Setting the Record Straight: Separating Fact from Fiction on Health Insurance Marketplace Fraud

The Paragon Health Institute has published a series of new reports once again alleging large-scale “fraud” in health care. This time their target is enrollment in the Insurance Marketplace. They then use these allegations as rationale for a set of policy recommendations, including allowing the enhanced premium tax credits to expire at the end of 2025. If adopted by Congress, this policy would result in millions of Americans losing health care coverage and increase the cost of health care coverage for many millions more. It is therefore imperative that policymakers understand that Paragon developed these allegations using inaccurate data, dubious assumptions, and an apparent lack of understanding of how health insurance actually works. We set the record straight on each of their claims.

Paragon’s definition of “fraud” is misleading. Paragon claims fraud exists when Marketplace sign-ups in a state exceed Paragon’s estimate of eligible enrollees. Their approach has serious flaws.

• Paragon develops its estimates of how many people may be eligible using self-reported Census data. However, the Census uses different income and household size definitions than the Marketplace so there is no possibility of the data matching. In addition, it has been well-documented that Census data undercounts low-income populations. Paragon did not adjust its data to account for these issues.
• The Census collects reported income, whereas Marketplace eligibility is based on projected income. Projected income will naturally fluctuate from final income, particularly for lower-income individuals who work in hourly or tip-based jobs and who are more likely to change jobs in a year. This is why the federal government has a reconciliation process with the IRS to ensure accuracy when validating eligibility.
• The self-reported Census income data is from two years prior. Despite it being very common for income to change over time, Paragon assumes, implausibly, that it stays static and compares current year Marketplace enrollment to two-year-old income data.

Paragon’s assumption that enrollees without medical claims automatically equate to fraud misunderstands how health insurance works and misrepresents Centers for Medicare & Medicaid Services data. Some enrollees not using medical care in a given year is quite common — and even necessary — across all insurance markets, including employer-sponsored plans. Risk pooling depends on a mix of healthy and sick enrollees where healthier individuals with few to no claims help offset the costs of sicker enrollees. Without healthy enrollees, premiums would skyrocket, and the market could collapse.

The CMS data on which Paragon relies also includes a substantial number of partial year enrollees, including individuals who may have switched plans mid-year and are included twice in the data. (That is why the dataset Paragon uses shows 33.5 million enrollees in 2024 when actual Marketplace enrollment was only 21.4 million enrollees.)

Paragon’s methodology is unclear, but even their own annualized estimates reduce the number of enrollees without claims by 3.9 million. Still, when describing the rise in enrollees without claims from 2021 to 2024, Paragon simply assumes fraud and omits two critical factors:

• Special enrollment period selections more than doubled from 2021 to 2024. This trend was driven by the loss of Medicaid coverage following the expiration of COVID-19 continuous enrollment, as well as the creation of a new SEP for low-income enrollees. These SEP enrollees often only have coverage for a few months in a year, making it far more plausible that they did not file any claims.
• From 2021 to 2024, the Marketplace population became younger. From 2021 to 2024, the share of enrollees under age 18 rose by nearly 20%, while those aged 55–64 dropped by 16%. A younger, healthier Marketplace risk pool will naturally increase the number of enrollees without claims.

The Marketplace has provided affordable, comprehensive health coverage for millions of Americans. Paragon’s attempts to discredit this coverage are based on faulty analytics and sweeping generalizations. Their alternatives — such as expanded access to short-term limited duration health plans — would effectively bring us back to an insurance market where health plans could deny coverage based on pre-existing conditions and place limits on coverage. While these “plans” may be less costly, that’s for a reason: They cover less care and often subject enrollees to surprise bills and medical debt.

As Congress weighs the extension of the EPTCs — a critical tool in supporting access to affordable care and coverage — it is imperative that the Marketplace debate is grounded in fact, not fiction.

Aaron Wesolowski is vice president of research strategy and policy communications at the American Hospital Association.

Isaac Nicchitta is associate director of health analytics policy at the American Hospital Association.