CDC Director Search, CMS Pilot & AI in Health – STAT News Roundup

Get your daily dose of health and medicine every weekday with STAT’s free newsletter Morning Rounds. Sign up here.

The way we seek health information is shifting, and increasingly, that involves turning to artificial intelligence. A new KFF poll reveals that one in three U.S. Adults have used AI chatbots for health-related advice in the past year, a figure that underscores a growing trend of individuals leveraging technology for immediate access to information. This reliance isn’t solely driven by convenience; affordability and access to traditional healthcare similarly play a significant role for roughly 20% of those using these tools.

Navigating the Rise of AI Health Information

The KFF poll, conducted primarily online with over 1,000 participants between late February and early March, highlights a complex relationship with AI in healthcare. While many are drawn to the instant answers these chatbots provide, a substantial proportion – 41% of users – are also willing to upload personal medical data, like test results and doctor’s notes, for more tailored advice. This willingness comes despite a majority expressing concerns about data privacy.

Perhaps more concerning is that a significant number of individuals aren’t following up with healthcare professionals after receiving AI-generated advice. 58% of those who sought mental health information and 42% who inquired about physical health concerns did not consult a doctor afterward. This raises questions about the potential for misdiagnosis, delayed treatment, and the overall impact on the patient-physician relationship.

Deepfakes and Diagnostic Accuracy: A Growing Concern

The increasing sophistication of AI extends beyond chatbots and into medical imaging. Recent research, detailed in STAT News, demonstrates the difficulty radiologists face in identifying AI-generated, or “deepfake,” X-rays. A study involving 17 radiologists found that fewer than half could detect subtle anomalies in fabricated images. Even when alerted to the possibility of deepfakes, accuracy only improved to 75%.

This poses a potential risk to patient care, as misinterpreting a deepfake image could lead to incorrect diagnoses and inappropriate treatment plans. Researcher Mickael Tordjman emphasizes the need for heightened awareness and improved detection methods, stating that “Anyone could be confronted at some point with a fake X-ray and should be able to differentiate real and fake.” The study doesn’t currently indicate widespread disruption in health systems due to deepfakes, but the potential for harm is clear.

CDC Leadership and Vaccine Policy Shifts

Meanwhile, changes are unfolding at the Centers for Disease Control and Prevention (CDC). Robert Malone, a physician and ally of Robert F. Kennedy Jr., is stepping down from the CDC’s Advisory Committee on Immunization Practices (ACIP) following a federal judge’s decision to block the panel’s work. Malone’s appointment last year, under the Trump administration, coincided with shifts in vaccine policies, including a vote to delay the hepatitis B birth dose for infants.

The judge’s ruling and Malone’s subsequent departure reflect ongoing tensions surrounding vaccine policy and the influence of political appointments on scientific advisory bodies. Malone’s statements regarding the panel’s status have been described as “confusing,” according to STAT’s Chelsea Cirruzzo, highlighting the uncertainty surrounding the ACIP’s future.

CMS Pilot Program for Children with Complex Conditions

On the policy front, the Centers for Medicare and Medicaid Services (CMS) is launching a new $125 million pilot program aimed at improving care coordination for children with complex medical needs. This initiative will fund up to five states to test integrated approaches that connect physical and behavioral healthcare with community support services. The program builds upon previous efforts, including the Integrated Care for Kids Model, which CMS is currently evaluating for potential adjustments.

As STAT’s Mario Aguilar notes, the success of these models often hinges on the details of implementation, including eligibility criteria, payment structures, and evaluation methods. The announcement follows other recent CMS initiatives focused on incentivizing high-quality, cost-effective care through innovative models. More information about the CMS Innovation Center’s model portfolio changes can be found on the CMS website.

Emergency Room Visits and Dementia Care

A First Opinion piece in STAT highlights the challenges faced by patients with dementia and their families when seeking emergency care. Neurologist Gabriela Khazanov shares her personal experience taking her mother, who has dementia, to the ER, and underscores the need for improved protocols to better address the unique needs of this vulnerable population. Studies have shown that dementia patients are at increased risk of adverse outcomes in emergency settings, including prolonged hospital stays and readmissions.

Khazanov and Jay Baruch, who also wrote about similar experiences for STAT, advocate for relatively simple changes to the emergency room environment and procedures that could significantly improve the care provided to patients with dementia. The core message: emergency rooms are often ill-equipped to handle the specific needs of individuals with cognitive impairment.

What to Consider When Using AI for Health Information

The rise of AI chatbots as a source of health information presents both opportunities and risks. While these tools can offer convenient and immediate access to information, it’s crucial to remember their limitations. They are not a substitute for professional medical advice, and relying solely on AI-generated information can lead to misdiagnosis or delayed treatment. The KFF poll data suggests a need for greater public awareness regarding the potential pitfalls of using AI for health-related decisions.

the increasing sophistication of AI-generated medical images, like X-rays, underscores the importance of ongoing vigilance and the development of robust detection methods. Healthcare professionals must be equipped to identify and address potential deepfakes to ensure accurate diagnoses and appropriate patient care.

The evolving landscape of AI in healthcare demands a cautious and informed approach. Individuals should prioritize consultations with qualified healthcare professionals, critically evaluate information obtained from AI sources, and remain aware of the potential risks associated with relying on technology for medical advice. Ongoing research and policy development will be essential to navigate these challenges and harness the benefits of AI while safeguarding patient safety.

What’s the word? Test your knowledge with today’s STAT Mini crossword.