Source: CMS via ResDAC. Data Use Agreement required (no fee), 2-3 week approval. 5% random sample of all fee-for-service Medicare: inpatient, outpatient, physician, SNF, home health, hospice, DME.
The problem this solves: Aggregated Medicare reports flatten the questions that matter most. What happens to a patient after a denial? Which procedures add volume without improving outcomes? When a physician's office is acquired by a hospital, does the bill for the same visit double? Claim-level data answers these one event at a time, tied to the patient. Everything else is inference.
Source: CIVHC (civhc.org). Custom data extract with Data Use Agreement. Commercial, Medicare Advantage, Medicaid, and MA plan coverage. Standard researcher access.
The problem this solves: Commercial insurer prices are set in private. Hospital chargemasters are fiction. PBM rebates hide inside bilateral contracts. An all-payer claims database is the only way to see what a specific insurer actually pays a specific hospital for a specific procedure, what a plan pays a PBM versus what the PBM pays the pharmacy, and whether denial rates differ by payer or diagnosis. Colorado runs one of the most researcher-accessible APCDs in the country.
Grant writers, we need you: Philanthropic funders like the Colorado Health Foundation run RFPs that can underwrite work like this. If you have experience writing health-research grants, consider donating your time to partner with us on a submission. Reach out via the "Get in touch" link above.
Source: AHRQ HCUP State Inpatient Databases. ~$500/state. California: 400+ hospitals, 3M+ discharges/year. New York: 200+ hospitals. Florida: 200+ hospitals, one of the largest Medicare-age markets in the country. Student/nonprofit discounts available.
The problem this solves: One state is a case study. Three major state markets, covering roughly a quarter of the US population and every major insurance model, is national evidence. Hospital discharge data lets us map procedure volume, DRG severity drift, readmissions, and end-of-life intensity at the hospital level across California, New York, and Florida. Findings that replicate across all three are very hard to dismiss as regional artifacts.
Source: Turquoise Health. Free tier covers 14 procedures; full dataset covers thousands. Research partnership may reduce cost. Derived from CMS-mandated public hospital price disclosures.
The problem this solves: The CMS price transparency rule requires hospitals to publish their negotiated rates in machine-readable format. In practice the files are huge, inconsistent, and only partially compliant. Turquoise has already parsed, normalized, and cross-walked the universe into queryable records. Without it, we cite studies. With it, we can show a reader exactly what their own hospital charges their own insurer for their own procedure, and compare that to the hospital across town.
Source: Midpage (midpage.ai). $99/month, cancel anytime. Already configured as a live connector in our research environment. Just needs activation.
The problem this solves: Every regulatory failure in US healthcare leaves a paper trail in the case law. The Anti-Kickback Safe Harbor that shields intermediary fees. The hospital mergers the FTC lost. The fiduciary-duty litigation that could rewire incentive structures inside public companies. Legal research is how you trace a mechanism back to its origin and understand which levers can actually be moved, rather than guessing at symptoms.
Source: CMS Virtual Research Data Center (VRDC) via ResDAC. ~$35K first year ($23K renewal). Full DUA + IRB approval required. Virtual-only access (no data downloads). All outputs reviewed by CMS for patient privacy.
The problem this solves: The 5% sample reveals patterns. The full file follows every Medicare patient across every provider, every year. It's the data that connects a denial to the hospitalization three months later, quantifies whether vertically integrated systems route patients to captive providers at higher cost, and settles the fee-for-service-vs-Medicare-Advantage volume debate with individual-level evidence. This is what peer-reviewed research, regulatory citations, and congressional testimony are built on.