The machine-readable files arrived on schedule, and almost nobody knew what to do with them. When the Centers for Medicare & Medicaid Services’ hospital price transparency rule took full effect in January 2021, requiring hospitals to publish negotiated rates for every payer and every covered service in a standardized, machine-readable format, the animating assumption was roughly Econ 101: sunlight disciplines prices, informed consumers shop, and market competition does the rest. What actually happened was more interesting and considerably more ambiguous. The data appeared — billions of rows of it — and revealed a pricing landscape so heterogeneous, so riddled with encoding inconsistencies, and so disconnected from the actual transaction that occurs at the point of care that the transparency illuminated very little about the forces that actually set prices.
The Aggregation Problem
Turquoise Health emerged into this gap. The company aggregates, normalizes, and indexes the machine-readable files that hospitals are now required to publish, making them searchable and analytically accessible in ways the raw files are not. By 2023, the platform had indexed rates from more than four thousand hospitals — a genuine feat of data engineering that cannot be understated. A 2022 analysis by the Patient Rights Advocate found substantial non-compliance among large hospital systems, some publishing files that were technically present but practically inaccessible — gigabyte-scale JSON constructs requiring specialized parsing tools unavailable to ordinary benefits staff. Turquoise Health solved exactly that problem. The question worth asking, with some rigor, is what the resulting data actually tells you.
Negotiated rates, it turns out, are not prices in any economically coherent sense. A hospital may report dozens of different rates for the same CPT code across different payers, different plan types, and different contract structures — some based on fee schedules, some on percent-of-charges, some on case rates, some on DRG multiples with complex outlier provisions layered on top. These rates are often the starting point of a reimbursement relationship that includes stop-loss clauses, carve-outs for implants or specialty drugs, quality bonuses, and supplemental payments tied to utilization thresholds. The published figure may be technically accurate and clinically meaningless for predicting what any given encounter would cost a specific plan. The transparency rule produced data without producing comparability.
Information Democratization vs. Market Reform
This is not an argument against the platform’s value. Researchers and sophisticated payers have found genuine utility in the aggregated data — particularly for benchmarking, for identifying egregious rate outliers, and for understanding the broad contours of pricing in specific geographies. The company’s analytics have helped some employers see, for the first time, that they were paying two or three times what another plan in the same metropolitan area paid for the same procedure at the same facility. RAND Corporation’s hospital price transparency research has repeatedly demonstrated that commercial rates in consolidated markets run at two to three hundred percent of Medicare across most regions. Turquoise Health gives employers the facility-level evidence to make that argument in contract negotiations rather than merely citing aggregate studies.
The deeper structural problem is that price transparency, in a market characterized by substantial provider consolidation, does not necessarily produce price competition. In markets where one or two health systems command dominant market share — and a growing proportion of U.S. hospital markets fit exactly this description — publishing negotiated rates may actually entrench existing pricing hierarchies rather than disrupt them. A dominant system charging three hundred percent of Medicare has no competitive reason to lower prices because a platform now makes that visible. The network imperative is what preserves the leverage, not the secrecy. Turquoise Health reveals the price. It does not change the power relationship that produced it.
The Map and the Territory
There is also a temporal constraint worth naming. Negotiated rates are contractual artifacts that expire and renegotiate on cycles of one to three years, often with material changes in methodology — a shift from percent-of-charges to case rates can alter effective payment levels by tens of millions of dollars without triggering any regulatory disclosure. The Turquoise Health dataset captures a moment in a continuous renegotiation process, which means its strategic value has a half-life. A rate published today may reflect a contract signed under conditions — volume levels, payer mix, competitive pressures — that have since shifted materially. The snapshot quality of the data is a structural constraint that no amount of engineering can fully solve.
What the platform has arguably done best is create a new category of market participant: the price-aware employer or benefits consultant who enters contract renewal conversations with data that previously required expensive actuarial analysis or proprietary benchmarking. That is a meaningful democratization of information. But it is worth distinguishing between information democratization and market reform. They are not the same, and conflating them produces a satisfaction with disclosure that may crowd out harder conversations about whether the underlying market structure — consolidated, insulated from consumer pressure, and deeply dependent on insurance intermediation — is reformable at all without structural intervention. The filing cabinet is now open. The question is whether anyone with the power to act on its contents is willing to do so.
