UHC’s HouseCalls Report Shows Medical Record Support but Ignores Central Question

by Lauren Flynn Kelly

A recent UnitedHealth Group report attempted to demonstrate the appropriateness of diagnoses collected during home visits and submitted for risk adjustment purposes in Medicare Advantage. But researchers say the report does nothing to address ongoing concerns about MA insurers’ aggressive coding behavior and failed to address an important question: What happens after the home visit?

The report was prepared by FTI Consulting and focused on the Optum HouseCalls program, which sends advanced practice clinicians (e.g., nurse practitioners, physician assistants) to UnitedHealthcare members’ homes to assess their health status and ongoing care needs. Relying on the statistical sampling software used by the HHS Office of Inspector General, FTI generated a random sample of 200 home visits conducted during calendar year 2025. It found that 96.6% of the Hierarchical Condition Categories (HCCs) associated with those visits and submitted to CMS for reimbursement were supported in the medical record.

FTI pointed out that the 3.4% error rate, referring to those HCCs that were not adequately supported, was lower than the 4.94% program-wide net payment error rate reported by CMS and the 5% error rate used in HHS Risk Adjustment Data Validation (RADV) audits.

While the report provides some new information in that it compares diagnosis codes captured during a home visit with those found in medical records captured at the same home visit, “it doesn’t seem that useful,” weighs in David Meyers, Ph.D., associate professor and vice chair for the Department of Health Services, Policy, and Practice at Brown University School of Public Health. “The potential problem with the home health risk assessments is that they capture diagnoses that are not actually relevant to patient care.”

He continues, “This report does nothing to answer the question of whether these diagnoses are really needed, or if the assessments are a way to inflate risk scores.”

But that wasn’t a question the FTI report intended to answer. “Consistent with how CMS conducts RADV audits, FTI did not perform a clinical review or clinical validation audit,” a UnitedHealthcare spokesperson says. “FTI’s coding review of HouseCalls focused on whether the diagnosis codes submitted accurately reflected the conditions identified and documented by the provider.”

Report Follows Problematic OIG, Journal Findings

In prior reports looking at Medicare Advantage plans’ risk adjustment practices, UnitedHealth has been an outlier. In a July 2024 Wall Street Journal investigation linking an extra $50 billion risk-adjusted payments to MA insurers stemming from “hundreds of thousands of questionable diagnoses,” the Journal estimated that UnitedHealth received $8.7 billion in 2021 payments for additional diagnoses that were not treated. The report called out home visits as one way to gather questionable diagnoses.

In October 2024, the HHS Office of Inspector General found that diagnoses derived from health risk assessments and HRA-linked chart reviews but not appearing on any other service records led to roughly $7.5 billion in MA risk-adjustment payments for 2023. The OIG expressed concern about the lack of follow-up treatment for those diagnoses and pointed out that almost two-thirds of the $7.5 billion were related to in-home HRAs and HRA-linked chart reviews. OIG noted that UnitedHealth “stood out from its peers,” generating a disproportionate share of $3.7 billion in payments.

The FTI report is the latest in a series of independent assessments commissioned by UnitedHealth. In December 2025, UnitedHealth shared a set of third-party audits of its risk assessment, managed care and pharmacy benefits practices. Two of those were conducted by FTI; the third was prepared by Analysis Group. Together, those audits resulted in 23 process improvement recommendations. UnitedHealth said it has completed all of them.

Upon releasing the latest review, UnitedHealth CEO Stephen Hemsley said FTI’s findings “validate the integrity of the HouseCalls program, but we will continue to strive to ensure that only appropriately supported diagnoses are submitted to CMS.” He noted that through UnitedHealth’s Care Connectors team, patients who are identified as having new or worsening conditions during their home visit are linked to a Care Connector to schedule follow-up care or find additional resources. After the visit, Care Connectors “remain engaged” to ensure that members are getting the care they need, he added.

FTI’s evaluation, however, used only HouseCalls records, not records from providers’ offices or hospitals where care is usually provided, points out Paul Jacobs, Ph.D., a health economist who has previously studied coding intensity in MA.

“The question that matters for Medicare beneficiaries is: Are they receiving appropriate care for their needs?” says Jacobs, who is an associate scientist with the Department of Health Policy and Management at Johns Hopkins Bloomberg School of Public Health.

“The 2024 OIG report found that for 1.7 million enrollees, diagnoses like these appear on no other medical records and with no evidence they led to follow-up care. What, then, is a diagnosis worth if it never leads to care?”

Jacobs’ own research found that encounter-based risk scores for MA enrollees in 2021 were 7.4% higher when in-home health risk assessments and chart reviews were included.

Researcher: ‘Accuracy Isn’t the Point’

The broader problem, says Jacobs, is coding intensity, referring to the more comprehensive diagnosis capture occurring in MA versus traditional Medicare. “We know from more than a decade of research that coding intensity creates huge revenue surpluses for MA plans, and we fund those surpluses as taxpayers and by every Medicare enrollee through increased Part B premiums,” he adds.

“Some may find it comforting to know that UHC/Optum’s assessment program is following its own rules,” he continues. “However, even if UHC/Optum’s coding is clean, accuracy isn’t the point — we are paying billions in extra payments for codes that often do not translate into care.”

The Medicare Payment Advisory Commission (MedPAC) has repeatedly recommended to Congress that diagnoses found during health risk assessments should be excluded from risk adjustment in MA.

“The primary concern of MedPAC, OIG, and lawmakers is that the diagnosis codes captured during the home health risk assessments inflate risk scores for beneficiaries when these diagnoses aren’t actually relevant to that patient’s care needs,” says Meyers. “This report does nothing to address any of these concerns.”

Despite reservations about in-home assessments, CMS continues to count diagnoses from such interactions in the determination of risk-adjusted payments. In 2013 and 2014, the agency issued proposals that would have restricted the use of home-derived diagnoses, but those policies never made it into final rules.

Lauren Flynn Kelly has been writing about the business of health insurance and pharmacy benefits since the early 2000s. Up until its closure in August 2025, she served a variety of roles at AIS Health, including beat reporter on Medicare Advantage and executive editor leading a specialized team of health business journalists. When she’s not busy paddleboarding at the Jersey Shore, she is helping clients build their brand authority through the creation of blog posts, case studies, white papers and other strategic content.