← medicaid-provider-spending

Outlier Screen

Visuals in this post: EyePic Case Per-Enrollee Spending Concentration Top Providers Outlier Screen Provider Lookup Random Sample Methodology
data.nycuriosity.com › Medicaid Provider Spending › Outlier Screen
Medicaid July 2026

One query surfaces billing that grew fivefold in a year

Anyone with the public claims file can run this screen: providers that jumped from six figures to over $1 million and at least five times their prior-year billing. Most jumps have ordinary explanations. The point is that the first pass at oversight takes one query, not a subpoena.

Download CSV (members)
150 providers whose Medicaid billing grew at least fivefold in a single year
Every provider in the three states that went from at least $100,000 in one year to at least $1 million and five times its prior billing the next. Click a column header to sort.
Provider State Specialty Year Prior Year Paid Growth
Created by Tal Roded · NYCuriosity · Source: HHS Open Data, Medicaid Provider Spending by HCPCS

A screen is a starting point, not a verdict. Sudden billing growth has many legitimate causes: a new program, a merger, a state moving services into managed care, or simply better encounter reporting (which explains most extreme 2021 jumps in Texas). What a screen does is tell an auditor where to look first.

The EyePic companies would have appeared on screens like this. Harlem Eye Care's billing grew nine-fold between 2018 and 2019; Graham Eye Care went from $1,704 in 2019 to over $1 million in 2020. Neither was charged with anything until 2026.

Screen definition: providers with at least $100,000 in paid claims in the base year, at least $1 million the following year, and year-over-year growth of at least 5x. Sorted by growth multiple; the 150 largest fit those thresholds across NY, TX, and CA. Appearing here indicates only that billing grew quickly, not any wrongdoing.

Sources: HHS Open Data, Medicaid Provider Spending by HCPCS; CMS NPPES. Author's analysis; replication steps in the methodology.