Research Methodology

Definitive Healthcare Medicare Standard Analytics Files (SAFs) are part of the Limited Data Set (LDS) files released on a yearly and quarterly basis by the Centers for Medicare & Medicaid Services. The SAFs capture adjudicated claims and are 100% Medicare fee-for-service claims (Medicare Advantage not included). Claims adjudication refers to the determination of the insurer’s payment or financial responsibility after the member’s insurance benefits are applied to a medical claim to yield “final action” claims. The SAFs are available for all claim settings (e.g., inpatient, outpatient, home health, skilled nursing facility, and hospice).

The Definitive Healthcare commercial data set, which includes Medicaid, is sourced from some of the largest medical claim clearinghouses in the United States and includes a mixture of professional and institutional claims processed through those clearinghouses. Professional claims are generated for work performed by physicians, suppliers, and other non-institutional providers for both inpatient and outpatient services. Institutional claims are generated for work performed by hospitals, skilled nursing facilities, and other institutions for inpatient and outpatient services (e.g., use of equipment/supplies, laboratory, radiology). Definitive Healthcare aggregates claims data and reports as cases.

Patient-level, chronic disease–specific claims data derive from the Managed Care Digest Series^® Local Trends Summary^™ database. These data come from health care professional and institutional insurance claims, including all physician specialties and all hospital types. IQVIA gathers prescription activity from the National Council for Prescription Drug Programs (NCPDP). These data account for some 4 billion prescription claims annually, or more than 92% of the retail prescription universe and 72% of the traditional and specialty mail order universe.

Proprietary lab data derive from one of the largest independent commercial lab companies in the U.S. Patient information is de-identified, matched, and linked with other patient data assets (e.g., medical claims data). The most common attributes used are the de-identified patient ID, observation date, diagnosis, test name, test code, and test result.

Claims undergo a careful de-duplication process to ensure that when multiple, voided, or adjusted claims are assigned to a patient encounter, they are applied to the database, but only for a single, unique patient. Through its patient encryption methods, IQVIA creates a unique, random numerical identifier for every patient, and then strips away all patient-specific health information that is protected under the Health Insurance Portability and Accountability Act (HIPAA). The identifier allows IQVIA to track disease-specific diagnosis and procedure activity across the various settings where patient care is provided (hospital inpatient, hospital outpatient, emergency rooms, clinics, doctors’ offices, and pharmacies), while protecting the privacy of each patient.

The Centers for Medicare & Medicaid Services (CMS) provided data on the following: Medicare Advantage enrollment and Star Ratings, Medicare costs, accountable care organization (ACO) payments, readmission rate penalties, and Medicaid enrollment. In June 2024, CMS announced that, in light of court decisions, it would recalculate the 2024 Star Ratings for 2025 Bonus Payment purposes. This digest includes analysis of only the original release of the 2024 Star Ratings (CMS. [2024]. Update to 2025 Quality Bonus Payment Determinations. https://www.cms.gov/files/document/updateto2025qualitybonuspaymentdeterminations.pdf).

Medicare readmission rate penalty data for fiscal year (FY) 2024 are from the CMS Hospital Readmissions Reduction Program. For FY 2024, CMS calculates excess readmission ratios (the ratio of predicted readmissions to expected readmissions), dual proportions, and hospitals’ payments for each condition/procedure and overall using discharges that occurred during a non-contiguous 29-month period, including portions of 2019–2022. Medicare Shared Savings Program (MSSP) ACO performance data are for performance year 2022 and are current as of April 2024.

Data on Medicare fee-for-service (FFS) actual costs by setting are derived from CMS’s Geographic Variation Public Use Files (PUFs), which are current as of May 2024. Overall Medicaid enrollment data are from state Medicaid and Children’s Health Insurance Program (CHIP) applications, eligibility, determinations, and enrollment data, and are current as of December 2023. Information on these and other public sources listed can be found on their respective websites.

Data on HIXs are provided by PUFs from CMS. State-level HIX enrollment data come from the state-level PUF that includes total health plan selections in all 50 states plus the District of Columbia. The PUF provides state-level data on metrics such as average monthly premium, financial assistance, age, gender, metal level, self-reported race and ethnicity, rural location, household income as a percentage of the federal poverty level (FPL), and plan switching behavior among consumers with a plan selection. In addition, the state-level PUF includes data on dental plan selections and Basic Health Plan (BHP) enrollments.

CMS provided historical and projected data for the National Health Expenditure Accounts, as well as information on Medicaid enrollment trends. IQVIA served as the source of social determinants of health (SDoH) data for Type 2 diabetes patients at the county level. Cancer incidence rates are from the National Cancer Institute; CMS provided data on the location of the participants in the Enhancing Oncology Model. Maps in this digest were generated using R (R Core Team [2021]. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. https://www.R-project.org/).