensipar, which prevents bone disease. The company employs about 20,000 in 100 countries and last year generated revenue of about $23 billion.

Read More
https://insiderlouisville.com/business/humana-banks-on-big-data-to-boost-patient-health-and-profits/

Capitated Doc Is Indicted in First MA Upcoding Criminal Case in S. Fla.

Reprinted from MEDICARE ADVANTAGE NEWS, biweekly news and business strategies about Medicare Advantage plans, product design, marketing, enrollment, market expansions, CMS audits, and countless federal initiatives in MA and Medicaid managed care.

In the first criminal case the U.S. Attorney’s Office in South Florida has brought on alleged fraud via up-coding of Medicare Advantage diagnoses, the feds this month obtained a grand-jury indictment against a Palm Beach County physician accused of causing at least $2.11 million in excessive MA payments. At the time, Isaac Kojo Anakwah Thompson, M.D., was a capitated member of Humana Inc.’s MA provider network, but he no longer is in the network, says the company, which is not accused of wrongdoing in the indictment and contends it has repaid money as part of cooperating with the feds on the case.

Thompson pleaded not guilty in U.S. District Court in West Palm Beach, Fla., on Feb. 18, and a trial was set for March 23.

On the same date as Thompson’s plea, Humana disclosed Feb. 18 that it “recently” has received a request for information from the U.S. Department of Justice’s Civil Division about how it oversees risk-adjustment data in MA, including such aspects as medical-record reviews, use of health assessments and fraud-detection efforts. The company said in its Form 10-K filing with the SEC that it is cooperating with that request as well.

The grand jury in Florida on Feb. 3 indicted Thompson on eight counts of health care fraud that it said occurred between about January 2006 and April 2010. He allegedly did this by reporting to Humana “false and fraudulent diagnoses of Medicare beneficiaries enrolled in a Humana Medicare Advantage plan, thereby increasing the capitated payments that Medicare made to Humana and that Humana in turn made to” two entities in which Thompson was a principal.

Diagnoses Submitted Were for Serious Illnesses

The indictment charges that the claimed diagnoses the beneficiaries involved “did not suffer from” included ankylosing spondylitis (a chronic inflammatory disease of the spine), sacroiliitis (an inflammation in joints in the pelvis), inflammatory polyarropathy (five or more inflamed, swollen, tender joints) and major depressive affective disorder. Humana, which paid Thompson’s medical center about 80% of the MA capitation pay it got for beneficiaries who picked one of two Thompson entities as their primary care provider, “reported the false and fraudulent diagnoses to Medicare,” the indictment says.

The document adds that Thompson “obtained control of the fraudulent proceeds” that Humana paid to the two entities and “diverted these monies for his personal use and benefit, as well as that of others.” The charges carry maximum penalties that include 10 years of imprisonment.

Robert Nicholson, a Fort Lauderdale, Fla., attorney representing Thompson, told MAN Feb. 19 that his firm had “just entered” this case and had been told by the court not to comment on it to the media.

Asked by MAN to elaborate on the company’s role in the Thompson situation and investigation, Humana spokesperson Tom Noland said only, “We are cooperating fully with the authorities. Dr. Thompson is no longer a participating physician with Humana and was never a Humana employee. Humana has reimbursed the government to ensure that both the 20% [portion of Thompson’s billed charges kept by the insurer] and the 80% [Thompson portion] were paid back in full, thus making the government whole.”

He declined to comment on why Humana’s systems themselves wouldn’t have detected such large amounts for unusual diagnoses being billed by one of its capitated network providers or to say how much money the company reimbursed the government.

Asked whether in DOJ’s view Humana did anything wrong in the Thompson situation, a spokesperson for the U.S. attorney’s offices in south Florida told MAN, “Since this matter is ongoing, we will decline the opportunity to comment.” She also wouldn’t discuss the Humana filing.

http://aishealth.com/archive/nman022615-05

Diagnoses from 2/3/15 indictment:

..., ::::: :::: 0 u	Date	Entity	i'.:' «"J" :1:1:::    ..., <l)   .... i:o  ..::	HICN (last 5 d igits)	Fraudulent Diagnosis
1	2/8/ 2010	IM Med ical	M H	9471A	Inflammatory polyarthropathy
2	2/ 16/ 2010	IKAT	BB	2528A	Ankylosing spondylitis
3	2/ 16/ 2010	IKAT	ECi	5014A	Ankylosing spondylitis
4	2/ 16/2010	IKAT	RH	3396A	Ankylosing spondylitis
5	2/ 16/ 2010	IKAT	cs	4705A	Ankylosing spondylitis
6	4/ 5/2010	IM Med ical	RI	5697A	Major depressive affective disorder
7	4/9/2010	IM Med ical	TV	4666M	Major depressive affective disorder