Last winter, Brian Miller and Ted Cho published a 15-page white paper on the Innovation Valley of Death. Here, here. American Enterprise Institute; Cho is at UCSF, Miller at Hopkins.
In July, Miller announced a two-part series on CMS & Valley of Death, followed by FDA & Valley of Death. Linked In here. Find Part 1, Coverage Reform, online at Hoover Institution here.
The authors' key takeaways are three:
- Delays in Access to Innovation: Differing regulatory standards between FDA approval and Centers for Medicare and Medicaid Services (CMS) coverage delay access to the life-changing innovations in medical devices and health technology over the past several decades, with a median 5.7-year lag.
- Prior Regulatory Reform Efforts Have Lagged: For 30 years, policymakers and regulators have introduced advisory committees, new coverage pathways, and other process improvements, but access remains limited—underscoring the failure of these reforms.
- Urgent Need for CMS Reform: Policymakers and CMS can provide clarity of coverage principles, improve local coverage determination processes, and expand transparency to facilitate access to innovation.
As far as CED goes, I see little evidence of benefit, and a new complex (byzantine) CED proposal for renal denervation for hypertension looks very messy to me (blog). CMS does only 4 NCDs per year, yet if its TCET plan was fully implemented, it would be doing 12 (4 NCDs, 4 TCET NCDs, and 4 TCET NCDs coming up for re-appraisal.) No way.
###
AI Corner
###
Here's an AI summary (Chat GPT 5).
The Hoover Institution brief “Crossing the Valley of Death, Part 1: The Innovation Imperative and Medicare Coverage Reform” describes the persistent gap between FDA approval of new medical technologies and Medicare coverage sufficient to ensure patient access—a delay averaging about 5.7 years. This “Valley of Death” arises because innovators face two sequential regulatory hurdles: FDA’s safety and efficacy review, followed by CMS’s coverage determination under its “reasonable and necessary” standard. The agencies operate under different statutory mandates, resulting in misaligned processes and timelines that slow adoption of innovations and reduce their potential impact on patient care.
The authors frame the case for reform as both an innovation imperative and a moral obligation. While 45 percent of Medicare beneficiaries experience impairments in activities of daily living and over 6,000 rare diseases lack treatments, the United States has a strong history of medical innovation—producing more than 1,200 new drugs in the past 60 years and transforming care for conditions such as HIV and heart disease. Despite these successes, many conditions still impose enormous human and economic costs, such as insulin-dependent diabetes, which affects millions and drives billions in health care expenditures. The authors argue that U.S. policy should focus on removing inefficiencies that hinder timely adoption of proven innovations.
Medicare coverage reform has been attempted before. In 1999, under Administrator Nancy-Ann DeParle, CMS restructured its coverage analysis group and created the Medicare Evidence Development & Coverage Advisory Committee (MEDCAC). In 2021, the Trump administration introduced the Medicare Coverage of Innovative Technologies (MCIT) rule, which granted automatic four-year coverage for FDA-designated breakthrough devices. The Biden administration repealed MCIT in 2022 and replaced it with the Transitional Coverage for Emerging Technologies (TCET) pathway, which accepts only five devices per year and excludes diagnostics. In 2023, legislation (H.R. 1691) was introduced to restore MCIT-like transitional coverage.
The current coverage process includes National Coverage Determinations (NCDs), Local Coverage Determinations (LCDs) by Medicare Administrative Contractors (MACs), and Coverage with Evidence Development (CED).
(BQ: See Joe Grogan's WSJ Op Ed, nix CED, from June 2025.) (See also a WSJ op ed, nix CED for amyloid drugs, August 2025.)
MEDCAC’s activity has declined sharply in recent years, and the number of NCDs has dropped from a peak of 20 in 2003 to just four in 2022. CED, intended to accelerate coverage while gathering evidence, often stretches for years, drains resources, and rarely leads to full coverage—only three of 26 CED programs since 2005 have transitioned to routine coverage. The authors recommend a moratorium on new CED programs until CMS establishes defined milestones, completion criteria, and timelines. They also criticize CMS’s increasingly skeptical stance toward drugs granted FDA accelerated approval, noting that states like Oregon have sought to exclude such drugs from Medicaid coverage.
A central policy concern is the lack of a statutory definition for “reasonable and necessary.” CMS currently defines this administratively as safe and effective, not experimental or investigational, and appropriate for the Medicare population. Without a national definition, MACs apply their own interpretations, leading to inconsistent access based on geography. The authors recommend codifying broad evaluative principles in rulemaking, clarifying the division of responsibility between NCDs and LCDs, and maintaining local flexibility for positive customization.
Finally, the report proposes improvements to the LCD process, including imposing transparent timeframes for key steps, enhancing CAC membership with MEDCAC experts, posting CAC membership and technical questions at least 30 days before meetings [*], and holding annual CMS public meetings with MACs to review policy and operational lessons. The authors conclude that innovation is both an economic and moral imperative, and that streamlining, standardizing, and making coverage decisions more transparent can reduce inequities, improve access, and provide innovators with greater certainty, all in a way that should attract bipartisan support.
##
* Noridian has posted a meeting on molecular cancer tests, but by 8/11 offered no meaningful agenda for the 9/4 meeting.
