Free Healthcare Market Data: 12 Sources Founders Can Use

Three sources answer most healthcare market-sizing questions before you touch anything else: CMS for spending and claims, CDC for prevalence and mortality, and AHRQ for utilization and cost per encounter. Together they cover the three numbers almost every early healthcare TAM needs — and all three are free.

Quick Answer: Start with CMS (spending and claims), CDC (prevalence and mortality), and AHRQ (utilization and cost). Layer in Census and BLS to localize the picture, state agencies to fill regional gaps, and PubMed or a professional association to sanity-check what you found.

Paying a research firm to pull these numbers costs real money you don't have before you have revenue to justify it. Government health agencies, federal statistical offices, and academic databases already publish most of what a market-sizing exercise needs — the skill is knowing which one answers which question, not finding a vendor who claims to know for you. Resolving a sizing assumption this way is exactly the kind of work that belongs early in any structured startup idea validation process, and if your idea sits specifically in digital health, it slots into the market-sizing step of validating a healthtech startup idea — after you've confirmed the clinical problem is real, before you spend money building anything.

Selection criteria: coverage, granularity, and update cadence

Judge every source on three axes before you trust a number pulled from it: how much of the market it actually covers, how finely you can slice it, and how stale it is by the time you see it. A source can be excellent on one axis and useless on another — most bad market-sizing numbers come from a mismatch on one of these three, not from a source being "wrong."

• Coverage — does the dataset span the whole population and every payer, or only one segment (Medicare-only, one state, one condition)? A Medicare-only figure applied to an all-payer market will mislead you in a predictable direction.
• Granularity — can you filter to your specific condition, age band, geography, or care setting, or only to a broad category that buries the number you actually need?
• Update cadence — how often is it refreshed, and how much lag sits between the real-world event and the day it appears in the dataset? Claims data in particular can run a year or more behind.

No single free source clears all three axes for every question you'll ask. The practical fix is triangulation: pair a broad, laggy source against a narrower, faster one, and treat any large disagreement between them as a prompt to dig deeper — not as proof that one of them is simply wrong.

CMS: Medicare Claims and Spending Data

Best for: national spending and utilization benchmarks by procedure, service line, or diagnosis. The Centers for Medicare & Medicaid Services publishes claims and spending data through Data.CMS.gov and the National Health Expenditure Accounts, covering how much is spent, on what, and how that volume shifts year over year. Its native lens is the Medicare Fee-for-Service population, so treat any figure you pull as a Medicare-shaped signal, not a full-market number, unless you adjust for that.

Example query: look up spending and claim volume for a specific procedure or diagnosis code across CMS's public use files to see how utilization and average allowed cost have trended over recent years.

CDC: National Center for Health Statistics and WONDER

Best for: disease prevalence, mortality, and health-behavior rates broken out by state, age group, and demographic. The CDC's National Center for Health Statistics runs ongoing surveys — the National Health Interview Survey and NHANES among them — and publishes mortality, natality, and cancer data through its WONDER query tool, plus behavior data through the Behavioral Risk Factor Surveillance System.

Example query: pull the prevalence rate for a chronic condition by state through WONDER or BRFSS, then multiply against a state's adult population to build a first-pass estimate of your addressable patient count.

AHRQ: HCUP and MEPS for Utilization and Cost

Best for: how often people actually use the healthcare system and what it costs them when they do. The Agency for Healthcare Research and Quality maintains the Healthcare Cost and Utilization Project (hospital discharge and emergency-department data) and the Medical Expenditure Panel Survey, a household survey covering spending, utilization, and insurance status together.

Example query: use MEPS to estimate typical annual out-of-pocket or total spend for a condition category, then use that figure as the anchor for a willingness-to-pay assumption instead of guessing.

U.S. Census Bureau: Population, Insurance Coverage, and Business Patterns

Best for: the population denominators every market-sizing calculation needs underneath it — total people, income distribution, and insurance coverage rates. The American Community Survey, Small Area Health Insurance Estimates, and County Business Patterns together give you who lives where, what they earn, whether they're insured, and how many healthcare establishments already operate in a given county.

Example query: pull the uninsured rate by county from Small Area Health Insurance Estimates to flag markets where a self-pay product would face a harder sell than the national average suggests.

Bureau of Labor Statistics: Healthcare Workforce and Wage Data

Best for: sizing the supply side — how many of a given clinician or healthcare worker type exist, where, and what they're paid. The Occupational Employment and Wage Statistics program breaks this out by occupation and geography, which matters directly if your product sells to or staffs around a specific role.

Example query: pull employment counts and mean wage for a target occupation, such as registered nurses or medical billers, by state, to size a workforce-facing product's realistic seat count in your first launch market.

State Health Departments and All-Payer Claims Databases

Best for: regional granularity that national datasets can't give you, at the cost of consistency — coverage, format, and access rules vary widely from state to state. A number of states operate an all-payer claims database (APCD) that aggregates commercial, Medicaid, and Medicare claims into one regional view, alongside vital statistics offices and licensing boards that publish their own registries.

Example query: check whether your target state runs an APCD, then request or download its aggregate utilization reports for your specialty before assuming a national figure applies locally.

HRSA: Health Workforce and Shortage-Area Data

Best for: finding where care access is thin, which is often a proxy for unmet demand. The Health Resources and Services Administration maintains the Area Health Resources Files and designates Health Professional Shortage Areas and Medically Underserved Areas, alongside data on the federally funded health centers that serve them.

Example query: query HRSA's shortage-area tool for your target geography to check whether the access gap your product addresses is already officially recognized and mapped, which also helps when you're pitching grant-funded or public-health buyers.

FDA openFDA: Adverse Events, Recalls, and Device Clearances

Best for: competitive and safety-signal research on drugs, devices, and products already on the market in your category — not for sizing demand directly. The FDA's openFDA platform exposes an API over adverse event reports, device recalls, and device clearance records, including 510(k) and premarket approval filings.

Example query: search openFDA's device database for existing 510(k) clearances in your device category to map the competitive and regulatory landscape before you design your own submission strategy.

PubMed and NIH: Peer-Reviewed Prevalence and Outcomes Research

Best for: grounding your prevalence and outcomes assumptions in independently reviewed research instead of a vendor's marketing claim. PubMed indexes peer-reviewed literature across medicine and public health, and the National Institutes of Health's component institutes publish their own public datasets and research summaries on top of it.

Example query: search PubMed for recent peer-reviewed prevalence or incidence studies on your target condition, and use them to cross-check the figure you built from CDC or AHRQ data rather than relying on either source alone.

County Health Rankings & Roadmaps: County-Level Health Snapshots

Best for: a fast, pre-aggregated view of health outcomes and contributing factors at the county level, when you don't have time to build your own composite from raw federal files. The program — a collaboration between the Robert Wood Johnson Foundation and the University of Wisconsin Population Health Institute — packages dozens of federal and survey sources into one comparable county-by-county ranking.

Example query: compare rankings across your candidate launch markets to prioritize the counties where the underlying need your product addresses looks most acute.

Dartmouth Atlas of Health Care: Regional Utilization Variation

Best for: understanding that healthcare utilization varies by region independent of how sick the population actually is — a fact that explains why an identical product can land completely differently in two markets that look alike on paper. The Dartmouth Atlas project has spent decades documenting this variation across hospital referral regions.

Example query: compare regional utilization data for a relevant procedure or spending category to check whether your target region is historically a high- or low-utilization market before you build a go-to-market plan around it.

KFF: Health Policy and Survey Data

Best for: policy context and survey-based opinion and behavior data that the raw government datasets don't capture — how people say they feel about cost, coverage, and access, and how payer mix shapes the healthcare economy in a given state. KFF (the Kaiser Family Foundation) publishes ongoing health-policy research, state-by-state fact pages, and a recurring public-opinion tracking survey.

Example query: use KFF's state health facts pages to understand the payer mix in your target market — the same question that sits underneath mapping provider, patient, and payer relationships correctly in any healthtech validation.

Comparing all 12 sources by data type and use case

The table below lines up all twelve sources against what they primarily measure and the sizing question they answer best, so you can pick a starting point instead of opening all twelve tabs at once.

Read the table as a starting map, not a substitute for judgment: the "Best Use Case" column tells you which sizing question a source answers first, not the only question it can answer. Most credible healthcare market-sizing exercises end up pulling from four or five of these rows, not just one.

Key Takeaways

• CMS, CDC, and AHRQ cover the three numbers most sizing exercises need first — spending, prevalence, and utilization — and all three are free and public.
• Coverage, granularity, and update cadence are the three axes that make a number trustworthy or misleading; check all three before you use a figure, not just whether the source sounds authoritative.
• No single free source is complete on its own. Triangulate a broad national source against a narrower or faster one, and investigate disagreements instead of picking whichever number is more convenient.
• State-level detail usually means state-level sources. All-payer claims databases and state health departments fill gaps that CMS and CDC's national views leave open, at the cost of consistency across states.
• Workforce, competitive, and policy questions each have their own dedicated source — BLS for staffing, openFDA for competitors, and KFF for payer mix and sentiment — rather than expecting one dataset to answer everything.
• Peer-reviewed research is your check on government data, not a replacement for it. Use PubMed to confirm a prevalence figure looks reasonable against independently reviewed studies.

Frequently Asked Questions

Is CMS claims data actually free to use for market research?

Yes. CMS publishes claims, spending, and utilization data through Data.CMS.gov and related public use files at no cost and with no login requirement for most datasets. Some files require you to agree to a data use agreement before download, but that's a usage-terms step, not a paywall.

How current is free public healthcare data, and does the lag matter?

It varies by source, but lag is common — claims-based data like CMS's often trails by a year or more because claims take time to process and finalize, while survey-based sources like BRFSS publish on a more predictable annual cycle. For a market that's growing or shifting quickly, treat any lagged figure as a floor, and pair it with a faster proxy — search interest, provider interviews, or a recent survey — to catch what the lag might be hiding.

Can free public data replace a paid market research report?

For sizing and directional analysis, often yes — the underlying data behind most paid healthcare market reports is these same public sources, repackaged and interpreted. What free data won't give you is proprietary primary research, like a vendor's own customer survey or a competitor's private sales figures. Use free sources to build your own defensible estimate first, and reserve paid research for the specific gaps public data genuinely can't fill.

Which free source should I start with if I only have time for one?

Start with whichever of CMS, CDC, or AHRQ maps most directly to your product's core metric — CMS if you're selling into cost or reimbursement, CDC if your value proposition is tied to disease prevalence or outcomes, AHRQ if it's about utilization or how often people seek care. Pulling one solid number from the right starting source beats a shallow pass across all twelve.