CMS Releases Medicaid Spending Data: What Providers Need to Know

The Centers for Medicare & Medicaid Services (CMS) is increasingly focused on program integrity – addressing “fraud, waste and abuse” – within Medicaid and other health programs. This effort isn’t new, but it’s evolving, moving from a reactive “pay and chase” approach to one that relies more heavily on data analytics to proactively detect and prevent improper payments. In November 2024, CMS signaled its commitment to collaborative efforts with states, outlining opportunities for joint action. CMS’s work on fraud prevention spans various coverage types and provider settings, and on February 14, 2026, the agency released a new dataset intended to aid in these efforts.

The newly released dataset, containing provider-level spending information, is designed to assist identify unusual billing patterns. However, understanding what this data *can* and *cannot* tell us is crucial to avoid misinterpretations. The data includes information from 2018 through 2024, encompassing both fee-for-service claims and payments made by Medicaid managed care organizations. While the potential for identifying problematic trends is present, the dataset’s limitations require careful consideration.

What the Data Include – and Exclude

The dataset provides seven key pieces of information:

The National Provider Identifier (NPI) for the billing provider.
The NPI of the servicing provider (the individual or organization actually delivering the care).
The procedure code (HCPCS code) used for billing.
The month and year the service was provided.
The number of beneficiaries who received the service.
The number of procedures delivered (claim count).
The total amount paid for the services.

It’s important to note what’s *not* included. The data exclude institutional records – meaning spending at hospitals and other facilities – and all prescription drug information. These omissions are significant, as hospital care accounts for 37% of total Medicaid spending, making it the single largest expense category. KFF data highlights the substantial portion of Medicaid dollars allocated to hospital care, underscoring the limitations of this dataset for a complete picture of spending.

Beyond these broad exclusions, several other critical pieces of information are missing. The dataset doesn’t include data on Medicaid enrollment numbers, which fluctuate based on state policies, economic conditions, and demographic shifts. It also lacks information on the benefits covered by each state’s Medicaid program, payment rates, and patient diagnoses. Details about the place of service (in-person vs. Telehealth) and other modifiers that provide context to the services are absent.

Potential for Misleading Conclusions

Data analytics are a powerful tool, but relying solely on this dataset could lead to inaccurate conclusions. One issue is the varying level of specificity in the procedure codes. For example, the data categorize “personal care” as a single procedure, encompassing services ranging from 15 minutes to a full day. This contrasts with psychotherapy, which has separate codes for 30-, 45-, and 60-minute visits. As CMS itself illustrates in examples of how the data could be used, personal care appears as a significant outlier in spending, but this is partly due to the broad definition of the procedure code.

Another challenge lies in comparing providers. The dataset includes individual practitioners, group practices, clinics, and even entire county and state health departments. In CMS’s example, 10 of the 20 largest “providers” are actually government agencies responsible for both administering and delivering Medicaid benefits. States vary significantly in how they deliver these benefits, with many health departments directly providing services, particularly for behavioral health and individuals with developmental disabilities.

Finally, the data’s quality and creation process are not fully transparent. The data originate from the Transformed Medicaid Statistical Information System (T-MSIS), a rich data source, but one that can have state-specific data quality issues. CMS maintains a data quality atlas to identify potential problems, but it’s unclear how these issues were addressed when creating the publicly released dataset. For instance, CMS reports that in 2024 data, six states had unusable spending information, and another 16 had data of “high concern.” It’s unknown whether this problematic data was included in the public file or if a different version of T-MSIS was used.

The Impact of the Pandemic and Policy Changes

The period covered by the data – 2018 to 2024 – encompasses significant changes in Medicaid, particularly due to the COVID-19 pandemic. The pandemic led to increased enrollment through the continuous enrollment provision and heightened awareness of unmet needs for behavioral health and long-term care. As states expanded access to these services, both utilization and spending increased, driven by changes in coverage, eligibility, and provider payment rates. Without accounting for these contextual shifts, it’s tricky to interpret spending patterns accurately.

The Center for Program Integrity (CPI), established within CMS in 2010, has been working to shift Medicaid program integrity efforts from a reactive “pay and chase” model to a more proactive, data-driven approach. CPI’s work includes providing training to states through the Medicaid Integrity Institute and offering access to comprehensive data sets. This new dataset is a step in that direction, but its limitations must be acknowledged.

Looking Ahead: Data Use and Ongoing Evaluation

CMS intends for this data to be used to identify potentially problematic billing patterns, but it’s crucial to remember that correlation does not equal causation. Unusual spending patterns may indicate fraud, waste, or abuse, but they could also be the result of legitimate factors, such as increased demand for services or changes in state policies. Further investigation and contextual analysis are essential before drawing any conclusions.

The release of this dataset is part of a broader CMS effort to enhance Medicaid program integrity, as outlined in the Comprehensive Medicaid Integrity Plan for FYs 2024-2028. This plan emphasizes risk-based approaches, collaboration with states, and a commitment to equity. CMS and states will demand to continue working together to refine data analysis techniques and address the challenges of identifying and preventing fraud, waste, and abuse while protecting access to care for Medicaid enrollees.