by Kevin Schofield

---

This weekend’s read is a report by the Congressional Research Service on the formation of a new agency in the federal government: the Advanced Research Projects Agency for Health, or ARPA-H.

The Congressional Research Service is the legislative branch’s own research arm: It studies particular topics that members of Congress might want to pass legislation on, and it writes reports on what it finds. If that sounds like pretty dry reading material, well, often you’d be right. But sometimes, as in this case, it gives us some interesting insights on “how the sausage is made” in the halls of Congress and the kinds of difficult compromises that lead to a final bill.

The idea behind ARPA-H is simple: filling in a gap in how health-care-related research is funded that was made all too obvious by the COVID-19 pandemic over the past few years. To be sure, the federal government funds plenty of health-care-related research in academia and industry, primarily through the National Institutes of Health (NIH), the research arm of the Department of Health and Human Services. But most of that is “basic research”: incremental advances in our existing body of knowledge. Basic research is a critical component, but it tends to be slow and careful, with peer review by other academics and research scientists guiding both the funding decisions up-front and what gets published after the research is done. It’s also often criticized as risk-averse: Research that stays close to the prevailing thinking in the field tends to get funded and published more often, so there is a strong incentive for researchers not to rock the boat by challenging the status quo. Basic research methodically moves the field forward, but it rarely leads to the kinds of breakthroughs that disrupt and reorganize a field around a new paradigm.

In the middle of the 20th century, the U.S. Department of Defense established the original Advanced Research Project Agency (ARPA, later renamed DARPA) to pursue “high-risk, high-reward” projects that would not be funded by the traditional basic research ecosystem. Part of DARPA’s approach involves skipping the peer-review process altogether and instead enabling program directors to make their own decisions about what to fund — and to make those decisions quickly to speed the process along. This approach generates a pretty high failure rate — by design — but the successes more than make up for it. Among DARPA’s more notable successes was what we now call the internet, originally created as a high-speed network to allow for data and resource sharing between research sites DARPA was funding.

The idea for ARPA-H, as pitched by the Biden administration last year, was to follow the DARPA model: Focus on high-risk, high-reward research in medicine and healt hcare that could rapidly reshape those fields. The COVID-19 pandemic was the wake-up call: It led the federal government to make enormous investments to rapidly develop not only new vaccines and treatments for the virus, but also the new technologies needed to develop them quickly (such as mRNA vaccines). 

One of the difficult questions for Congress and the Biden administration to answer was how to organize this new agency. Creating it from scratch would take longer, as it would need to build its own infrastructure and best practices, which seems particularly wasteful when there are already several departments that fund health-care-related research; but on the other hand, placing ARPA-H inside an existing organization, such as NIH, runs the risk that the risk-averse nature of the parent organization will prevent ARPA-H from reaching its intended focus on high-risk, high-reward research that moves quickly. Ultimately, the government arrived at a compromise: to place ARPA-H within NIH, but have its director report to the secretary of Health and Human Services. The hope is that it will enjoy the best of both worlds: leveraging NIH infrastructure, but keeping its decision-making independent. Congress also required ARPA-H to submit its annual budget request independently of the NIH budget so the risk-averse NIH culture wouldn’t have a chance to weigh in on ARPA-H’s riskier bets. Finally, Congress requires that ARPA-H not be housed at NHI facilities so as to insulate it from the NIH internal culture, and that it be housed in multiple, geographically distributed sites. At least initially, it has settled on three sites: one in D.C. to focus on “stakeholder engagement” and operations; a “customer experience” hub; and an “investor catalyst.” 

What makes the Congressional Research Service report interesting to read is that it lays out the debate over these issues and how they were resolved. Given the current political polarization in our government, it actually feels reassuring to read that behind the scenes, the right conversations were happening, and members of Congress came together to solve problems — not just to fill a critical gap in health care research funding, but also to sort out the best way to set it up for success. Congress gave ARPA-H $1.5 billion in startup funds in 2023, and increased it to $2.5 billion for the 2024 fiscal year.

ARPA-H is borrowing one more fundamental principle from its predecessor, DARPA: the “Heilmeier Catechism.” George Heilmeier, who ran DARPA from 1975 to 1977, created a list of eight questions to guide his program directors’ thinking about what to fund and how to evaluate the results:

• What are you trying to do? Articulate your objectives using absolutely no jargon. 
• How is it done today, and what are the limits of current practice? 
• What is new in your approach, and why do you think it will be successful? 
• Who cares? If you are successful, what difference will it make? 
• What are the risks? 
• How much will it cost? 
• How long will it take? 
• What are the midterm and final “exams” to check for success? 

The leaders at ARPA-H have added two more questions to the original list, to help customize it for the health care space:

• To ensure equitable access for all people, how will cost, accessibility, and user experience be addressed?
• How might this program be misperceived or misused (and how can we prevent that from happening)?

According to the CRS report, ARPA-H has selected four initial “focus areas” for its work: health science futures, scalable solutions, proactive health, and resilient systems (longer descriptions of each of these areas are in the report). 

Again, this report provides something we rarely receive from the federal government these days: reassurance that smart, thoughtful people spent time considering and debating difficult issues, reached reasonable compromises, and kicked off a new program that addresses an existing problem. 

Advanced Research Projects Agency for Health (ARPA-H): Overview and Selected Issues

---

📸 Featured image by Monkey Business Images/Shutterstock.com.

---