RxEconomics Logo - Health Economics and Outcomes Research Consultancy specializing in AI-driven analytics

Real-World Evidence and Claims Data Analytics Consulting

Real-world evidence consulting turns real-world data—insurance claims, electronic health records, registries, and national surveys—into evidence that payers, regulators, and clinicians trust. We design and execute claims-based studies end to end: framing the question, selecting the right dataset, constructing defensible cohorts, applying rigorous causal methods, and delivering publication-ready results.

We do it fast. Comprehensive real-world evidence generation typically takes us 3-4 months rather than the 9-12 months common at traditional consultancies—typically 3× faster—because we pair 28+ years of health economics experience with the AI-accelerated HEOR workflow we have built into every stage of analysis. Speed never substitutes for rigor; it comes from removing the friction around it.

Why Does Real-World Evidence Matter Now?

Clinical trials demonstrate efficacy under controlled conditions. Real-world evidence shows what actually happens in routine care—who receives a therapy, whether they stay on it, what outcomes they achieve, and what it all costs. Payers increasingly demand that evidence for formulary and contracting decisions, health technology assessment bodies weigh it in value determinations and market access decisions, and the FDA now considers it in regulatory decision-making.

Real-world evidence also feeds directly into value demonstration. The utilization, cost, and outcomes estimates we generate from claims data become the inputs to the cost-effectiveness and budget impact models at the core of our health economics and outcomes research practice—one team, one evidence chain, no handoffs.

Which Real-World Datasets Do We Analyze?

Most consultancies say "claims data" and leave it there. We name our datasets, because dataset selection determines whether your study can answer its question at all.

  • MarketScan. Commercial claims from employer-sponsored plans—a workhorse for studying working-age populations, treatment patterns, and cost outcomes at scale.
  • Optum. Commercial claims with linked clinical elements, useful when your question needs both utilization detail and richer patient context.
  • Medicare Fee-for-Service. National administrative claims for beneficiaries in traditional Medicare—the standard for studying older adults, chronic disease burden, and post-acute care.
  • Medicare Advantage & Part D. Plan-level medical and prescription drug event data for the growing majority of beneficiaries in managed Medicare.
  • Medicaid. State administrative data—including MAX and TAF/T-MSIS research files and State Drug Utilization Data (SDUD)—covering low-income and disabled populations; essential for access, adherence, and policy questions in safety-net care.
  • Electronic Health Records (EHR). Clinical depth claims cannot capture—labs, vitals, and disease severity—for studies that need physiologic detail.
  • Patient Registries. Disease-specific longitudinal data with outcomes and clinical measures curated by specialty societies and sponsors.
  • MEPS. The Medical Expenditure Panel Survey—nationally representative data on healthcare use, expenditures, and insurance coverage for U.S. households.
  • NHIS. The National Health Interview Survey—population-level health status, access, and behavior measures that anchor burden-of-illness and policy analyses.

RxEconomics analyzes client-provided, licensed, or public-use datasets; dataset selection depends on the population, the outcome, your access rights, and the study question. We integrate multiple sources when a single database cannot carry the question—and if you are unsure which dataset fits your study, we will tell you in the first conversation, including when the honest answer is that none of them can.

What Claims Analytics Capabilities Do We Offer?

Retrospective Claims Studies

We design retrospective database studies that characterize treatment patterns, healthcare resource utilization, and costs for a defined population—burden-of-illness analyses, drug utilization reviews, and cost-of-care benchmarks. Our published work in this area includes a Medicaid burden-of-illness study of chronic hepatitis C and direct-acting antiviral use in The American Journal of Managed Care.

Comparative Effectiveness Research

When you need to know how a therapy performs against real-world alternatives, we build matched or weighted cohorts from claims and EHR data and estimate comparative outcomes with methods designed to withstand peer review and payer scrutiny—the same standard our healthcare economic expert witness work must meet under courtroom examination.

Medication Adherence Analysis

Adherence is where our track record runs deepest. Dr. Roebuck's Health Affairs study "Medication Adherence Leads To Lower Health Care Use And Costs Despite Increased Drug Spending" is widely cited, and companion work in Medical Care and Health Affairs extended those findings to Medicaid populations. We measure adherence, model its drivers, and quantify its downstream effects on utilization and cost. Those findings circulate well beyond the journals—they have been used in corporate professional education from Abbott's a:care program and in payer communications making the business and clinical case for adherence programs.

External Control Arms

For rare diseases and single-arm trials, we construct external control arms from claims, EHR, and registry data—defining eligibility criteria that mirror the trial protocol, addressing confounding explicitly, and documenting every design decision for regulatory and HTA audiences.

Proof: A Multi-Year Site-of-Care Research Program

Working with the Employee Benefit Research Institute (EBRI), RxEconomics has produced a repeatedly updated series of published analyses of hospital outpatient department (HOPD) markups relative to physician offices and other sites of care—spanning medical-benefit adjudicated drugs, oncology therapies, biologics and biosimilars, laboratory and imaging services, and other outpatient services. The series shows what disciplined claims analytics can do: same clinical service, different site, quantified price difference.

That work now informs the policy debate. The oncology site-of-care analysis was cited in 2026 written witness testimony before the House Energy and Commerce Subcommittee on Health in support of site-neutral payment reform.

How Do We Keep the Methods Rigorous?

Claims data reward careful methods and punish careless ones. Three disciplines anchor every study we deliver:

  • Causal inference. Propensity score matching and weighting, difference-in-differences designs, instrumental variables, and fixed-effects panel models—chosen to fit the question, not the fashion. These are the methods behind Dr. Roebuck's publications in Health Affairs, Medical Care, and the American Economic Review.
  • Cohort construction. Transparent inclusion and exclusion criteria, continuous-enrollment requirements, washout periods, and validated code-based algorithms (ICD-10, CPT, HCPCS, NDC)—documented so your reviewers can reproduce every step.
  • Sensitivity analysis. We stress-test findings across alternative specifications, follow-up windows, and outcome definitions, and we report what moves and what does not. That is what peer-reviewed rigor means in practice—and it is why our clients count on exceptional accuracy.

How Does the FDA Treat Real-World Evidence?

The 21st Century Cures Act directed the FDA to evaluate the use of real-world evidence in regulatory decision-making, and the agency's real-world evidence program now includes a framework and a growing body of guidance on study design, data standards, and the use of EHR and claims data in submissions. The practical implication for you: real-world studies intended for regulatory audiences must be designed to regulatory standards from day one—prespecified protocols, fit-for-purpose data, and transparent methods. We build studies that way regardless of audience, so the evidence holds up wherever you take it.

What Does a Claims-Based Engagement Look Like?

Every engagement starts with a scoping conversation—your question, your timeline, and the dataset realities that shape both. From there we deliver a written protocol, interim readouts as the analysis progresses, and final deliverables built for their destination: a manuscript, a value dossier, a payer presentation, or a regulatory submission.

Who Leads Your Study?

Dr. M. Christopher Roebuck, PhD, MBA, President & CEO of RxEconomics, personally leads every engagement. He brings 28+ years of experience in health economics and outcomes research, including service as Director of Strategic Research at CVS Caremark, and has authored 120+ publications & research reports—with peer-reviewed work in Health Affairs, Medical Care, and the American Economic Review. His publication record is independently verifiable through Google Scholar and ORCID. Founded in 2011 and based in Miami, FL, RxEconomics serves clients nationally—and when you engage us, the expert whose name is on the research is the one doing it.

Frequently Asked Questions

Real-world data (RWD) are the raw materials—claims, EHR records, registry entries, and survey responses generated in routine care. Real-world evidence (RWE) is what disciplined analysis produces from them: findings about treatment effectiveness, safety, utilization, and cost that can support clinical, payer, and regulatory decisions. The distance between the two is methodology, and that is precisely what we provide.

It depends on the population and the outcome. Commercial questions typically point to MarketScan or Optum; questions about older adults point to Medicare Fee-for-Service or Medicare Advantage and Part D data; safety-net and policy questions point to Medicaid; and nationally representative cost or access questions point to MEPS or NHIS. We assess fit—population coverage, follow-up length, and outcome capture—before recommending a dataset, and we integrate multiple sources when one alone cannot answer the question.

Our AI-enhanced approach typically delivers results 3× faster than traditional consultancies. Comprehensive real-world evidence generation might require only 3-4 months rather than 9-12 months, and a focused retrospective claims analysis can move faster still. Timelines depend on data access and scope, which we define precisely at the outset.

Yes—within a defined framework. Since the 21st Century Cures Act, the FDA has issued a real-world evidence framework and guidance addressing when and how RWE can support regulatory decisions, including the use of claims and EHR data. Acceptance hinges on fit-for-purpose data and rigorous, prespecified design—standards we build into every study.

For claims analytics specifically, the gains concentrate in code-set construction and cohort iteration: generative AI drafts and cross-checks the HCPCS, CPT, ICD-10, and NDC code lists that define exposures and outcomes, which shortens dataset preparation and makes it practical to stress-test alternative cohort definitions early in the study rather than after a misstep. Every AI-assisted code list and cohort specification is expert-verified before it drives an analysis, ensuring exceptional accuracy. For a deeper look, see our analyses of the current AI-in-HEOR landscape, the shift toward an AI-native research cycle, and the road ahead to 2030.

Put Your Claims Data to Work

Bring us the question keeping your evidence plan stalled—we will tell you which dataset can answer it, how long it will take, and what it will cost. You will hear back within one business day.

Request Your Consult