Yale CRRIT

Demographic Data Supporting FDA Authorization of AI Devices for Alzheimer Disease and Related Dementias This cohort study examines the availability and representativeness of data supporting US Food and Drug Administration–authorized artificial intelligence and machine learning devices to treat Alzheimer...

This team was comprised of @hopkinsmedicine.bsky.social Krista Chen, CRRIT Co-Director @jsross119.bsky.social, @yalesph.bsky.social Adam Cohen, @pennmedicine.bsky.social Jason Karlawish, and @hopkinsmedicine.bsky.social Ravi Gupta (4/4)
🔗 jamanetwork.com/journals/jam...

Lack of transparency raises concerns regarding underperformance, underdiagnosis, & inequalities in care recommendations. Enforcing data transparency guidelines can help promote safe use of devices. (3/4)

Key point: Transparency of evidence across demographics for AI/ML devices was limited. Transparency is important for understanding appropriate application, esp for demographic representativeness of training & validation datasets. (2/4)

ICMYI: AI/ML-enabled medical devices have potential for earlier diagnosis & symptom monitoring for Alzheimer’s disease & other dementias. A new @jama.com current study led by @hopkinsmedicine.bsky.social student Krista Chen looks at demographic representativeness of these devices (1/4)

Exclusion of Older Adults from Obesity Treatment Pivotal Trials of GLP-1RAs and GIP/GLP-1RAs - Journal of General Internal Medicine Journal of General Internal Medicine -

This team comprised of @yaleschoolofmed.bsky.social Alissa Chen, @keckschoolusc.bsky.social Yixuan Liang, @yaleschoolofmed.bsky.social Kasia Lipska, and CRRIT Co-Directors @jsross119.bsky.social and @reshmagar.bsky.social

Full analysis: link.springer.com/article/10.1...

Many older adults have obesity, yet few were included in clinical trials for GLP1s. Study exclusion criteria (depression, prior cancer) may contribute to the under-enrollment of adults ≥ 65 years (1/2)

Regulatory change could improve biosimilar access in the US Tiffany E Jiang , Reshma Ramachandran , and Joshua J Skydel argue that US requirements for interchangeability are not supported by evidence and hamper use of biosimilars Biological medicines (biologi...

In a recently published article in @bmj.com, @yaleschoolofmed.bsky.social student Tiffany Jiang, CRRIT Co-Director @reshmagar.bsky.social & Joshua Skydel (Yale Rheumatology) investigate how eliminating these requirements can increase access to biosimilars (2/2)

🔗 www.bmj.com/content/390/...

Biosimilars provide low-cost options for patients and expanded access to treatment, but use of these treatments is hampered by @fda.gov requirements for interchangeability (1/2)

Compounded GLP-1 Receptor Agonists: Opportunity for Compassionate Conversation - Journal of General Internal Medicine Journal of General Internal Medicine -

In a new @journalgim.bsky.social viewpoint, @yaleschoolofmed.bsky.social & Yale Internal Medicine Ashwin Chetty and Dr. Alissa Chen Chen describe the importance of clinicians proactively discussing these treatments with patients. (2/2)

Full viewpoint: link.springer.com/article/10.1...

Many clinicians do not recommend compounded GLP-1RAs and are uninformed when patients seek these medications, which can leave patients vulnerable to misinformation. (1/2)

Revenue, Patents, and New Generic Competition for Prescription Drugs in the USA - Journal of General Internal Medicine Journal of General Internal Medicine -

Full analysis: link.springer.com/article/10.1...

High brand-name drug prices fall once a generic enters the market. In a new @journalgim.bsky.social article, Ravi Gupta, CRRIT Co-Director @jsross119.bsky.social, and colleagues from @portalresearch.org assess associations between patents, revenue, and generic competition.

Dr. Ross noted these letters “would allow other sponsors & scientists to better understand what precluded drug approval — was it safety concerns or lack of efficacy — & ideally advance on that’ to bring new drugs to market” (2/2)

F.D.A. Posts Collection of Letters Outlining Concerns About New Drugs

ICYMI: @fda.gov released a limited set of denial letters for drugs that were ultimately approved. CRRIT Co-Director @jsross119.bsky.social recently told @nytimes.com why these letters are important (1/2)

🔗 www.nytimes.com/2025/07/10/h...

CRRIT Co-Director @reshmagar.bsky.social provided public comments at the Prescription Drug User Fee Act (PDUFA) VIII Kickoff Meeting @fda.gov yesterday. Dr. Ramachandran discussed how the user fee process could be reformed to be more patient-centered & transparent.

Validating Surrogate Endpoints to Support FDA Drug Approval As surrogate markers are increasingly being accepted by FDA to support approval of new drugs and biologics, it is imperative that patients and clinicians understand whether such novel endpoints are re...

Surrogate markers are increasingly being used to support @fda.gov approvals instead of measuring clinical outcomes. In a @scientistsorg.bsky.social memo @reshmagar.bsky.social, Joshua Wallach & @jsross119.bsky.social outline how FDA can make transparent & strengthen evidence. fas.org/publication/...

F.D.A. to Use A.I. in Drug Approvals to ‘Radically Increase Efficiency’

The @fda.gov says it will “radically increase efficiency” with AI and embrace “radical transparency.” But as @reshmagar.bsky.social told @nytimes.com, agency officials recently kicked off a six-city listening tour—with Pharma CEOs—behind closed doors. Here’s the scoop:

Do companies continue promoting drugs granted
@fda.gov accelerated approval even after failed confirmatory trials? Yes—though less aggressively, finds Dr. Maryam Mooghali’s #ASCO25 Merit Award-winning research.

🔗 meetings.asco.org/abstracts-pr...

...her team found that information published in scholarly journals is becoming more accessible over time (see below ⬇️)—giving patients and their caregivers more agency to make the health decisions best for them. #InfoAsHealth
@jsross119.bsky.social @genpatient.bsky.social

Access to Information Cited in National Organization for Rare Disorders Reports This cross-sectional study describes the availability and access cost to the public of research findings and other clinical insights relevant to rare diseases.

For patients with rare diseases, high-quality health info can be scarce. Research led by CRRIT post-doc Mengyuan Fu, PhD, on @nordrare.bsky.social’s free online disease reports found that 50% of sources cited were unavailable online or behind paywalls (~$168 per report). But... 🧵

Congrats to Reshma Ramachandran MD, MPP, MHS (Yale NCSP '20–'22), winner of the 2025 Bernard Lown Award for Social Responsibility! 👏 A powerful voice for health equity. Learn More: bit.ly/reshma-lown2025
#HealthEquity #SocialResponsibility #YaleNCSP #PharmaReform #PublicHealth@reshmagar.bsky.social

Some more photos of our wonderful CRRIT trainee team presenting their research at #SGIM25! And we managed to get everyone together for dinner right by the beach 🏖️.

So proud of our CRRIT @yaleschoolofmed.bsky.social students & @ncspyale.bsky.social fellows for presenting their research projects on the evidence underlying @fda.gov approvals, FDA safety actions, costs of care, drug pricing, & more at #SGIM25! @jsross119.bsky.social @reshmagar.bsky.social #MedSky

Implications for Public Health Regulation if Chevron Deference Is Overturned This Viewpoint describes implications for medicine and public health if the US Supreme Court decides to overturn or narrow Chevron deference.

After the Supreme Court overturned Chevron deference last year, courts—not agencies—get final say on ambiguous laws. But judges often lack scientific expertise, and the ruling on lab-developed test shows the risks.

For more on public health post-Chevron:

A court decision on lab-developed tests sets a dangerous precedent An under-the-radar court ruling may be among the most significant harbingers of the crises to come for the FDA and public health.

In March, a Texas judge tied @fda.gov’s hands on its ability to regulate lab-developed tests—to prevent, for example, another Theranos. In @statnews.com, CRRIT’s @ktkadakia.bsky.social, @jsross119.bsky.social, & @reshmagar.bsky.social explain the dangers of this ruling:

Our co-Director's research & policy work was featured by @yaleschoolofmed.bsky.social including what led to her & @jsross119.bsky.social re-establishing CRRIT which brings together expertise across campus to study how drugs, devices, etc are regulated & how that can be improved for patients. #MedSky

This study was authored by Dr. Maryam Mooghali, Dr. Sanket Dhruva, Hollin Hakimian, Dr. Vinay Rathi, Kushal Kadakia (@ktkadakia.bsky.social), and Dr. Joseph Ross (@jsross119.bsky.social).

Nonconcurrent controls can be justified—but they risk bias, selective reporting, and confounding. To protect patients and earn trust in nonconcurrent controls, @fda.gov could require justifications of and disclose why these controls are used.

We found that:
- Among 88 high-risk devices approved,
- 52 (59.1%) used ≥1 trial with nonconcurrent controls, and
- The @fda.gov docs of just 3 (3.8%) included justifications for using nonconcurrent controls.

In this study, we reviewed @fda.gov approvals of high-risk therapeutic devices from 2019–2023 to see how often nonconcurrent controls were used—and whether @fda.gov justified their use.