The Centers for Medicare & Medicaid Services (CMS) released a dataset on February 14, 2026, containing provider-level spending data for Medicaid services, intended to help identify potential fraud, waste, and abuse. The release follows a November 2024 comprehensive plan outlining CMS’s strategy for Medicaid integrity through fiscal year 2028, and builds on efforts dating back to 2010 with the establishment of the Center for Program Integrity (CPI).
The dataset includes information on services rendered between 2018 and 2024, encompassing both fee-for-service claims and those processed through Medicaid managed care organizations. Specifically, the data provides the National Provider Identifier (NPI) for both the billing and servicing provider, the Healthcare Common Procedure Coding System (HCPCS) code for the service, the month and year the service was provided, the number of beneficiaries seen, the number of procedures delivered, and the total amount paid.
However, the dataset excludes significant portions of Medicaid spending, notably institutional care and prescription drug costs. Hospital care alone accounts for 37% of total Medicaid expenditures, making its omission a substantial limitation. The data lack crucial contextual information necessary for a comprehensive assessment of spending patterns. This includes enrollment numbers, benefit designs, state-specific payment rates, and diagnoses associated with the procedures. The absence of place of service information – whether a service was delivered in-person or remotely – and other modifiers further restricts the data’s analytical utility.
CMS acknowledges the potential for misinterpretation when using the data in isolation. A key concern highlighted by the agency is the varying granularity of procedure codes. For example, the code used for “personal care” encompasses a wide range of service durations, from 15 minutes to a full day, while codes for psychotherapy are more precisely defined by visit length (30, 45, or 60 minutes). This inconsistency can skew comparisons between different service categories. Without including institutional spending, personal care appears as the largest spending category in CMS’s example, a result that would change with a more complete dataset.
Another challenge lies in the comparability of providers. The data include individual practitioners as well as large organizations like state and local government agencies. In CMS’s illustrative analysis, ten of the twenty largest “providers” were identified as state or local government entities that both administer and directly deliver Medicaid benefits, particularly for behavioral health and developmental disabilities. Variations in how states structure their Medicaid delivery systems further complicate comparisons.
The quality of the underlying data, sourced from the Transformed Medicaid Statistical Information System (T-MSIS), similarly presents a concern. CMS maintains a “data quality atlas” to identify potential issues, and the agency’s own reporting indicates data quality concerns in several states. Specifically, CMS reported in 2024 that six states had unusable spending data, and an additional sixteen states had data of “high concern.” It remains unclear how these data quality issues were addressed in the publicly released dataset.
The release of this data also occurs against a backdrop of significant changes to Medicaid enrollment and service utilization driven by the COVID-19 pandemic. The continuous enrollment period initiated during the pandemic led to increased enrollment, while heightened awareness of behavioral health and long-term care needs drove increased demand for those services. Changes in state policies regarding coverage, eligibility, and provider payment rates further influenced spending patterns between 2018, and 2024.
CMS has not yet announced a specific timeline for follow-up analysis or enforcement actions based on the released data. The agency continues to collaborate with states on program integrity efforts, as outlined in the November 2024 plan, and offers training through the Medicaid Integrity Institute.
